CMTS Motorola BSR PDF

Introduction, Page 0
Implementing the BSR for Basic Data Services
IRI-100
Implementing the BSR for Basic

Data Services
Rev. 1.1 3/02
MOTOROLA and the Stylized M Logo are registered in the US Patent & Trademark Office. All other
product or service names are the property of their respective owners. © Motorola, Inc. 2002.
Page 0
Introduction
Introduction Implementing the BSR for Basic Data Services (IRI-

100) is an instructor-led hands-on course presented
by Motorola Broadband Communications Services to
employees, customers and partners. This five day
course is intended to prepare students to install,
configure and deploy the Motorola Broadband
Services Router 64000 (BSR64000) and BSR1000 in
distributed CATV environments.
Page 1
Course Overview
Course Overview Attendees will learn how to install and configure

BSR1000s to provide basic data services and how to
aggregate this I.P. traffic using BSR64000s.
As this is an introductory course, some time will be
spent on communication theory and the background
technologies that are integral to this implementation.
This course is designed for CATV operator Headend
technicians and network engineers responsible for
implementing high-speed data over cable services.
Page 2
Course Objectives
Course Objectives Upon successful completion of this course, you will

be able to perform the following tasks:
Ô Describe the process involved in moving data
from Customer Premises Equipment to the CMTS.
Ô Explain the transport mechanisms used to deliver
data services in a CATV environment.
¾Differentiate addressing schemes used in a cable
data architecture.
Ô Interconnect BSR64000 and BSR1000s according
to a provided specification.
¾Configure the BSR1000 in a distribution hub to
provide transparent I.P. services from CPE to CMTS.
¾Configure the BSR64000 as an I.P. aggregator for
distribution hubs using dynamic and static routing.
¾Successfully register cable modems with the BSR
providing basic data services.
Ô Troubleshoot, and when possible, resolve basic
implementation-related issues.
Page 3
Prerequisites
Prerequisites To fully benefit from this course, you should already

possess certain prerequisite skills and knowledge.
This includes:
Ô A familiarity with CATV systems and I.P.data
networking
Page 4
Topic Order
Topic Order The Big Picture Module 1

Communications Models Module 2
Media and Encoding Module 3
Installation and Boot-up Module 4
Unique Identification Module 5
Initial Configuration Module 6
CMTS Configuration Module 7
CM/CMTS Communication Module 8
Path Determination Module 9
Dynamic Routing (RIP) Module 10
OSPF Configuration Module 11
Course Labs Appendix A
Course Lab Diagrams Appendix B
Reference Material Appendix C
Page 5
Welcome
High bandwidth plant upgrades, digital video, interactive services, multi-megabit data networking,
Internet access, and telephony.
Today’s broadband landscape is undergoing great change, creating new opportunities. Multiple
providers will vie with each other for customers by offering new revenue-generating applications
and superior network performance. Your broadband network will have to distinguish itself in a
sophisticated environment that supports video, Internet access, information and a host of
interactive services. This is a world where quality and reliability are essential ingredients of the
services you provide.
Motorola’s technical and operational training courses prepare your staff to deliver dependable,
high quality services to your customers. As networks become more complex, your personnel will
need training on cutting-edge technology to be able to support advanced architectures.
Motorola offers many training courses with curricula in broadband network design, operations,
maintenance, testing and product specific technology. The courses are taught by instructors with
a combined experience of over 150 years in the industry.
The instructional design of our courses is focused on a 50/50 mix of lecture and hands-on
laboratory training with Motorola equipment. We emphasize real-world experiences that help
students understand the concepts necessary so they can then demonstrate their skills on actual
products in a supervised laboratory setting.
Please enjoy the class; we certainly enjoy providing it.
Page 6
How We Develop Our Classes
We use The Instructional Systems Design Model, which is a task-oriented model for
systematically developing training. While it can be used in any instructional setting, it is
typically used in business for job performance improvement.
Î
MoP Curriculum
Performance Î Course
Broad area of responsibility or expertise
Ability Î Module
Unit of work or expertise
Component Î Lesson
1st level of breakdown
Step Î Learning Interaction
2nd level of breakdown
Skill/Knowledge Î Teaching Point
What performer must know or do on the job
Page 7
General Information
Ô Please: Ô More:
¾ Turn off cell phones and beepers ¾ Class 9:00 – 5:00
¾ Come back from breaks on time
¾ Ask lots of questions
¾ Share your experiences
¾ Work with your teammates
Page 8
Module 1, Page 0
Module 1
The Big Picture
Rev. 1.1 11/01
Page 0
Module 1, Page 1
Introduction
Introduction Cable systems have developed devices and

architectures geared towards supporting multiple
services, including video and data. We will examine
the devices and architectures currently in use.
Importance It is important to know what components make up the

cable system in order to be able to examine the
hardware and configurations necessary to provide data
services.
Lesson Overview We will look at the big picture and drill down through
the system components and architectures in use to
provide current cable services.
Page 1
Module 1, Page 2
Objectives
Objectives Upon successful completion of this module, you will be

able to perform the following tasks:
ÔDifferentiate signal sources within cable systems
ÔExplain the generic architecture of an HFC system
¾Differentiate between signal sources, conditioning and
distribution components
ÔIdentify components of typical HFC systems and
explain their functions
Page 2
Module 1, Page 3
Topic Order
Topic Order The Big Picture Page 5

Traditional Headend Functions Page 6
System Components Page 7
Headend Output Page 9
Distribution System Access Page 10
Distribution Plant Page 11
Tree and Branch Architecture Page 12
Fiber Backbone Architecture Page 13
Hybrid Fiber Coax Architecture Page 14
Fiber Nodes Page 15
Splitters Page 16
Directional Couplers Page 17
Subscriber Taps Page 18
Page 3
Module 1, Page 4
Topic Order (cont.)
Topic Order Amplifiers Page 19

Forward Signal Path Page 20
Return Signal Path Page 21
Customer Premises Page 22
Customer Drop Page 23
Customer Premises Equipment Page 24
Page 4
Module 1, Page 5
The Big Picture
Our ultimate goal, when looking at the path from signal source(s) to customer
premises, is to pass multiplexed voice, video and data services. This is done in
stages, beginning with signal reception/collection/origination. These signals are
then processed (or conditioned) and multiplexed into a channel plan that may
include voice video and data. It is at this point that they are put on the
distribution plant for transmission to and reception by the customer. We will look
at each of these steps and facilities in detail (and see that they can be further
divided architecturally) as we move forward.
Page 5
Module 1, Page 6
Traditional Headend Functions
Headends are the nerve center of CATV (Community Antenna Television)

systems. Signals are brought into the headend via satellite, microwave, antenna
and digitally through fiber optic systems. Data signals may be originated at
these sites, which are closest to the end user. All of the these signals gathered
at the headend are processed (or conditioned) and combined into the forward
spectrum that is sent out to the subscriber. The headend also receives signals
from the end user data devices on the return system.
The basic forward system components of a headend system include:
•Antennas (satellite, microwave and off-air)
•Video modulators and RF processors
•Data processing equipment
•RF combining
•Fiber Optic Transportation devices
Signal collection in the headend is accomplished with Satellite Dishes (TVRO),
VHF and UHF Antennas, Microwave Antennas, Videotape Players, Telephone
Lines (DS-0), Character Generators, etc.
Signal Processing/Network Control functions use Stereo Encoders, Data
Encoders, Addressable Controllers, Operational Support Systems, Status
Monitoring Systems, etc.
Distribution Plant Access/Multiplexing equipment includes Network Combiners,
Post Amplifiers, Broadband Optical Transmitters, Amplifiers, Optical Splitters,
Return Path Optical Receivers, etc.
Page 6
Module 1, Page 7
System Components
SUBSCRIBER
CONTENT
SOURCES HEADEND SYSTEM SYSTEM
Locally Encoded DHEI IRT 1000 DCT-1000/1200
Analog Signals
QPSK
Digital Satellite 64QAM C6U
Multiplexes
I Mult. IRT 2000
Digital Satellite QPSK DCT-2000
Split Multiplexes C6U
64 QAM
Digital Satellite QPSK Q Mult.
Signals MPS
DS3 64/256QAM C6U DCT-5000+
Digital Transport
Digital Broadcast 8-VSB

Signals
OC-3
Ethernet RPD 1000 DOCSIS
NC 1500
DAC 6000 Cable Modems
The
Internet NC 2000 OM 1000
NETSentryTM
PSTN Digital Domain KLS 1000
Manager
Video Server
Content DANIS/DLS HFC
Interactive
Application HCT 1000
Servers CMTS
Headend system components provide for signal collection, signal

processing/network control and distribution system access/multiplexing of signal
sources. This means the creation of a channel plan (or multiple channel plans)
that stack video, voice and data signals and deliver them to the home. We begin
with multiple signal sources, which we condition and combine in order to place
them on the plant.
Page 7
Module 1, Page 8
System Components (cont.)

Cable Modem Combiner/
Termination System Splitter
TX
Video Fiber
Signals
RX Coax
Trunk
Amp
Cable Tap
Line
Modem Extender
Splitter
Home PC or MAC
Subscriber
Television
The BSR’s transmitter is connected to the combiner and multiplexed with the
video signals. The data channels are combined with the video channels and
other services before hitting the laser.
The levels at the fiber node are critical to the correct operation of the data
services - especially important are the upstream (return) path nodes. Each node
must be characterized to ensure it will meet the requirements for passing data
on the upstream. It must be characterized for multiple upstream data channels.
Page 8
Module 1, Page 9
Headend Output
The individual channels from the RF modulators and heterodyne processors are
combined using the network combiner.
The combined signals are amplified by a low distortion post amplifier to provide
the proper input signal levels to the optical transmitters.
Page 9
Module 1, Page 10
Distribution System Access
The conversion from coaxial transmission to fiber in the cable industry has led
to a decrease in the amount of headends. Headends can feed their signals to
hub sites via fiber in both digital and analog formats. This increased the size of
the area being fed from a single site and at the same time reduced the amount
of processing equipment needed to provide video services. The fiber
infrastructure has also paved the way for new products and services such as
high speed data and digital telephone DOCSIS is a “last mile technology” and
the CMTS is deployed in the hub site. Today’s Headends and hubs are typically
redundant and can service a much wider area with increased reliability.
Signal is applied to the forward laser at the headend. Laser manufacturers
specify the RF input level to the laser. Typically Analog video channels are at a
higher level than digital signals at the input to the laser.
If the RF power level is too high distortions can occur (called clipping).
If the RF level is too low the C/N ratio suffers, potentially leading to degradations
in services.
Always refer to the manufacturers specifications when applying digital signals to
a laser.
Page 10
Module 1, Page 11
Distribution Plant
The distribution plant consists of the cables, strands, actives and passive
devices that comprise the broadband network, delineated from the egress of the
headend to the side of the customer premises.
Page 11
Module 1, Page 12
Tree and Branch Architecture
Originally intended as a means to provide TV signals to remote areas, CATV

systems are now in every major population center and many rural areas as well.
Early systems were point-to-multipoint, feeding signals from a headend site out
to the subscribers. Amplifier cascades could run from 25-45 actives deep on a
trunk run of the Tree and Branch configuration. Attempts at returning signals
from the subscriber to the headend (multipoint-to-point) failed. This was
because of the large amount of amplifiers funneling noise and distortions back to
the headend. Poorly shielded cable used in the house drop system added
interfering signals into the mix.
As HFC plant was built and the number of homes passed per serving area
decreased it was recognized that upstream transmission of signals may be
possible. As early as 1957 experimentation with Pay Per View services on
cable systems took place In Bartlesville, OK.
Today systems use a hybrid of fiber and coax (HFC) with short cascades of 1 to
4 amplifiers and minimal power supplies. This paved the way for offering digital
services that are dependant on “clean plant.”
Page 12
Module 1, Page 13
Fiber Backbone Architecture
Optical fibers transport data between optical node and the CMTS in the cable
headend. Different fiber is used to transport data downstream than to transport
data upstream, hence the use of optical splitters and combiners. At the
headend, the CMTS downstream output is combined with other downstream
cable signals (such as analog and digital video services), while the CMTS
upstream input is picked from the upstream cable spectrum.
Page 13
Module 1, Page 14
Hybrid Fiber Coax Architecture

Early systems were point-to-multipoint, feeding signals from a headend site out
to the subscribers. Amplifier cascades could run from 25-45 actives deep on a
trunk run of the Tree and Branch configuration. Attempts at returning signals
from the subscriber to the headend (multipoint-to-point) failed. This was
because of the large amount of amplifiers funneling noise and distortions back to
the headend. Poorly shielded cable used in the house drop system added
interfering signals into the mix.
possible. As early as 1957 experimentation with Pay Per View services on
cable systems took place In Bartlesville, OK.
Today systems use a hybrid of fiber and coax (HFC) with short cascades of 1 to
4 amplifiers and minimal power supplies. This paved the way for offering digital
Page 14
Module 1, Page 15
Fiber Nodes
DFB (Distributed Feedback) laser with 1310 nm operational wavelength using

single mode fiber with < 0.4 dB loss per kilometer. Included in the system are
Photodetector Receivers, Fabry-Perot Return Lasers and Optical Return Path
Receivers.
Signals are originated at various locations throughout the plant and are
combined together at the node for transportation back to the headend. The
levels from all these locations must be at the same exact amplitude when
presented to the input of the return laser. Unlike forward amplifiers, return
amplifiers have set inputs and variable outputs depending on the losses from the
output to the input of the next active device.
Page 15
Module 1, Page 16
Splitters
Splitters accept signal input in order to divide it into two or more identical
outputs. In the cable world, splitters range in size, dependant on the the size of
the coaxial cable meant to be serviced. Furthermore, splitter design will be
affected by the number of times the signal is meant to be split (you will find 2-
way, 3-way, 4-way splitters, etc.). The more splits, the greater the signal loss on
any given output as compared to the input.
Page 16
Module 1, Page 17
Directional Couplers
Couplers are passive devices used in cable systems to divide and combine
signals. A directional coupler has at least three ports: line-in, line-out, and the
tap. Signal passes between line in and line out ports with loss referred to as the
insertion loss. A smaller portion of the signal power applied to the line-in passes
to the tap port. A signal applied to the tap port is passed to the line in port less
the tap attenuation value (for return signal). The tap signals are isolated from
the line out port to prevent reflections. A signal applied to the line out port
passes to the line in port and is isolated from the tap port. Some devices provide
more than one tap output line (multi-taps).
Page 17
Module 1, Page 18
Subscriber Taps
Typically passive devices used in the “last mile” of the HFC. They are used to
join feeder cable to drop cable, leading to the customer home. Taps come in a
number of different varieties, including 4-port, 8-port, etc. Each port is designed
to feed signal to and from the home.
Page 18
Module 1, Page 19
Amplifiers
Amplifiers are used to maintain unity gain in the distribution system.

Unity gain compensates for transmission losses that occur when signal levels
fall below design standards predetermined to maintain network performance.
Page 19
Module 1, Page 20
Forward Signal Path
The signal flow in a basic amplifier is split by the diplex filter into two paths.
The high RF signals (50 to 750 MHz) are routed from the input (headend side) of
the amplifier, processed, amplified and recombined through a second diplex
filter.
The low RF signals (typically 5 to 40 MHz) flow in the opposite direction
(customer side to the headend).
Page 20
Module 1, Page 21
Return Signal Path
The signal flow in a basic amplifier is split by the diplex filter into two paths.
The high RF signals (50 to 750 MHz) are routed from the input (headend side) of
the amplifier, processed, amplified and recombined through a second diplex
filter.
The low RF signals (typically 5 to 40 MHz) flow in the opposite direction
(customer side to the headend).
Page 21
Module 1, Page 22
Customer Premises
Customer premises begins at the home and not only encompasses all customer
premises equipment that depend on delivered signal, but, all internal distribution
equipment, like cable, splitters, etc.
Page 22
Module 1, Page 23
Customer Drop
House Splitter
Drop cables connect the tap ports to the customer premises. House splitters are
used to provide multiple connections at the customer premises. A bonding
screw/clamp is provided on the splitter for grounding.
Page 23
Module 1, Page 24
Customer Premises Equipment
ÔCommonly Referred to as CPE

¾ Television
¾ Set Top Box
Customer Premises
¾ Cable Modem
» P.C.
Cable Modem
Splitter HFC Plant
Set Top Box

Personal Computer
Television
From a cable data perspective, the CPE - Customer Premises Equipment is

(today) primarily the consumer’s PC. CPE connects to the Cable Modem via an
Ethernet or USB port. The Cable Modem connects to the 2-way cable plant
coax via a standard CATV drop and jack. DOCSIS-enabled set-top boxes are
another example of CMTS-managed customer premises equipment that we are
seeing more of in the field.
Page 24
Module 2, Page 0
Module 2
Communications Models
Rev. 1.0 11/01
Page 0
Module 2, Page 1
Introduction
Introduction High speed cable data communications share

the imprint of two converging technological
traditions – r/f and I.P. engineering. This
convergence allows us the ability to view both
sides within a single, unified framework
Importance Frameworks and models provide a means of

characterizing and understanding technologies.
As such, viewing both the r/f and I.P.
technologies within a single framework will lead
to a greater understanding of both.
Lesson Overview We will look at models as they relate to

communications technologies. Specifically, we
will discuss protocols, specifications,
frameworks and standards within this context.
Page 1
Module 2, Page 2
Objectives
Objectives Upon successful completion of this module, you will

ÔExplain the value of using frameworks in building
specifications for networking technologies
ÔDifferentiate the necessary functions within a
communications system
ÔCorrelate Model layers to Broadband component
functions
ÔIdentify the organizations who are responsible for
maintaining standards
Page 2
Module 2, Page 3
Topic Order
Topic Order Components of a Conversation Page 5

A Layered Approach to Communication Page 6
Who Makes the Rules? Page 7
Models of Interoperability Page 8
What is a Communications Standard? Page 9
Requests for Comments Page 10
What is a Communications Protocol? Page 11
The ISO’s OSI Model Page 12
Remembering the OSI Model Page 13
The Application Layer – Layer 7 Page 14
The Presentation Layer – Layer 6 Page 15
The Session Layer – Layer 5 Page 16
The Transport Layer – Layer 4 Page 17
The Network Layer – Layer 3 Page 18
Page 3
Module 2, Page 4
Topic Order (cont.)
Topic Order The Data Link Layer – Layer 2 Page 19

The Physical Layer – Layer 1 Page 20
Overhead Page 21
Encapsulation Page 22
How the Layers Work Together Page 23
DOD’s TCP/IP Protocol Suite Page 24
The Host-to-Host Model Page 25
Introduction to DOCSIS Page 26
DOCSIS Reference Architecture Page 28
DOCSIS Protocol Model Page 29
DOCSIS Protocol Stack Page 30
Networked Components Page 31
Layers and Devices Page 32
Devices in the HFC Page 33
Page 4
Module 2, Page 5
Components of a Conversation
Ô Unique identification of partner
Sue Ruth ÔLocal (physical)
Hello Ruth. Hello Sue.
ÔRemote (logical)
Sue
Up for Lunch?
Ruth
Sure.
ÔUsing intermediaries
Sue Ruth
Ô Shared language
Later, Ruth. Bye Sue. ÔDialect
Ô Shared media
sound waves
Ô Taking turns
Consider a conversation between two people. There are a number of common

components. As an example, both involved parties will want to uniquely identify
one another to ensure that the correct person is listening and responding. This
becomes more important as more people are within hearing range. The same is
true in the data networking world. In fact, this becomes even more interesting
when communicating parties are not within earshot. In cases like this, these
parties will have to use intermediaries to converse.
Page 5
Module 2, Page 6
A Layered Approach to
Communication
Ô A parallel can be found between
CONCEPT CONCEPT
network and human Communications
Lunch Lunch Ô Human communications may be
divided into three layers
LANGUAGE LANGUAGE
English English ¾ The physical layer
MEDIUM MEDIUM » Sight
Air Air » Sound
¾ The language layer
» English
sound waves » Sign
¾ The concept layer
» Quantum Physics
» Fishing
Two communicating parties must use the same medium, typically sound waves.
A common language that will be understood by both parties must be decided
upon. Each party involved requires some idea of the content and concept of the
conversation. Each communicating party must follow a set of rules governing
each layer. These rules are called protocols.
Page 6
Module 2, Page 7
Who Makes the Rules?
Ô International Telecommunication Union (ITU)

Ô American National Standards Institute (ANSI)
Ô Electronics Industries Association (EIA)
Ô Telecommunications Industries Association (TIA)
Ô Institute of Electrical and Electronics Engineers (IEEE)
Ô International Standards Organization (ISO)
Ô Internet Engineering Task Force (IETF)
Ô Internet Activities Board (IAB)
Ô CableLabs (formerly MCNS)
Who makes the rules? There is no one source for standards. A wide variety of
organizations are involved in the development of internetworking standards.
Below is a list of some of the standards organizations and their web sites:
ITU http://www.itu.int/
ANSI http://web.ansi.org/default_js.htm
EIA http://www.eia.org/
IEEE http://www.ieee.org/
ISO http://www.iso.ch/welcome.html
IETF http://www.ietf.org/
IAB http://www.iab.org/
CableLabs http://www.cablemodem.com
Page 7
Module 2, Page 8
Models of Interoperability
ISO’s DOD’s
OSI Model TCP/IP Suite
Application T F T
D
e T F
N
Presentation
l P T
Application S
n P
e
Session IEEE t Addressing
Transport
and Transport TCP UDP
Port Number
ANSI Protocol Number
Network Standards Internet IGMP IP ICMP IP Address
802.2 SAP/Type Address
Data Link E
t
MAC Address
h
8 8 F
e Network
0 0 D
Physical r Interface
n 2 2 D
e . . I
t 3 5
Networking and internetworking are like playing games. If we all play by the same
rules, the game is fair, we can play, and we understand what is happening.
However, if we don’t play by the same rules, we have chaos and compatibility
issues. Try playing a card game and give everyone different rules and don’t allow
the people to talk. One person thinks high score wins, another is sure it is low
score, another wants to deal three cards, and yet another deals one card face up!
Networks are no different.
Interoperability between vendors is critical and more likely to occur if vendors play
by the same rules. The industry is full of communication models including various
standards and protocol stacks to assist us with standardization. After all, we
wouldn’t be driving automobiles if someone didn’t think about the overall design as
they put the individual pieces together.
Many models have been developed over the years and some are more popular
than others. Models are “guidelines” for communications in the networking world.
When a standard model is followed it becomes possible for two nodes that have
nothing in common other than physical connectivity to share information.
Page 8
Module 2, Page 9
What is a Communications
Standard?
ÔRule, condition, or requirement

ÔDe jure standards
¾ Official; by legislation
¾ Endorsed by standards body
ÔDe facto standards
¾ By fact!
¾ Borrowed
ÔProprietary standards
¾ DEC, NetWare, SNA
ÔNon-proprietary standards
¾ TCP/IP, OSI
Standards are the most important issue in the communications and networking
systems industry. The one constant in this industry is change. The minute a
vendor updates their hardware or software, we need to figure out how we can talk
to it.
De jure standards are endorsed by standards organizations such as the American
National Standards Institute (ANSI) and the Institute of Electrical and Electronics
Engineers (IEEE). ANSI and IEEE have sanctioned many standards that others
have adopted. Standards that others have borrowed are referred to as de facto
standards. Some standards are proprietary in nature and others are open. DEC,
NetWare, and SNA are more proprietary in nature than TCP/IP or OSI which
promote interoperability. Non-proprietary standards are often referred to as open
standards. All of these standards are prevalent today.
Page 9
Module 2, Page 10
Request for Comments
ÔInternet Society (ISOC)

¾ Internet Architecture Board (IAB)
ÔAnyone can propose a new standard
ÔRFCs go through stages
ÔOnce published, RFCs never change
ÔUpdates get a new RFC number
Ôwww.rfc-editor.org
No organization really owns the Internet, but many are responsible for its direction
and operation. The Internet Society (ISOC) and the Internet Architecture Board
(IAB) are responsible for the Internet standards process. ISOC is a global
organization responsible for standards and protocols that allow the Internet to
function. IAB is a technical advisory group responsible for actually setting
standards, publishing RFC’s, and overseeing the standards process. The Internet
Engineering Task Force (IETF), Internet Assigned Numbers Authority (IANA), and
Internet Research Task Force (IRTF) are among the groups under IAB.
Request for Comments (RFCs) are the documents that contain TCP/IP standards
that define protocols. All TCP/IP standards are always published as RFCs. In
many ways this Internet standards process is governed by consensus. If
someone wishes to propose a new standard, they submit an Internet Draft, which
can become an RFC. Documents go through maturity levels and then receive a
classification (Required, Recommended, Elective, Limited Use, or Not
Recommended). If changes are proposed to the standard, a new RFC is created;
an existing RFC is never modified after publication.
RFC’s are available online or for purchase. An excellent site for viewing RFC
information is http://www.rfc-editor.org/.
Page 10
Module 2, Page 11
What is a Communications
Protocol?
A communications protocol is an agreement among parties on a format and

sequence of control messages to be exchanged between the parties. What
language are you speaking? Are there customs you typically follow? Protocols
are nothing more than rules governing transmitting (Tx) and receiving (Rx) data.
A protocol suite or stack is a hierarchical layering of protocols that work together
to carry out communications. Think of it as the world of subcontractors that we
must hire to accomplish an overall project, or departments that carry out their own
duties to make sure company goals are met. Each department, subcontractor, or
“layer” must trust that the other layers will participate and perform properly in
order for things to run smoothly.
Examples of protocol suites include TCP/IP, OSI, DEC, AppleTalk, Novell
NetWare, IBM SNA, etc. Our focus is on the TCP/IP protocol suite, since it is used
for world-wide communications and networking, including the Internet. Let’s
compare the ISO’s OSI model to the widely used TCP/IP protocol suite.
Page 11
Module 2, Page 12
The ISO’s OSI Model
Application 7 7 Application
Presentation 6 6 Presentation
Session 5 5 Session
segments segments
Transport 4 4 Transport
3 packets packets 3
Network Network
frames frames
Data Link 2 2 Data Link
bits bits
Physical 1 1 Physical
Host “A” Host “B”
The International Organization for Standardization (ISO) developed the Open

Systems Interconnection (OSI) model during the 1970's to facilitate interoperability
between vendors. Understanding the purpose and structure of the OSI model is
key to understanding networking protocols, utilities, and how they work together to
carry out communications.
The sending computer starts at the Application layer and works its way down to
the Physical layer with each layer adding instructions (a header) for what it’s peer
layer on the receiving computer is to do with the data. The Data Link layer adds a
trailer for error checking and sequencing, and the data is converted to 0’s and 1’s
to go across the wire. Starting at the Physical layer, the receiving computer’s NIC
removes the bits from the wire, transforms them into parallel data for the computer
(Data Link layer frames) and continues up the stack until the packets get to the
Application layer.
Data that is too large for transmission across the media is broken into more
manageable chunks to be processed and sent to its destination. This means that
all packets don’t necessarily take the same path. However the packets are put
back together in the correct order at their final destination.
Page 12
Module 2, Page 13
Remembering the OSI Model
Again Again All All

People Preston People People
Support See Seem Seem
Threaten To To To
Not Ned Need Need
Do Dumped Dr Data
Please Phyllis Pepper Processing
How will you ever remember the layers? Here are a few mnemonics to assist.
Keep in mind that the layers are arranged in a vertical hierarchical manner
because tasks in the upper layers are dependent upon those performed in the
lower layers. Use the OSI model as a reference point to all networking topics. It is
a reference model or conceptual blueprint of how communications should occur.
The industry uses a layered model because it eases troubleshooting. Networking
complexity is broken into more manageable layers so developers can focus on
one layer without changing all the code. Standards do promote interoperability.
Page 13
Module 2, Page 14
The Application Layer — Layer 7
Ô Service use and advertisement

¾ File
¾ Print
¾ Message
¾ Application
¾ Database
Ô FTP
Ô HTTP
Ô Telnet
Ô SMTP
Ô SNMP
Ô Gateways (all layers)
Layer 7, the top layer of the OSI model, provides shared services. Servers
advertise file, print, message, application, and database services. Common
Application layer services include X.400 mail services, Simple Mail Transfer
Protocol (SMTP), Post Office Protocol (POP3), Internet Messaging Access
Protocol (IMAP4), X.500 directory services, Telnet, File Transfer Protocol (FTP),
HyperText Transfer Protocol (HTTP), and Simple Network Management Protocol
(SNMP).
The Application layer is the layer that the user works with directly. It is the start of
communications. Application gateways are required if the applications are
different. For example, if my company uses Outlook and yours uses Notes, email
gateways assist in the delivery of the incompatible systems. Perhaps you have an
IBM mainframe connected on your LAN. An SNA gateway is specific to this
application. Gateways work at the upper layers of the OSI model. A gateway is a
protocol translator.
Page 14
Module 2, Page 15
The Presentation Layer — Layer 6
Ô “The Translator”
Ô Encryption
¾ Private key
¾ Public key
Ô Compression
¾ MP3
¾ MPEG, JPEG
Ô Character codes and file syntax
¾ ASCII
¾ EBCDIC
Layer 6, the Presentation layer of the OSI model, is the translator. It presents data
to the Application layer on the receiving end. The Presentation layer coverts
formats such as American Standard Code for Information Interchange (ASCII) to
Extended Binary Coded Decimal Interchange Code (EBCDIC).
Translation, encryption/decryption, and compression/decompression features are
associated with this layer. TIFF, JPEG, MIDI, and MPEG are some examples of
graphic and multimedia formats used at this layer. Public and private key
encryption schemes operate here. Secret or private key technologies use the
sender’s private key to encrypt/decrypt. Rivest-Shamir-Adleman (RSA) is an
example of this same key method where the private key is kept by the owner and
the public key is published. Public key technologies use the recipient’s public key
to encrypt and the recipient’s private key to decrypt. Data Encryption Standard
(DES) is an example of where the private key is randomly generated for each use.
Secure Sockets Layer (SSL) and Transport Layer Security (TLS) are the most
common security protocols on the Internet.
Page 15
Module 2, Page 16
The Session Layer — Layer 5
Ô “The Operator”
¾ Dialog
¾ Establish session
¾ Administer session
¾ Tear down session
Ô Communications
¾ Simplex
¾ Half-duplex
¾ Full-duplex
Ô NetBIOS/sockets communications
Ô Network login/drive mappings
Layer 5, the Session layer of the OSI model, is the the “dialog” layer. Think of it
as the “operator” or switchboard person who initiates, manages, and tears down
the logical communication session.
Session layer communications are concerned with dialog between network nodes
or devices. Three phases of communications include session establishment, data
transfer, and session release. Three modes of operation include simplex, half-
duplex, and full-duplex communications. Simplex is like a radio station broadcast
where one device is sending and others are receiving; simplex is a one-way
transmission. Half-duplex is like walkie-talkies. It supports the transmission of
data in both directions, but only one direction at a time. Full-duplex is two-way
communications, transmitting and receiving at the same time. Typically, one pair
of wires or frequency is used to transmit and another is used to receive with full-
duplex. A telephone conversation uses full-duplex communications.
Some examples of session layer protocols and interfaces include Network File
System (NFS), Structured Query Language (SQL), Remote Procedure Calls
(RPC), NetBIOS communications, Sockets, network login, and drive mappings.
Page 16
Module 2, Page 17
The Transport Layer — Layer 4
Ô Delivery
¾ Connection-oriented/connectionless
¾ Reliable/unreliable
Ô End-to-end communications
¾ Error control and flow control
¾ ACKs and NAKs
¾ Sequencing
¾ Hides the lower layer intricacies
from the upper layers
Ô Application multiplexing
Ô TCP/ UDP
Layer 4, the Transport layer of the OSI model, is responsible for end-to-end
delivery and control, or accuracy. It hides lower-layer intricacies from the upper
layers by deciphering the packets to see if the data inside the packets is in the
correct format. Layer 4 assists with error detection and correction; whereas layer
2 is typically concerned with only error detection.
The Transport layer is responsible for application multiplexing. The services of
layer 4 both segment and reassemble data from upper layer applications and
unite it on to the same data stream. Client hosts can access the same applications
on a remote host through various port numbers.
Delivery can be connection-oriented or connectionless. It can also be reliable or
unreliable. Connection-oriented means establishing a connection before
communications can occur. Connectionless does not require an established
connection before communications can occur. Transmission Control Protocol
(TCP) is an example of a connection-oriented reliable protocol, and User
Datagram Protocol (UDP) is an example of a connectionless unreliable protocol.
Reliable protocols provide delivery through acknowledgments and flow control
(certified mail); whereas unreliable protocols (regular mail) rely on the upper
layers for reliability.
Page 17
Module 2, Page 18
The Network Layer — Layer 3
Ô Path Determination
Ô Routing
Ô Logical Addressing 192.168.0.0
Ô Routers exchange routing information
Ô Last-hop router determines the node
Ô IP
Ô ICMP
Ô ARP
Layer 3, the Network layer of the OSI model, is responsible for logical addressing
and path determination for internetworking networks. Routers examine the packet
or datagram to see where the packet is coming from and where it is going.
Packets can take different routes to reach their destination network. Routers look
at the network portion of the logical address and talk to other routers to determine
what to do with the packet. The destination router handles the local delivery to the
appropriate node.
The Network layer is like driving home. You listen to the radio, and if the route you
normally take is under construction or there is an accident, you take an alternate
route. Routers check their routing tables to find an alternate route just as humans
check maps or stop at gas stations for directions.
Internet Protocol (IP) is an example of a Network layer routed protocol. Internet
Control Message Protocol (ICMP) and Address Resolution Protocol (ARP) are
helper protocols. ICMP is for status and error reporting, and ARP is for resolving
IP addresses to MAC addresses.
Page 18
Module 2, Page 19
The Data Link Layer — Layer 2
Ô How does the network work?

Ô Network Interface Card
¾ Packages raw bits into frames
¾ Drivers
¾ The MAC address 00-00-0c-12-34-56
Ô Bridges and switches
Ô LAN topologies such as Ethernet
Ô WAN protocols such as Frame Relay
Ô Logical topologies and access methods
Layer 2, the Data Link layer of the OSI model, transmits frames from node to node
depending on the hardware address. Layer 2 converts frames to bits to go out on
the wire on the sending side and packages the raw data bits into frames on the
receiving end. Frames are Protocol Data Units (PDUs) that contain control bits for
the beginning and end of the frame and a trailer for error checking. Think of
frames as picture frames and how the frame marks the edges of a picture. Bits
are 1's and 0's which ultimately are the only protocol data units that computers
understand.
The Data Link layer encompasses various LAN and WAN protocols such as
Ethernet, Token Ring, Fiber Distributed Data Interface (FDDI), Copper Distributed
Data Interface (CDDI), High Level Data Link Control (HDLC), and Frame Relay.
Bridges and switches operate here.
The IEEE 802 Project, which was named for the year and month it began, divides
the Data Link layer into two sublayers, the Logical Link Control (LLC) and Media
Access Control (MAC). LLC provides software controls to manage multiple
protocols, and the MAC layer defines how devices access the wire. An example
of an access method is Carrier Sense Multiple Access with Collision Detection
(CSMA/CD) for Ethernet.
Page 19
Module 2, Page 20
The Physical Layer — Layer 1
ÔWhat does the network look like?

ÔTransmit bits 00001111
Ô Mechanical specifications
¾ Connectors
¾ Cabling
¾ Pin-outs
Ô Category 5 cable
Ô EIA/TIA-232
Ô Physical topologies and signaling
Layer 1, the Physical layer of the OSI model, sends and receives bits. Bits are 1's
and 0's. Starting at the Physical Layer, the receiving computer’s NIC removes the
bits from the wire, transforms them into parallel data for the computer (Data Link
layer frames) and continues up the stack until the packets get to the Application
layer.
The Physical layer coordinates the rules for bit transmission. Cables, connectors,
hubs, and modems operate here. All devices on a network are connected through
the Physical layer. Physical topologies such as bus, ring, and star and other
mechanical and electrical specifications are defined at layer 1. Examples include:
EIA/TIA-232, V.35, Category 5 cable, and High Speed Serial Interface (HSSI).
Page 20
Module 2, Page 21
Overhead
Ô Additional information
Sue Ruth
Ô In support of data transfer
Hello Ruth. Hello Sue. Ô Conversation set-up
Ô Acknowledgement
Sue Ruth
Up for Lunch? Sure.
Ô Unique identification
ÔBi-directionally
Sue Ruth
Later, Ruth. Bye Sue.
Ô Transfer of information
Ô Conversation end
Ô Some not as obvious
sound waves
Ô Indication of language
Ô Length of message
There is some amount of overhead involved in passing data, not just between
devices, but between layers of the network stack. This includes addition of
sending and receiving address to the message in the form of a header. The
inclusion and use of error checking and correction mechanisms is another
example. These additional components are required for data passing, both locally
and remotely, but, take up precious processing and bandwidth resources.
Page 21
Module 2, Page 22
Encapsulation
ENCAPSULATION User information Data
DE-ENCAPSULATION
Data Segments
Segments Packets
Packets Frames
Frames Bits
To review the OSI model communications process, the sending computer starts at
the Application layer and works its way down to the Physical layer with each layer
adding instructions (a header) for what the peer layer is to do with the data. The
Data Link layer adds a trailer for error checking, and the data is converted to 0's
and 1's to go across the wire. Starting at the Physical layer, the receiving
computer’s NIC removes the bits from the wire, transforms them into frames for
the computer (Data Link layer) and continues up the stack until the packets get to
the Application layer. Larger chunks of data are broken into more manageable
chunks to be processed and sent to their destination.
This entire process is referred to as encapsulation. The sending computer starts
at the upper layers (Application, Presentation, Session) with user information that
is referred to as messages or data. The Transport layer encapsulates the data
into segments. The Network layer encapsulates the segments into packets or
datagrams. The Data Link layer encapsulates the datagrams into frames, and the
Physical layer requires bits to traverse the wire. After the bits get to their
destination the de-encapsulation process begins. Each layer reads the header
instructions from its peer, discards the header, and sends the protocol data units
up the stack to the destination computer.
Page 22
Module 2, Page 23
How the Layers Work Together
Sending Receiving
DATA
App App
Application Application Application

AH DATA
Layer Protocol Layer
Presentation Presentation Presentation

PH DATA
Session Session Session

SH DATA
Transport Transport Transport

TH DATA
Network Network Network

NH DATA
Data Link Data Link Data Link

DH DATA DT
Physical Physical
BITS
Layer Layer
Here is another representation of the OSI model including the encapsulation/de-

encapsulation process. Notice how each layer adds a header so the peer layer
has the instructions to no what to do with the data.
The ISO’s OSI model helps us determine which components can exist on a
network and how they operate. This includes hardware, software, standards,
protocols, and topologies. This is a modular approach to understanding network
communications. We will discuss the layers in more detail and compare them to
the widely used TCP/IP suite of protocols.
Page 23
Module 2, Page 24
DOD’s TCP/IP Protocol Suite
OSI TCP/IP
Application 7
Presentation 6 Application
Session 5
Transport
Transport 4
(host-to-host)
Network 3 Internet
Data Link 2
Network Interface
Physical 1
Another popular model is the Department of Defense (DOD) TCP/IP protocol

suite, also referred to as the Host-to-Host model. Instead of the 7-layer OSI
model approach, the Host-to-Host model uses 4 layers. Some vendors refer to
the lower two layers as the Link layer and the next layer as the Network layer.
The Application layer is equivalent to the OSI Application, Presentation, and
Session Layers. The Transport layer is equivalent to the OSI Transport layer.
The Internet layer is equivalent to the OSI model Network layer, and the Network
Interface layer encompasses the OSI Data Link and Physical layers.
TCP/IP is more than the two protocols in its name. It is a robust set of enterprise
networking protocols that work together to connect dissimilar systems and a
method of gaining access to the Internet.
Page 24
Module 2, Page 25
The Host-to-Host Model

File Data
FTP FTP
Client Data Server
Application Application
FTP FTP Data FTP
TCP TCP Data TCP
IP IP Data IP
Ethernet EH Data ET Ethernet
10BaseT BITS 10BaseT
The Host-to-Host’s TCP/IP model protocols work together to carry data to it’s
destination. Let’s look at a practical example using an FTP application.
A client wants to transfer a file using File Transfer Protocol (FTP). FTP is the
Application layer protocol. Notice how FTP carries the data to the Transport layer
by way of Transmission Control Protocol (TCP). TCP encapsulates FTP, adds its
own header, and sends it on to the Internet layer. At the Internet layer, the
Internet Protocol (IP) encapsulates the data, adds an IP header, and sends it to
the Network Interface layer, where the Ethernet header is added. The Ethernet
frame gets translated to 1's and 0's to go across the 10BaseT wire.
The receiving computer, an FTP server in this example, pulls the bits off the wire,
strips the peer headers, and moves the FTP request up the protocol stack.
The DOD model and its corresponding protocols will be covered in more detail
throughout the course. We have defined networking; discussed types,
components, and requirements to help us make decisions; explored standards
and protocols through industry models; and introduced TCP/IP.
Page 25
Module 2, Page 26
Introduction to DOCSIS
Ô1992 -> 1996 proprietary systems

ÔStandards effort in IEEE 802.14 stalled
ÔMajor MSO’s wanted
¾ Lower costs through retail cable modems, multiple vendors
ÔCM to CMTS interface & system operation
ÔIndustry standard
¾ Specs & certification
¾ Started as RFI, grew into family of specs
ÔSystem level set of specifications against a reference architecture
» Enabling IP (transparently to the customer) over cable
Developed by CableLabs and approved by the ITU in March 1998, Data Over
Cable Service Interface Specification defines interface standards for cable
modems and supporting equipment.
With certification from CableLabs, manufacturers will be able to produce cable
modems for retail, so consumers no longer have to depend on leased cable
modems from their cable providers.
Other devices that recognize and support the DOCSIS standard include HDTVs
and Web enabled set-top boxes for regular televisions.
DOCSIS specifies downstream traffic transfer rates between 27 and 36 Mbps over
a radio frequency (RF) path in the 50 MHz to 750+ MHz range, and upstream
traffic transfer rates between 320 Kbps and 10 Mbps over a RF path between 5
and 42 MHz. But, because data over cable travels on a shared loop, individuals
will see transfer rates drop as more users gain access.
In 1998, there were 1.2 million cable modems installed across the United States
with an average price of $245 per unit, and by 2004, research reports predict
there will be 24.3 million units installed across the US with an average price of $50
per unit.
Page 26
Module 2, Page 27
Introduction to DOCSIS (cont.)
ÔData Over Cable Service Interface Specifications

¾ System-level specifications against reference architecture
» DOCSIS 1.0
• Radio Frequency (RF) Interface Specification 1.0
• Operations Support System Interface (OSSI) 1.0
• Baseline Privacy Interface Specification (BPI)
• Cable Modem Termination System - Network Side Interface
• Cable Modem to Customer Premise Equip. Interface (CMCI)
• Cable Modem Telephony Return Interface Specification
• Acceptance Test Plan 1.0
» DOCSIS 1.1
• Baseline Privacy Plus Interface Specification (BPI+)
• DOCSIS 1.1 CMCI Acceptance Test Plan
• DOCSIS 1.1 RFI Acceptance Test Plan
» DOCSIS 2.0
• Currently Interim Standards
Standards effort in IEEE 802.14 were stalled and Major MSOs wanted to lower
their costs through retail cable modems and interoperability of multiple vendors’
equipment. DOCSIS 1.0 is effectively complete. Product is available and being
installed in hundreds of CATV systems. DOCSIS 1.1 cable modems are available
and CMTS are currently being tested. This will add technology that makes
possible Quality of Service (QoS) guarantees needed for newer services like
DOCSIS-based telephony over cable. Another CableLabs effort -- PacketCable --
is merging its efforts with DOCSIS to realize this.
DOCSIS is really a family of documents. These are available from the public area
of the CableLabs cable modem website at
http://www.cablemodem.com/specifications.html. DOCSIS defines a
Reference Architecture for an overall data communication system optimized to
connect consumer PCs to the internet using the DOCSIS-compliant Cable
Modems and DOCSIS-compliant Cable Modem Termination Systems. These
specifications really started as just RFI and grew into family of specifications that
have been accepted by the MCNS and IEEE.
DOCSIS 2.0, the newest standard will provide 30 megabits of upstream capacity
for each 6MHz cable TV channel over cable hybrid fiber/coax networks. Current
DOCSIS 1.0 modems provide a maximum of 4 megabits. DOCSIS 1.1 provides
10 megabits.
DOCSIS 2.0 will include two modulation techniques -- synchronous code-division
multiple access (S-CDMA) and advanced frequency agile time division multiple
access (A-TDMA). Equipment will be backward-compatible with versions 1.0 and
1.1.
Since the major changes in DOCSIS 2.0 are at the physical layer (hence the
name Advanced PHY), the impact on existing CM/CMTS implementations should
be minimal.
Page 27
Module 2, Page 28
DOCSIS Reference Architecture
DISTRIBUTION HUB OR HEADEND

DS TV signals
LOCAL
SERVER
Cable
FACILITY OTx
COMBINER
Modem
Termination CMTS Cable Modem
System Termination FIBER
Network Side CABLE DS = >54 MHz
MOD System
Interface TERMINATION PLANT
NETWORK
Downstream
(CMTS-NSI) F.O.
LAN
RF Interface
NODE
DEMOD Cable Modem
SPLITTER
Internet Backbone
to RF Interface CM CPE
ORx
OSS COAX
Interface Baseline Privacy CABLE
Network
Cable
(OSSI) Interface (BPI) Modem PLANT
OPERATIONS SECURITY & Termination
SUPPORT ACCESS US signals System
US = <42 MHz
SERVER CONTROLLER Upstream
RF Interface
Cable Modem to
REMOTE SERVER FACILITY CPE Interface
(CMCI)
The CPE - Customer Premise Equipment is (today) primarily the consumer’s PC.
CPE connects to the Cable Modem via an Ethernet or USB port. The Cable
Modem connects to the 2-way cable plant coax via a standard CATV drop and
jack.
A two-way coax cable transports data to/from the HFC optical node that feeds the
home, using traditional 2-way coax cable technology. This technology transports
signals higher than 54 MHz in the downstream direction, and signals lower than
42 MHz in the upstream direction, by RFI specification.
data upstream, hence the use of optical splitters and combiners. At the headend,
the CMTS downstream output is combined with other downstream cable signals
(such as analog and digital video services), while the CMTS upstream input is
picked from the upstream cable spectrum.
On the CMTS network side interface, a LAN in the headend connects the CMTS
to the Internet and a number of support servers required by DOCSIS specification
(i.e., DHCP,ToD and TFPT), plus (typically) data caching servers and provisioning
systems. The support servers are often physically located at another Regional
Data Center, that may be tens or hundreds of miles from the headend. Caching
servers are co-located with the CMTS at the cable headend.
Page 28
Module 2, Page 29
DOCSIS Protocol Model
DHCP TFTP SNMP SMTP Ô Well known application

TOD HTTP FTP Telnet protocols
UDP TCP Ô Well known transport protocols
Ô Well known network protocol
IP Protocol
Data Link Encryption
Ô DOCSIS Specific Protocols

Media Access Control
Downstream Transmission
Convergence Sub-layer (MPEG- 2)
Physical Media Dependant Layer
The CM and CMTS operate as forwarding agents and also as end-systems

(hosts). The protocol stacks used in these modes will differ. The principal function
of the cable modem system is to transmit Internet Protocol (IP) packets
transparently between the headend and the subscriber location. Certain
management functions also ride on IP, so that the protocol stack on the cable
network is as shown above (this does not restrict the generality of IP transparency
between the headend and the customer). These management functions include,
for example, supporting spectrum management functions and the downloading of
software.
CM and CMTS as Hosts
CMs and CMTSs will operate as IP and LLC hosts in terms of IEEE Standard 802
[IEEE802] for communication over the cable network. The protocol stack at the
CM and CMTS RF interfaces is shown above.
The CM and CMTS MUST function as IP hosts. As such, the CM and CMTS
MUST support IP and ARP over DIX link-layer framing (see [DIX]). [Note: DIX link-
layer framing refers to the frame format only; the CM and CMTS are not subject to
the DIX 64 byte limit on minimum frame size.] The CM and CMTS MAY also
support IP and ARP over SNAP framing [RFC-1042]. The CM and CMTS also
MUST function as LLC hosts. As such, the CM and CMTS MUST respond
appropriately to TEST and XID requests per [ISO8802-2].
Page 29
Module 2, Page 30
DOCSIS Protocol Stack

OSI Stack Data-Over-Cable Interface Specifications
Cable Modem Termination System Cable Modem (Ethernet CMCI)

Security Security
DHCP TFTP SNMP Mgmt DHCP TFTP SNMP Mgmt
Higher Layers Applications DOCSIS Applications Applications
Control
Transport TCP/UDP Messages TCP/UDP TCP/UDP
IP
Network IP IP
Forwarding 802.2/DIX LLC 802.2/DIX LLC 802.1d

Bridging 802.2/DIX
Data LLC
Data Link Link Security Link Security
Link Layer
802.3/DIX
DOCSIS MAC DOCSIS MAC MAC
U/S TDMA (minislots) U/S TDMA (minislots)

Physical PHY Layer D/S TDM (MPEG) D/S TDM (MPEG) 802.3
10 Base-T
Cable PMD Cable PMD
CMTS-NSI Cable Transmission Network CMCI

Interface to/from WAN Interface
Management only to/from Customer
Data Premises Equipment
Data forwarding through the CMTS may be transparent bridging or may use
network layer forwarding (routing, IP switching). Data forwarding through the
cable modem is link layer transparent bridging. Forwarding of IP traffic must be
supported. DOCSIS was designed against as a CSMA shared medium with the
RFI defining layers one and two only. Layer three is handled by standard IP
technologies by definition.
Conceptually, the CMTS forwards data packets at two abstract interfaces:
between the CMTS-RFI and the CMTS-NSI, and between the upstream and
downstream channels. The CMTS MAY use any combination of link-layer
(bridging) and network-layer (routing) semantics at each of these interfaces. The
methods used at the two interfaces need not be the same.
Forwarding between the upstream and downstream channels within a MAC layer
differs from traditional LAN forwarding in that: A single channel is simplex, and
cannot be considered a complete interface for most protocol (e.g., 802.1d
spanning tree, Routing Information Protocol per RFC-1058, etc.) purposes.
Upstream channels are essentially point-to-point, whereas downstream channels
are shared-media.
For these reasons, an abstract entity called the MAC Forwarder exists within the
CMTS to provide connectivity between stations within a MAC domain.
Page 30
Module 2, Page 31
Networked Components
Video on OC-3c ATM 10/100 Ethernet
Demand
QPSK Modulators
QPSK
FIBER TRANSPORT
DOWNSTREAM
ANALOG
BROADCAST FIBER TRANSPORT

VIDEO
HFC PLANT
QPSK
UPSTRAEM
FIBER TRANSPORT QAM64/256
Service QPSK/QAM16
Provider
Router/CMTS
ATM Switch ATM to Ethernet

Uplink
Giga-Bit Switch Router

NETWORK TRANSPORT SYSTEM
Router (GSR)
Downstream RF
Optical Transmitter /
Receiver Upstream RF
Baseband (OC-3, etc.)

Router/CMTS
REB 15-FEB-00
We can look at the components involved in providing cable data services as being
either on the HFC, in the network transport system or as parts of the WAN. Each
of these pieces act, when dealing with data transport (unique identification of
where to send data), at specific layers, as we will see in the next couple of slides.
Page 31
Module 2, Page 32
Layers and Devices
Physical (L1) Hub
Datalink (L2) Switch/Bridge
Network (L3) Router
Devices are attributed to specific layers in much the same way that protocol
function is. The rule of thumb for determining at what layer a network device
operates is to look at which layer it makes it’s forwarding decisions. From this
perspective, something like a hub is a Layer 1 device, since it makes no
forwarding decisions – it serves as a conduit for the data, enabling it to go from
one device (or segment) to another.
Switches and bridges use Layer 2 addressing to make their forwarding decisions,
while routers go to Layer 3.
Intelligent hubs and switches include additional features that enables an
administrator to monitor the traffic passing through the device and to configure
ports and features, typically using the Simple Network Management Protocol
(SNMP). Although assigned unique identifiers at Layer 3 (specifically I.P.
addresses), they remain Layer 1 and 2 devices respectively, due to how they still
make their forwarding decisions. Many times you will hear these devices referred
to as either Managed or Manageable devices (e.g., managed hub, manageable
switch, etc.).
Page 32
Module 2, Page 33
Devices in the HFC
Physical (L1) Fiber, Coax,

Nodes, Amps
Datalink (L2) CM/CMTS
Network (L3) CMTS/Router
Cable data devices are nothing more than specialized data networking devices.
As such, they too can be categorized as Layer 1, 2 or 3 devices.
What is traditionally referred to as the HFC Plant consists of Layer 1 devices,
specifically, fiber, coax, forward and reverse nodes, splitters, taps, etc.
Cable modems make their forwarding decisions at Layer 2, making them Layer 2
devices. As with switches, cable modems are manageable (by DOCSIS OSSI
specification), making them managed Layer 2 devices. Some vendors have
recently introduced home cable devices with Layer 3 path determination
functionality – which, by definition, makes them no longer cable modems.
Devices like this are called cable modem routers (or some variant), because of
this.
The DOCSIS Specification allows for cable modem termination systems (CMTS)
to act at either layer two or layer three.
Page 33
Module 3, Page 0
Module 3
Media and Encoding
Rev. 1.0 11/01
Page 0
Module 3, Page 1
Introduction
Introduction Cable data communications have the imprint of

two converging technological traditions – r/f
and I.P. engineering. Much of what we see in
terms of physical media and encoding is
shared across both sides.
Importance An understanding of the media and encoding

methods within the cable data system is critical
in order to be able to properly plan and
implement support according to their needs.
Lesson Overview We will look at cable data networking

topologies, media choices and encoding
methods.
Page 1
Module 3, Page 2
Objectives

ÔDescribe physical topologies
ÔExplain the difference between baseband and
broadband signaling
ÔIdentify cabling types and connectors
ÔExplain the process of modulating a digital signal
onto an analog carrier
Page 2
Module 3, Page 3
Topic Order
Topic Order The HFC Network Page 6

Media for Cable Data Networking Page 7
Data Network Topologies Page 8
Physical Bus Topology Page 9
Physical Star Topology Page 10
Physical Star Logical Ring Page 11
Mesh Topology Page 12
WAN/Network Transport Interfaces Page 13
Ethernet Page 14
5 – 4 – 3 Rule Page 15
Fast Ethernet Page 16
Gigabit Ethernet Page 20
Gigabit Ethernet Protocol Architecture Page 21
Twisted Pair Cable Page 22
Unshielded Twisted Pair (UTP) Page 23
Page 3
Module 3, Page 4
Topic Order (cont.)
Topic Order Unshielded Twisted Pair Connectors Page 24

UTP Straight-Through Page 25
UTP Crossover Page 26
Fiber Optic Cable Page 27
Fiber Optic Cable Composition Page 28
Coaxial Cable Composition Page 29
The RF Spectrum Page 30
Off-Air vs. Cable Page 33
Amplitude and Wavelength Page 34
Frequency and Wavelength Page 35
Frequency, Amplitude and Wavelength Page 36
Waves Page 37
Attenuation Page 38
Modulation Page 40
Analog Signal Page 41
Page 4
Module 3, Page 5
Topic Order (cont.)

Topic Order Amplitude Modulation Page 42
Baseband Signaling Page 43
Broadband Signaling Page 44
Frequency and Phase Modulation Page 45
Digital Signal Page 46
Analog to Digital Conversion Page 47
Bi-Phase Shift Keying (BPSK) Page 48
Bits and Symbols Page 49
Quadrature Modulation Page 51
Constellation Page 52
Quadrature Modulation
Constellations Page 55
64 and 256 QAM Constellations Page 57
Decision Boundaries Page 59
Constellation Display Page 60
Page 5
Module 3, Page 6
The HFC Network

Cable Modem Combiner/
Termination System Splitter
TX
TV Fiber
Channels
RX Coax
Trunk
Amp
Cable Modem Line Tap

Extender
Splitter
Set Top Box
Personal
Home
Computer
Subscriber
Television
•BSR transmit is connected to the combiner with the video signals.

• The data channels are combined with the video channels and other services
before hitting the laser.
• The levels at the fiber node are critical to the correct operation of the data
services - especially important are the upstream (return) path nodes. Each node
must be characterized to ensure it will meet the requirements for passing data on
the upstream. It must be characterized for multiple upstream data channels, i.e.,
10.
• Distribution: downstream is not usually a problem, upstream may have to be
tweaked.
• Home: only one splitter or DC coupler between the cable and the modem.
Page 6
Module 3, Page 7
Media for Cable Data Networking
DOWNSTREAM
Operational
HFC PLANT
Support
Systems
UPSTRAEM
FIBER TRANSPORT QAM64/256
Service Provider QPSK/QAM16
Router/CMTS
Wide Area Network (WAN) NETWORK TRANSPORT SYSTEM HFC
Border Router Switch
Downstream RF
Optical Transmitter /
Receiver Upstream RF
(OC-3,
Baseband Ethernet,
Router/CMTS etc.)
Page 7
Module 3, Page 8
Data Network Topologies
ÔPhysical topologies (layer 1)

¾ Bus
¾ Star
¾ Ring
¾ Mesh
ÔLogical topologies (layer 2)
¾ How does the network work?
Physical layer networking is layer 1 of the OSI model. It sends and receives bits
and coordinates the rules for transmission. Topologies, signaling types, cables,
connectors, hubs, and modems operate here.
What the network looks like is referred to as physical topology. How it works is
called logical topology. Common physical topology types include bus, star, ring,
and mesh with star being the most popular.
Page 8
Module 3, Page 9
Physical Bus Topology
Ô Ethernet
¾ Backbone or trunk segment
¾ Small workgroups
Ô Installation
Ô Terminators
Ô Cost
Ô Troubleshooting
Ô Data transmission
Ô Hubs
A physical bus topology consists of a linear transmission medium that is

terminated at both ends. It looks like a clothesline where clothes (nodes) attach
directly, making it difficult to troubleshoot. Any break in the bus or lack of
termination causes the entire network to become inoperable. However, the failure
of one workstation will not cripple the network.
The bus is typically represented as a straight line, although it usually is a series of
cables connected together like Christmas tree lights. The “clothesline” is referred
to as a trunk or backbone segment, and each drop can be a client, server, printer,
etc. All devices on the bus hear everything. Data is transmitted to all nodes, but
only the destination computer accepts it. Hubs regenerate signals; however, there
is no signal regeneration from one computer to another.
This is an economical use of cable and an excellent choice for small networks. It
is easy to add workstations, and repeaters can be used to boost or regenerate the
signal. The disadvantages include troubleshooting and heavy traffic situations.
Page 9
Module 3, Page 10
Physical Star Topology
ÔEthernet 10BaseT
To another Hub
ÔInstallation and cost
Hub (optional) ÔTroubleshooting
ÔWorks like a bus
topology
Workstation
Segment
A star or hub-and-spoke physical topology connects nodes via point-to-point links

to a central hub or concentrator. The central hub works like the bus physical
topology in that all devices on the hub hear everything. Signals are transmitted to
all nodes connected to the hub, but as in the bus topology, only the destination
computer accepts it. There is no signal regeneration from one computer to
another, although the hub does regenerate the signal for transmission.
Troubleshooting is much easier with a star topology. The failure of one device
does not usually affect the rest of the devices unless the main server or
centralized hub is the device that fails. The possible area of failure for a given
node consists of that workstation, the cable connecting it to the hub, and the port
on the hub itself. Centralized diagnostics and management features are built into
many hubs.
Although easy to add and troubleshoot, a star topology can require more cable
than bus or ring topologies. A good example of a physical star would be 10BaseT
Ethernet. Note however, that it is still a logical bus. We call this a “physical star,
logical bus” network. A large majority of modern networks are configured in this
way.
Page 10
Module 3, Page 11
Physical Star Logical Ring
ÔToken Ring and FDDI

¾ Physical star
¾ Logical ring
ÔInstallation and cost
ÔTroubleshooting
ÔToken passing
A ring topology joins all the nodes together to form a closed loop of point-to-point
links between stations. This is an active topology: each workstation acts as a
repeater to minimize signal degradation. Failure of any one device in a true ring
can cripple the network, and thus is not used often today.
A token is passed around the ring in one direction. If a station has the token it can
talk. The ring topology provides equal access for all computers in this respect.
Message verification is performed by each and every computer. It is a “hey, is this
for me?” approach with two outcomes. One: “if the data is for me, I will send it up
the stack for processing.” Or two: “if the data is not for me then I’ll regenerate the
signal and send it to my neighbor.” Once the originating computer receives back
the original token it sent out, that token can be released, and another one created.
Token Ring networks combine the advantages of the star and bus topologies
through use of a physical star, logical ring. A ring topology looks like a star and
the ring is built inside of the hub device to bypass problem stations. Problem
isolation and network reconfiguration are definite issues in a ring topology.
Page 11
Module 3, Page 12
Mesh Topology
ÔThe mesh topology
ÔWAN topology
ÔBackbone redundancy
ÔNo single point of failure
ÔOther topology terms
¾ Point-to-point
¾ Multi-point
¾ Clouds
¾ Hybrid
Number of links = n (n-1)/2,

where n is the number of devices
The mesh topology is typically used in WANs or hybrid situations for redundancy.
There are multiple active paths in which routers can determine the best route. A
mesh topology is great for a backbone because there is no single point of failure.
However, cost and complexity make this topology impractical in most LANs,
especially those LANs with lots of devices.
Troubleshooting is more complicated in a mesh topology and fault tolerance is
maximized. However, it is difficult and expensive to install and reconfigure.
Depending upon the design, bandwidth may be wasted too.
Other topology types are point-to-point, multi-point, cloud, and hybrids of bus, star,
ring, and mesh.
Page 12
Module 3, Page 13
WAN/Network Transport Interfaces
ÔCurrently Supported
¾ Packet Over Sonet (POS)
¾ Ethernet
¾ Fast Ethernet
¾ Gigabit Ethernet
The BSR1000 currently supports a single 10/100 Ethernet interface, while the
BSR64000 supports POS and Gigabit Ethernet interfaces too.
Page 13
Module 3, Page 14
Ethernet
Mbps Twisted Pair

10BaseT To another hub
Baseband or switch
IEEE 802.3 1024 stations
Hub per network
100 meters
328 feet
Today’s Ethernet networks refer to a family of LAN implementations. The coax

implementations are 10Base2 and 10Base5, the UTP implementations are
10BaseT or 100BaseT, and the fiber implementation is 10BaseFL. The 10BaseT
implementation is installed in most networks today, although 100BaseT is used in
many new installations. The fiber implementations are more for distance and
security.
Ethernet 10BaseT standards allow for 10Mbps, baseband signaling, typically
using UTP cable. 10BaseT provides connectivity between the end user devices
and hubs or switches. The standard specifies 1,024 stations per network with an
end-to-end cable distance of 100 meters (328 ft). Practically speaking, 10baseT
should be kept to a maximum length of 90 meters. This will allow for a 10 meter
(total) patch cable between the wall jack and the workstation NIC for both ends.
Although 2-pair category 3 and category 4 cable support 10BaseT, many choose
to install category 5 cable to allow for growth into 100BaseT Fast Ethernet.
10Mbps and 100Mbps Ethernet may exist on the same LAN at the same time. If
you wish to integrate dual speeds, this is handled efficiently with switches.
Page 14
Module 3, Page 15
5-4-3 Rule
No Stations No Stations
R R
R R
30 Stations 30 Stations
30 Stations
185 meters x 5 segments = 925 meters
One Collision and Broadcast Domain
10Base2 abides by the 5-4-3 rule: 5 segments connected by 4 repeaters with a

maximum of 3 segments populated with nodes. Layer 1 devices such as
repeaters are used to extend segments and regenerate signals. However, these
devices perform no filtering which means they propagate collisions and
broadcasts. As you can see, Thinnet is not very scalable. Since each segment
between repeaters can only carry 30 hosts, there is a maximum of 90 hosts per
collision domain. Compare this with 10BaseT and newer technologies.
Thinnet is much more difficult to troubleshoot than other mediums, such as
10BaseT. When a connection is lost using Thinnet, although there is no central
connection point, one disconnection can render the entire network segment
unusable. Troubleshooting in a 10BaseT environment is much easier, because
failure is isolated (normally) between a workstation and a hub, or hub to hub.
Although the 5-4-3 Rule allows us to extend the cable segment with the two
Inter Repeater Links (IRLs) where there are no stations, the network is still
one big collision and broadcast domain. The hubs do not provide any
filtering, but instead repeat everything whether good or bad.
Page 15
Module 3, Page 16
Fast Ethernet
Mbps Twisted Pair

100BaseT To another hub
Baseband or switch
IEEE 802.3u
Hub
Fast Ethernet 100BaseT gives high-performance workstations 100Mbps access to

the servers. Although not used as a total desktop solution, it provides connectivity
between the access and distribution layers, and the distribution and core layers of
the network. Think of the access layer as the users, hubs, and switches at the
user layer (car dealership) that are connected to the distribution layer (car
distribution center) by way of routers, that are connected to the core layer (car
factory). The core layer consists of multilayer switches to make the network more
efficient. Today, 10Mbps/100Mbps solutions are popular at the access layer,
100Mbps solutions at the distribution layer, and 100Mbps/1000Mbps solutions at
the core.
Fast Ethernet 100BaseT standards allow for 100Mbps, baseband signaling, using
UTP or fiber cable with 100Mbps cards and hubs. The following are various
media implementations of 100BaseT:
100BaseTX 2 pair Cat 5 100 meters Star topology RJ-45
Connector
100BaseT4 4 pair Cat 3, 4, 5 100 meters Star topology RJ-45
Connector
100BaseFX multi-mode fiber400 meters Shared MIC (ST)
multi-mode fiber2000 meters Point-to-point
single-mode fiber2k – 10k
Page 16
Module 3, Page 17
Fast Ethernet (cont.)
• Built on Ethernet principles

• Runs at 100 Mbps
• Uses same frame types, lengths, and formats
• Still CSMA/CD
• Same MAC layer
•New physical layer
100 Mbps
100 10 Mbps
Mbps
100 Mbps
Fast Ethernet or 100BaseT is based on CSMA/CD.

Fast Ethernet can use existing 10BaseT cabling.
Fast Ethernet reduces the duration of time each bit is transmitted by a factor of 10.
This enables the packet speed to increase tenfold from 10 Mbps to 100 Mbps.
Data can move from 10 Mbps to 100 Mbps without protocol translation or changes
to application and networking software.
Fast Ethernet also maintains the 10BaseT error control functions as well as the
frame format and length.
Fast Ethernet can run over UTP and fiber.
Page 17
Module 3, Page 18
100 Mbps
200 Mbps
100 Mbps
In full-duplex operation, 100 Mbps is available in each direction.
Full duplex provides bi-directional communication, meaning that 100 Mbps is

available for transmission in each direction.
By adding another pair of wires and removing collision detection, the connection
speed is doubled.
In the terms of Fast Ethernet, essentially 200 Mbps of throughput is the theoretical
maximum per full-duplex Fast Ethernet connection.
This type of connection is limited to a node-to-node connection and is typically
used to link two Ethernet switches together.
Page 18
Module 3, Page 19
II am
am sending
sending data
data
10 Mbps at
at 10
10 Mbps.
Mbps.
II am
am sending
sending data
data 100 Mbps
at
at 100
100 Mbps.
Mbps.
Allows devices at each end of a network link to automatically exchange

information about the link capabilities
The autonegotiation function is an optional part of the Ethernet standard that

makes it possible for devices to exchange information about their abilities over a
link segment. This, in turn, allows the devices to perform automatic configuration
to achieve the best possible mode of operation over a link.
At a minimum, autonegotiation can provide automatic speed matching for multi-
speed devices at each end of a link Multi-speed Ethernet interfaces can then take
advantage of the highest speed offered by a multi-speed hub port.
Page 19
Module 3, Page 20
Gigabit Ethernet
Mbps Copper
1000BaseCX
Baseband LX Fiber To another switch
IEEE 802.3z SX
Switch
Fast Ethernet offers a speed increase ten times that of the 10BaseT Ethernet
specification, while Gigabit Ethernet offers ten times that of 100BaseT Ethernet.
They are both based on an extension to the IEEE 802.3 specification.
Gigabit Ethernet is fast becoming a standard for campus backbones. In 1998 and
early 1999, the IEEE ratified the 802.3z standards for Gigabit Ethernet over fiber
(1000BaseLX and 1000BaseSX). The copper standard is specified in 802.3ab.
Page 20
Module 3, Page 21
Gigabit Ethernet Protocol

Architecture
IEEE
IEEE 802.2
802.2 LL LC
LC IEEE 802.3z
Gigabit Ethernet
IEEE 802.3 IEEE
IEEE 802.3
802.3
Ethernet CSMA/CD
CSMA/CD IEEE
IEEE 802.2
802.2 LLC
LLC
IEEE802.3
IEEE802.3
Physical CSMA/CD
CSMA/CD or
or Full-Duplex
Full-Duplex
Physical Layer
Layer
Media
Media Access
Access Control
Control (MAC)
(MAC)
FC-4
FC-4 Upper
Upper Layer
Layer 8B/10B
8B/10B Encode/Decode
Encode/Decode
Mapping
Mapping
FC-3
FC-3 Comm
Comm Serializer/Deserializer
Serializer/Deserializer
Service
Service
ANSI X3T 11 Connector
Connector
FC-2
FC-2 Signaling
Signaling
Fibre Channel
FC-1
FC-1
Encode/Decode
Encode/Decode
FC-0
FC-0 Interface
Interface
and
and Media
Media
802.3z includes specifications of Media Access Control, topology rules, the

Gigabit Media Independent Interface and three physical layer interfaces using a
common link encoding (1000Base-SX, 1000Base-LX and 1000Base-CX. 802.3z
was approved as a standard in June 1998.
802.3z is not a stand-alone document, it includes both changes and additions to
the 802.3 standard to add 1000BM/s operations to the previously defined 10 and
100 Mb/s speeds of operation. The IEEE 802.3 standards catalog page lists the
most current document that includes the specification for Gigabit Ethernet
operations.
Draft 802.3ab adds a Gigabit Ethernet physical layer (1000Base-T) for operation
over Category 5 unshielded twisted pair writing to the 802.3 standard.
Page 21
Module 3, Page 22
Twisted Pair Cable
ÔUnshielded Twisted Pair

Ô Shielded Twisted Pair
Ô Ethernet and Token Ring
Ô American Wire Gauge
Ô Standards
¾ EIA/TIA
¾ IBM
Twisted pair cabling is used in Ethernet and Token Ring networks. The solid or
stranded twisted pairs are insulated in an outer jacket for protection. Individual
pairs are twisted to assist with reducing crosstalk and other EMI sources. The
number of twists and their spacing are specified by industry standards. The pairs
may or may not be individually shielded. The shielding or braid may just be within
the outer jacket. Plenum and PVC types are available.
Wire thickness is measured by the American Wire Gauge (AWG). As thickness of
the conductor increases, the AWG number decreases. For example, 14 AWG is
physically larger in diameter than 22 AWG. The Electronic Industries
Association/Telecommunications Industries Association (EIA/TIA) and
International Business Machines (IBM) are among the standards bodies of the
cable industry.
Twisted pair cabling is divided into two main categories:
Unshielded Twisted Pair (UTP)
Shielded Twisted Pair (STP)
Page 22
Module 3, Page 23
Unshielded Twisted Pair (UTP)
ÔConnectors
¾ RJ-45 (4 pr)
¾ RJ-11 (2 pr)
Ô Connectivity
¾ Patch cable to wall jack
¾ Wall jack to punch down block
¾ Punch down block to patch panel
¾ Hubs to patch panel
Ô 10BaseT, 100BaseT
The most common uses of UTP include telephone wiring and local area networks.
There are various categories of cabling with category 5 being the most popular
cable type for new installations. It is designed to reliably support more than
100Mbps over 24 AWG wire. It has more twists per foot and better insulation
material than category 3. Category 3 may already be installed for voice, but it may
only support up to 10Mbps, depending upon the installation method.
Although UTP suffers from attenuation and EMI, it is a very practical and
economic choice for today’s networks. RJ-45 connectors are used to make patch
cables that connect the node to the wall jack or hub. The wall jack is connected to
a punch down block, typically located in one or more wiring closets throughout the
facility. The punch down block is connected to a patch panel that provides
connectivity for hubs, printers, and hosts. RJ-45 connectors are typically used for
LAN connectivity, while RJ-11 connectors support phones. RJ-45 houses 8 wires
or 4-pair, and RJ-11 houses 4 wires or 2-pair. One pair is for transmitting and the
other for receiving.
Common implementations of UTP include 10BaseT and 100BaseT (Fast
Ethernet).
Page 23
Module 3, Page 24
Unshielded Twisted Pair

Connectors
Wire Pair:
T is Tip
Pin R is Ring
1 Pair 3 T2
2 Pair 3 R2
1
3 Pair 2 T3
4 Pair 1 R1
8 5 Pair 1 T1
6 Pair 2 R3
7 Pair 4 T4
The RJ-45 Connector 8 Pair 4 R4
Page 24
Module 3, Page 25
UTP Straight-Through
Cable 10BaseT/
Straight-Through Cable
100BaseTX Straight-Through
8 1
Hub/Switch Server/Router
1 8
Pin Label Pin Label
1 TD+ 1 8 1 8
1 RD+
2 RD- 2 TD-
3 TD+ 3 RD+
4 NC 4 NC
5 NC 5 NC w g w b w o w br w g w b w o w br
g o b br
g o b br
6 TD- 6 RD-
7 NC 7 NC Wires on cable ends
8 NC 8 NC
are in same order.
A straight-through cable is wired pin 1 to pin 1, pin 2 to pin 2, pin 3 to pin 3, and
pin 6 to pin 6. The other wires are typically connected for strength and future use.
Tx is for transmit and Rx is for receive.
A straight-through cable is required to connect the following:
PC to Hub
PC to Switch
Switch to Router
Typically, straight-through cables are used to connect unlike devices (e.g., a
server to a switch, a switch to a router, etc.).
Page 25
Module 3, Page 26
UTP Crossover
Cable 10BaseT/
100BaseT Crossover Crossover Cable
Hub/Switch Hub/Switch 8 1
Pin Label Pin Label 1 8
1 RD+ 1 RD+ 8 1 8 1
2 RD- 2 RD-
3 TD+ 3 TD+
4 NC 4 NC
5 NC 5 NC br
w ww w b o g
br w g w b w o w
6 TD- 6 TD- br b g o g br o b
7 NC 7 NC
8 NC
Some wires on cable
8 NC
ends are crossed.
A crossover cable is wired pin 1 to pin 3, pin 2 to pin 6, pin 3 to pin 1, and pin 6 to
pin 2. The other wires are typically connected for strength and future use. Tx is
for transmit and Rx is for receive.
A crossover cable is required to connect the following:
PC to PC (or Server)
Switch to Switch
Hub to Switch
Router to Router
Hub to Hub
PC to Router
Page 26
Module 3, Page 27
Fiber Optic Cable
Ô Single mode 2km-10km point-to-point

Laser
Core Cladding
LEDs
Ô Multi-mode 400m shared/2km point-to-point

Ô Distance
Ô Security
Fiber is a physical medium capable of conducting modulated light transmission.

Compared with other transmission media, fiber-optic cable and its components
are more expensive. However, fiber is not susceptible to Electro-Magnetic
Interference (EMI), and is capable of higher data rates.
There are single-mode and multi-mode specifications. Single mode fiber optic
cable has higher bandwidth than multimode fiber, but it requires a monomode
laser light source. It has a narrow core that allows light to enter only at a single
angle. Multi-mode fiber optic cable supports propagation of multiple frequencies
of light using light emitting diodes (LEDs).
100BaseFX uses two strands of multi-mode fiber optic cable per link for shared
media. To guarantee proper signal timing the link cannot exceed 1,312 feet
(400 meters) in length. Point-to-point uses single-mode fiber optic cable and is
rated up to 2 kilometers. This is based on the IEEE 802.3u standards.
Page 27
Module 3, Page 28
Fiber Optic Cable Composition
Loose
Configuration
Cladding Tight
Core Configuration
Plastic Strength
Sheath Wires
Special tools are required to install and test fiber optic cable to ensure it is within
certain specifications. Fiber can be run within a building or between buildings,
and must be protected accordingly. The glass or plastic conducting core is
protected by a cladding and plastic sheath that is either surrounded by kevlar
strengthening strands or a protective gel within the outer jacket.
Fiber-to-the-desktop is still not a viable solution in many environments due to the
associated costs. Many of these costs are hidden. It is not enough to just price a
roll of fiber cable. Installation, troubleshooting, and component costs must be
considered. For example, compare the cost of NICs, hubs, switches, routers, etc.
that support fiber optic cable.
Page 28
Module 3, Page 29
Coaxial Cable Composition
Outer Jacket Shielding
Dielectric Insulator
Conducting
Core
Ô Familiar technology
Ô Reasonable security
Ô Voice, data, video
Ô RG-59 family of cables (RG-6)
¾ Cross section of a typical piece of 75ohm
coaxial Cable
» Used to deliver signals to the home
Coax cable is a familiar technology with reasonable security that is moderately

priced and primarily used in Ethernet physical bus topologies. The shielding
makes it more resistant to crosstalk (noise) and attenuation (signal loss over
distance) than twisted pair. It is good for voice, data, and video over long
distances. The outer protective jacket is either made of polyvinyl chloride (PVC)
that gives off poisonous gasses when it burns or is plenum rated for fire
resistance.
Coax cable comes in many varieties including:
RG-8 or RG-11 Standard Ethernet (50 ohm impedance)
RG-58/U (solid) (50 ohm impedance)
RG-58 A/U (stranded) (50 ohm impedance)
RG-58 C/U (military spec of RG-58 A/U) (50 ohm impedance)
RG-59 or RG-6 Cable TV (75 ohm impedance)
RG-62 ArcNet (93 ohm impedance)
Typical coaxial cable is designed to conduct signals along the
center conductor. The shielding, dielectric insulator and protective
jacket are designed to keep unwanted signal from bleeding into (or
out of) the system and to protect the center conductor from damage.
Multiple frequencies are transmitted on the same wire and is
separated in such a way as to minimize their interfering with one
another.
Page 29
Module 3, Page 30
The RF Spectrum
ÔElectronic communication is the transmission, reception and
processing of Information (voice, video, data) over media
ÔElectromagnetic energy can be oscillated at different frequencies
allowing for multiple “channels”
¾ The RF spectrum ranges from 1 cycle (Hertz) to light
¾ In the United States, the r/f spectrum is controlled by the government
(FCC)
» The FCC assigns numbers or letters to different “channels”
An electronic communications system is comprised of three basic components:

A transmitter
A transmission medium – air, cable, fiber
Receiver(s)
The electron flow in an ac circuit is continually reversing. Each time it
reverses it is said to be alternating. Two of these alternations is said to be
a cycle.
Page 30
Module 3, Page 31
The RF Spectrum (cont.)
ÔChannel assignment was

needed as a standard for
receivers
ÔSections of the spectrum
are divided into different
Arts & Entertainment
applications
High-Speed Data #1
High-Speed Data #2
High-Speed Data #2
High-Speed Data #1
Telephony Data #1
Telephony Data #2
¾ Radio
Telephony Data
Impulse PPV
¾ Broadcast TV
PBS
HBO
CNN
Satellite
NBC
¾
ABC
CBS
5 15 40 52 550 750 ¾ Microwave

Return Frequencies Forward Frequencies ÔThe off air spectrum is
controlled by the FCC and
separated by geographical
distances
VHF TV channels are separated from the FM band and the UHF TV channels.
The FCC set up geographical boundaries for off-air broadcasters. Adjacent TV
channels are never in the same market area to prevent interference. This is
because as signals bounce off the ionosphere, they can impact other channels on
the same frequency in different locations. In fact, if conditions are right a channel
500 miles away can be picked up off an antenna.
Page 31
Module 3, Page 32
The RF Spectrum (cont.)

HAM 10mHz HAM
Off Air
HAM HAM
paging services
15mHz land land land land

CB
mobile mobile mobile mobile
wwv wwv
inter
nat'l
shortwave shortwave shortwave
short
wave
5mHz 7mHz 14mHz 21mHz 27-27.5mHz 30.5-32mHz 36mHz
CATV Return 28-29.7mHz 34mHz 38mHz

40mHz
DATA 1 DATA 2 Telephony
Status
IPPV IPPV
Monitoring
ÔOff-air and CATV r/f spectrum utilize the same frequencies

¾ Support different channels
ÔCable is a closed circuit infrastructure
¾ Keeps the off-air signals out
¾ All the available bandwidth can be used for transporting services that are
chosen by the operator not the FCC
In an off-air environment the spectrum between 5-42Mhz contains short wave,

citizens band, WWV signals. That same spectrum in CATV systems may contain
video and data signals.
In a perfect world none of the signals on a CATV system escape to the air waves
and no off-air signal is introduced to the cable plant outside of the headend. For
the most part the FCC does not dictate the channel line-up on CATV systems.
The exceptions are:
The must-carry rule
If a cable operator can pick up an off-air channel, within signal
limits, it must be carried on the cable system.
Off-air channels in a given geographic area must have the same channel
assignment on a cable system sharing that same area.
Page 32
Module 3, Page 33
Off-Air vs. Cable
Ô CATV systems have advantages over

off-air signals
¾ Noise and other random disturbances
can be managed better
¾ Signal levels at all points can be
calculated
¾ Through unity gain signals are kept at
a constant/known level
¾ Digital signals such as QAM are not
reliable over the air
» Allows higher throughput in CATV
Off-air signals can be affected by:

•Atmospheric conditions – rain, fog, cloud cover
Rain – attenuates signals
Terrain – blocks signals
•Electrical disturbances – poor power line insulators, welding machines
Because QAM signals us both amplitude and phase modulation off-air amplitude
related disturbances can cause unrecoverable errors.
Page 33
Module 3, Page 34
Amplitude and Wavelength
Maximum Positive Amplitude

Amplitude
Wavelength (λ) Maximum Negative Amplitude
A carrier signal is a frequency in a communications channel modulated to carry

analog or digital signal information.
It can be described by it’s attributes:
Amplitude – the maximum absolute value (positive or negative) reached
by a voltage or waveform
Wavelength – the distance between two points in the same phase in
consecutive cycles of a wave
Page 34
Module 3, Page 35
Frequency and Wavelength
• Frequency is a measurement of number of cycles a

wave undergoes in a fixed time period
• Wavelength is the distance between two points in the
same phase in consecutive cycles of a wave
• When using a fixed time period as a reference:
Higher frequencies
will have shorter
ÔLower frequencies will wavelengths
have longer wavelengths
By definition, higher frequencies go through more cycles within the time same
period as lower frequencies.
Page 35
Module 3, Page 36
Frequency, Amplitude and

Wavelength
Amplitude y
qu enc
F re
Tim
e
View of Time domain

Measurements View of Frequency
Domain
Measurements
As mentioned before, a signal can be characterized by it’s attributes. One way to

look at an electrical signal is to use an oscilloscope to see how the signal varies
with time. This is very important information; however, it doesn't give you the full
picture.
To fully understand the performance of your device/system, you will also want to
analyze the signal(s) in the frequency-domain. This is a graphical representation
of the signal's amplitude as a function of frequency The spectrum analyzer is to
the frequency domain as the oscilloscope is to the time domain.
In the time domain, all frequency components of the signal are summed together
and displayed. In the frequency domain, complex signals (signals composed of
more than one frequency) are separated into their frequency components, and the
level at each frequency is displayed.
Page 36
Module 3, Page 37
Waves
ÔMeasuring Waves
¾ Period
» Shortest time a point takes to return to
initial position
λ » Represented by symbol T
¾ Frequency
» Number of vibrations per second
λ » Measured in hertz (Hz)
» Represented by symbol f
» f= 1/T
¾ Wavelength
» Shortest distance between peaks or
troughs
» Represented by symbol λ
» Velocity=λf
¾ Amplitude
» Distance from crest to equilibrium
» Used to measure energy of wave
Page 37
Module 3, Page 38
Attenuation
• Attenuation is the loss of energy by a signal over the

course of it’s propagation
• This means a loss in a wave’s amplitude
It takes more energy to travel the same distance at higher frequencies. The
higher the frequency the greater the loss (attenuation). A signal at 860Mhz will
lose more level than a signal at 50mHz when it is passed through an equal length
of cable.
Page 38
Module 3, Page 39
Attenuation (cont.)
• Lower frequencies travel down the middle of the center

conductor
• Higher frequencies travel on the surface of the conductor
• Called the Skin Effect
Higher frequencies travel along the surface of the conductor. This is called the
Skin Effect and makes the higher frequencies on a CATV system more
susceptible to imperfections and attenuation. The result of this phenomenon is
greater signal loss as the frequency increases.
For example, the signal loss for channel 116 (750 MHz) is over 4 times greater
than the loss for channel 2 (55 MHz) passed through the same coaxial cable.
Page 39
Module 3, Page 40
Modulation
ÔModulation
¾ Process of imprinting carrier wave with information
» Amplitude
» Frequency
» Phase
Since carrier waves, by definition, have known, unchanging attributes, any

changes to those attributes can be measured by a receiver and interpreted as
carrying imprinted information. Any of these attributes can be altered individually,
or, in combinations.
Page 40
Module 3, Page 41
Analog Signal
Ô An analog video signal

consists of three
distinctive carriers:
¾ Visual
¾ Color
¾ Aural
Ô These three carriers
comprise a video
channel and occupy a
6 MHZ bandwidth.
Typically, an analog television signal will use amplitude modulation (varying the
amplitude of a wave to propagate information) for it’s visual information.
Frequency modulation (varying the frequency of a carrier wave to encode it with
information) is used for the aural sub-carrier. Color information is typically sent
using a combination of amplitude and phase modulation.
Page 41
Module 3, Page 42
Amplitude Modulation
ÔAM (Amplitude Modulation)

¾ The energy of the signal
changes with the modulation
Carrier » Average power power of
the carrier changed as a
function of baseband
signal
Baseband Signal
Amplitude Modulated Carrier
Although AM signals travel further distances over the air they are susceptible to
electrical interference such as poor insulators on high tension lines. AM signals
are also attenuated by terrain as encountered in the loss of signal through tunnels.
FM signals can be recovered by the receiver at much lower levels.
Page 42
Module 3, Page 43
Baseband Signaling
ÔDigital signaling is common

ÔOne signal frequency at a time
ÔBi-directional communications
ÔUses repeaters (amplifiers) to regenerate the signal
ÔExamples—10Base2, 10Base5
A B C A B
B
C
There are two major types of transmission: baseband and broadband. Baseband
signaling typically uses digital signaling and broadband typically uses analog
signaling. Analog implies a continuous varying operation, in contrast with digital
which is broken up into numbers or discreet states.
Baseband transmission as used in Ethernet is bi-directional. The entire capacity of
the cable is taken up by a single transmission. Repeaters are used to regenerate
the signals.
Examples include 10Base2 and 10Base5.
Page 43
Module 3, Page 44
Broadband Signaling
Ô Integrated data, voice, and video
Ô Analog signaling is common
Ô Half-duplex or unidirectional
Ô Multiple simultaneous frequencies
Ô Frequency Division Multiplexing
Ô Uses amplifiers to boost signal
Ô Examples—Cable TV
A
B
C
Broadband signaling integrates data, voice, and video across multiple

simultaneous frequencies. This is commonly referred to as Frequency Division
Multiplexing (FDM). It is half-duplex or unidirectional and amplifiers are used to
boost the signal.
Think about how cable TV works. Many stations are broadcast at once, but when
you turn on the TV, you can only watch one station at a time. Computers do the
same thing. Multiple sources can transmit at once on different frequencies,
provided the receivers can tune in to only one signal frequency at a time.
Page 44
Module 3, Page 45
Frequency and Phase Modulation
ÔFM (Frequency Modulation)

¾ Amplitude doesn’t change
Carrier ¾ Frequency of the carrier is
changed as a function of the
baseband signal
ÔPM (Phase Modulation)

Baseband Signal ¾ Shift in phase of carrier
signal
Frequency Modulated Carrier
Although AM signals travel further distances over the air they are susceptible to
electrical interference such as poor insulators on high tension lines. AM signals
are also attenuated by terrain as encountered in the loss of signal through tunnels.
FM signals can be recovered by the receiver at much lower levels.
Page 45
Module 3, Page 46
Digital Signal
ÔDigital signals are

comprised of binary bit
streams that are
modulated in various
degrees of efficiency
¾ FSK - lowest efficiency
¾ QPSK - more efficient
» Reliable
3mHz 3mHz
¾ QAM - most efficient
6mHz
» Less robust
The more efficient the modulation type, the more susceptible it is to imperfections
in the physical plant. Factors such as modulation error rate (carrier to noise), are
critical to the success of using digital signal.
Digital signals are capable of carrying more video channels in a 6 MHZ bandwidth
than analog. Six digital channels can be transmitted in the same amount of
bandwidth it takes to send one analog channel.
FSK (Frequency Shift key) signals are robust but can only provide low bandwidth.
They are used on CATV plant for Impulse pay-per-view and status monitoring
systems.
QPSK (quaternary phase shift key) is a angle modulated constant amplitude
signal.
QPSK can function in fairly poor carrier-to-noise and carrier-to-interference
environments.
QAM (Quadrature Amplitude Modulation) is a form of modulation where the digital
information is carried in both the amplitude and phase of the transmitted carrier.
Although QAM is capable of handling higher data rates it needs a higher carrier-
to-noise and carrier-to interference ratio than QPSK signals.
Page 46
Module 3, Page 47
Analog to Digital Conversion
Although r/f waves are analog in nature, by selecting set, discrete amplitude
ranges to map to digital symbols, it becomes possible to modulate an analog
carrier with digital information.
Page 47
Module 3, Page 48
Bi-Phase Shift Keying (BPSK)

Binary Data
1 1 In Phase
Amplitude
0
180° Out
-1 0 of Phase
• Simplest form of digital modulation

• Data is transmitted by reversing the phase of the carrier
• Amplitude of the carrier remains constant
• Robust transmission method
Consumes significant bandwidth.
As an introduction to digital modulation, we will look at BPSK. The phase of the

carrier is altered by 100 degrees to indicate a binary data point and left in phase to
indicate a different binary value. Because there are only two possible
transmittable values, it is a highly reliable method of transmission. For that same
reason, it consumes a great deal more bandwidth than QPSK or QAM modulation,
which we will examine later in this module.
Page 48
Module 3, Page 49
Bits and Symbols
Bits
• Symbol
Group of bits 3 [10] Symbol
(Group of Bits)
Represented by combining
amplitude and phase 1 11 In Phase
Amplitude
modulation
The more amplitude levels, the 0
higher the symbol rate -1 01 180° Out
of Phase
-3 00
A higher data rate can be achieved by adding amplitude modulation to the carrier
in addition to the phase modulation. By having multiple levels of amplitude,
groups of bits can be transmitted. A group of bits that is represented by a
particular level and phase of the carrier is called a symbol.
Page 49
Module 3, Page 50
Bits and Symbols (cont.)
• Bits grouped into symbols

• Phase and amplitude modulated and transmitted
Symbol 00 10 01 11 10 01 00
Amplitude -3 +3 -1 +1 +3 -1 -3
Bits are grouped into symbols and the appropriate phase and amplitude is
modulated and transmitted.
Page 50
Module 3, Page 51
I Channel
Carrier
Phase
+ = Carrier
Amplitude
Q Channel
Carrier
Phase 90° Carrier Phase Shift
Shifted
Quadrature Modulation is a method of amplitude modulation that allows two

channels to be carried at the same frequency effectively doubling the bandwidth
that can be carried. A form of Quadrature Modulation has been used for many
years in analog television to carry the two components of the color subcarrier. By
modulating two carriers at exactly the same frequency but shifted by 90° both the
amplitude and phase of the carrier is modulated.
Page 51
Module 3, Page 52
Constellation
• Signal is separated into two

independent components:
• Inphase (I)
• Quadrature (Q) Q Channel
• The location on the
Constellation determines the I
and Q components amplitude I Channel
• The amplitudes of the I and Q
channel are derived from the
rectangular coordinates of the
carriers amplitude and phase
Page 52
Module 3, Page 53
Constellation (cont.)
• Polar display
• Provides simple way to view amplitude and phase
• Signal interpreted relative to the carrier
• Magnitude represented as the distance from the center
• Phase as the angle
e
ud
nit
ag
Phase
M
0 degrees
Page 53
Module 3, Page 54
Constellation (cont.)
• Signal changes
• AM changes the magnitude only
• PM changes the phase only
• Can be used together
e
ud
nit
Phase
ag
Phase Phase
0 degrees
M
0 degrees 0 degrees
Page 54
Module 3, Page 55
Constellations
• The phase and amplitude of the carrier at any given time

determine the location on the Constellation and the I and Q
channels can be derived from this information.
Location On
Constellation
Phase of
Carrier
Amplitude
of Carrier
Page 55
Module 3, Page 56
Constellations (cont.)
• The symbol which each location on the constellation represents is

chosen using a technique known as Gray Coding.
• Gray coding insures that any adjacent location will only be one bit
different, reducing the effect of an error.
10 11
00 01
Page 56
Module 3, Page 57
64 and 256 QAM Constellations
• By adding more levels to the I and Q channels, higher

data rates can be carried
The higher the number of levels, the more effect there will be
from noise or interference
64 QAM uses 8 levels in the I direction and 8 levels in the Q
direction for a total of 8 squared or 64 symbols
256 QAM uses 16 levels in the I direction and 16 levels in the Q
direction for a total of 16 squared or 256 symbols
Page 57
Module 3, Page 58
64 and 256 QAM Constellations

(cont.)
64 QAM Constellation 256 QAM Constellation

Q Amplitude Q Amplitude
15
7
13
11
5
9
3 7
5
1 3
-7 -5 -3 -1 1 3 5 7 I Amplitude 1
-1 -15-13 -11 -9 -7 -5 -3 -1 1 3 5 7 9 11 13 15 I Amplitude
-1
-3
-3
-5
-5 -7
-9
-7 -11
64 Possible -13
-15
Combinations of I and Q
256 Possible
Combinations of I and Q
Page 58
Module 3, Page 59
Decision Boundaries
• Each location on the constellation is framed by

decision boundaries
• If the signal falls within these boundaries, the correct data
will be received
• If because of noise or other interference it falls in an
adjacent area the data will be in error
1
-7 -5 -3 -1 1
-1
-3
Correct Locations Fall -5

Within Decision
-7
Boundaries
Locations in Error Fall Out

Decision Boundaries
Page 59
Module 3, Page 60
Constellation Display
• On a test instrument the locations on the constellation build

up over time
• The shape and distribution are indicators of types of impairments
-7 -5 -3 -1 1
-1
-3
Buildup of -5
Dots
-7
Page 60
Module 4, Page 0
Module 4
Hardware Installation and Boot-up
Rev. 1.0 11/01
Page 0
Module 4, Page 1
Introduction
Introduction We will begin by examining both the

BSR64000 and BSR1000. We will focus on
physical installation and boot-up processes as
a means of understanding their architectures.
Importance Knowing how to install and understanding the

boot processes will enable us a baseline to use
for troubleshooting implementations.
Lesson Overview We will begin by exploring the physical aspects

of the BSR64000 and BSR1000, including
interfaces, cards and installation. We will also
examine the boot process.
Page 1
Module 4, Page 2
Objectives
Objectives Upon successful completion of this module, you

will be able to perform the following tasks:
ÔRack and power a BSR1000 and BSR64000
ÔInstall cards in the BSR64000 chassis
ÔIdentify the port and resource modules on a
BSR64000
ÔDescribe the BSR64000 boot-up process
¾Examine the status of resource modules
Page 2
Module 4, Page 3
Topic Order
Topic Order Installation Tasks Page 5

Chassis Page 6
Current Resource Modules Page 7
Installing Resource Modules Page 8
Installing I/O Modules Page 9
Architecture Page 10
SRM Resource Module Page 11
SRM Resource Module LEDs and
Alarms Page 12
SRM I/O Module Page 13
POS Resource Module Page 14
POS I/O Module Page 15
CMTS Resource Module Page 16
CMTS I/O Module Page 17
Page 3
Module 4, Page 4
Topic Order (cont.)
Topic Order Gigabit Ethernet Resource Module Page 18

Gigabit Ethernet I/O Module (back) Page 19
50 A Variable Speed Fan Trays Page 20
Fan Control Curve Page 21
Rack Mounting Page 22
Installing DC Power Source Page 23
BSR64000 DC Connections Page 24
Console Connection Page 25
Show Chassis Status Page 26
Boot Process Page 27
Boot-Up Console Messages Page 29
Show Version Page 33
Page 4
Module 4, Page 5
Installation Tasks
Ô Basic Installation tasks:

¾ Racking
» Mounting and securing in rack
¾ Installing modules (BSR64000)
» Resource Modules
» I/O Modules
¾ DC power (BSR64000)
» Connecting grounding cables
» Connecting power cables
Page 5
Module 4, Page 6
BSR64000 Chassis
Redundant CMTS
(optional)
SRM Resource
Modules
FAN Assemblies POS/Gig E/CMTS

Resource Modules
The BSR 64000 uses a midplane design that pairs a resource module with a
matched I/O (Input/Output) module. There are 16 slots, labeled slots 0 through
15, in the BSR64000 chassis.
Resource Modules perform most of the tasks associated with processing data
while I/O modules provide physical connections to the networks (with network
interfaces – i.e., POS, r/f, etc.).
The BSR 64000 supports the following Resource Module and I/O Module pairings:
•Supervisory Resource Module (SRM) and SRM I/O module (must be placed in
slots 7 and 8)
•DOCSIS 1:4 Resource Module and I/O Module (redundant DOCSIS must be
placed in slot 6)
•OC3 POS Resource Module and I/O Module
•Gigabit Ethernet Resource Module and I/O Module
The only other existing limitation is that slot 15 has a special design consideration.
Slot 15 will accept and work fine with a CMTS module in non-redundant mode
ONLY. RF redundancy is NOT supported in slot 15. A CMTS module in that slot
will not be protected by Motorola’s leading RF redundancy solution. The RF
backplane does not support a redundant connection to this slot and the redundant
CMTS I/O housed in slot 6 does not support this slot.
Page 6
Module 4, Page 7
BSR1000 Interfaces
Ô The BSR1000 has fixed LAN, WAN and user interfaces
• The user interface is for local configuration

and troubleshooting
The WAN interface is for

connection to the I.P. network and
the necessary servers for modem
operation
The LAN interface is on the RF

side and connects the CMTS to
the modems
The DOCSIS reference architecture specifies the bounds for these interfaces.
The LAN interface consists of one downstream and four upstream r/f connections.
These are connected to the HFC plant and provide the connectivity to CPE
through cable modems.
The 10/100 Ethernet port acts as the WAN/MAN interface. It provides for I.P.
connectivity to the headend, access to DHCP, provisioning and Time of Day
(ToD) servers.
Each port is given a label of a port type and port number. From example “cable
0/0.” To configure a fixed interface, you would specify the interface by type and
numbering using this convention.
A DB-9 pin RS-232 interface allows for direct console access.
Page 7
Module 4, Page 8
Current Resource Modules
ÔChassis and resource modules that are currently shipping

¾ Chassis
» Redundant DC power distribution and stepping
» Variable control fans with multiple controls
• Redundant fan assemblies (either one can cool fully loaded chassis)
» RF Protection Switching Backplane
» 50 amp chassis
¾ SRM
» S/W upgradeable with reboot
» One active 10/100 port for management or data (slow path support only)
¾ CMTS
» 1x4 module
¾ POS HSIM (High Speed Interface Module)
» 1 Active OC3 POS interface
» 1 Active 10/100 Ethernet
¾ Gigabit Ethernet HSIM
» 1 port Gigabit Ethernet NIM (Network Interface Module)
Page 8
Module 4, Page 9
Installing Resource Modules
Switch Actuator and

Module Ejectors
ESD Front Jack
Captive Screw
ESD Wrist Strap
To install a new Resource Module:

1. Connect an antistatic wrist strap to the ESD grounding jack located at the
lower left corner of the chassis front, then put the strap on one of your wrists.
2. Loosen the two captive screws on the slot cover and remove the slot cover.
Make sure to save the slot cover for reuse.
3. Remove the module from its packaging, and hold it vertically so that you can
read its labeling.
4. Extend the module ejectors away from the module front panel.
5. Place the module edges into the slot guides, then push the module toward the
back of the chassis until it stops
6. Push the module ejectors toward the front panel, until you meet resistance, to
mate the module firmly with its midplane connector.
7. Push the switch actuator up to apply power to the module.
8. Tighten the captive screws using a Phillips screwdriver. Alternate between
screws when tightening them to allow the module to seat properly in the
backplane.
Page 9
Module 4, Page 10
Installing I/O Modules
ESD Socket (Rear)
ESD
Wrist
Strap
Captive Screws
To install a new I/O Module:

1. Connect an antistatic wrist strap to the ESD grounding jack located at the
lower left corner of the chassis front, then wrap the strap around one of your
wrists.
2. Loosen the two captive screws on the slot cover and remove the slot cover.
Make sure to save the slot cover for reuse.
3. Remove the I/O module from its packaging, and hold it vertically so that you
can read its labeling.
4. Place the module edges into the slot guides, then push the module firmly into
the midplane connector.
5. Tighten the captive screws using a Phillips screwdriver. Alternate between
screws when tightening them to allow the module to seat properly into the
midplane.
6. Connect all the appropriate cables for the specific I/O Module installed.
Page 10
Module 4, Page 11
Architecture
HSIM1 HSIM2 HSIM3 High Speed Network Interface Modules

Wire rate flow forwarding per interface module
Supervisory
Supervisory Supervisory
Supervisory Fully redundant route calculation/
Routing
Routing Routing
Routing
Module Module switch fabric in Supervisory Routing
ModuleAA ModuleBB
Module (SRM)
DOCSIS1 DOCSIS2 DOCSIS11 DOCSIS13 High density DOCSIS

modules
1xN redundant
1XN RF Switch
Automatic RF line switchover
Primary Midplane Connection

RF
RF11 RF
RF22 RF
RF33 RF
RF44 RF
RF1313
Secondary Midplane Connection
The BSR 64000 features not only a distributed architecture but also dedicated
hardware IC’s (Integrated Circuits) to handle various tasks such as packet
forwarding and classification. It distributes the packet handling intelligence and
control throughout the platform. Other, conventional systems use a centralized
processor scheme where all packet handling and management must go through a
single, dedicated CPU in order to be properly managed. The BSR 64000
distributes route forwarding, classification and filtering to each module such that
most traffic never goes through a single, central processing engine.
All modules connect to the primary and secondary SRMs via redundant Ethernet-
based control buses that allow the flow of control and management information
from the SRM to the resource modules. The two central slots in the chassis are
reserved for the primary and secondary SRMs to enable 1:1 redundancy. Typical
configurations will include a Network Interface Module for upstream metropolitan
or wide area connectivity and a Network Interface Module for high-speed LAN
connectivity to server applications/caches and up to 13 CMTS Modules.
The system provides 2.1 Gbps, bi-directional, non-blocking data packet transport
between the modules in a cross-connect manner, thus offering a redundant, 64
Gbps fabric. Also, it is important to note that there is no processor involved in
normal packet forwarding, filtering or classification decisions. These processing-
intensive tasks are all done in dedicated FPGA's (Field Programmable Gate
Arrays).
Page 11
Module 4, Page 12
SRM Resource Module
Must go in reserved slots 7&8
Contains:
• NVRAM
• Non Volatile Storage
• 64MB FLASH
• Console Connection
• Stratum 3 clock
The Supervisor Resource Module (SRM) uses a Motorola MPC750 PowerQUICC

II Microprocessor (with a bus speed of 66 MHz and a core speed of 266 MHz )
that acts as the master and runs the operational software and a Motorola
MPC8260 PowerQUICC II Communications Processor (with a bus speed of 66
MHz, a core speed of 166 MHz and a communication edge speed of 133 MHz
leading to 140 million instructions/ second) acts as a system and memory
controller handles several functions during operation, specifically:
•Runs all routing protocols supported by the BSR 64000 and provides forwarding
information to the other resource modules
•Serves as a centralized control system -- managing the other Resource Modules
installed in the BSR 64000 chassis
•Hosts the system’s 16x16 2.0Gb/s switch fabric that provides the connectivity
between modules installed in the BSR across the midplane
•Supports NVRAM and removable flash (PCMCIA card) to buffer syslog
messages and the operating software image
The SRM provides two E1/T1 interfaces that are used primarily to receive
synchronization timing from a Building Integrated Timing Supply (BITS). BITS
timing information received over the E1 or T1 network is used for synchronization
distribution in the BSR64000 system.
The SRM interfaces with each CMTS and Network Interface Module in the
BSR64000 system through fast Ethernet communication links.
Page 12
Module 4, Page 13
SRM Resource Module LEDs and

Alarms
FACE PLATE LEDs
LED SIGNATURE FACE PLATE :

LED ORIENTATION
State Fail Status Alarm
Fuse Fail or On Off Off

Module Inserted
but latch is open
FAIL
Module In Reset On On On
STATUS
Running Off On Off
(No alarms) ALARM
FAN TRAY TOP POWER, FAIL

Red and green LEDs assembly, green LED indicating top fan tray has power, red
LED indicating top fan tray has an error.
FAN TRAY BOTTOM POWER, FAIL
Red and green LEDs assembly, green LED indicating bottom fan tray has power,
red LED indicating bottom fan tray has an error.
MINOR ALARM, MAJOR ALARM, CRITICAL ALARM
Amber and 2 red LEDs assembly, amber LED indicating minor alarm, middle red
LED indicating a major alarm and the top red LED indicating a critical alarm.
Page 13
Module 4, Page 14
SRM I/O Module (back)
Alarm
T-1 Bits Clock
Input A
Input B
Console Port
10BaseT Port
Associated with the SRM is the SRM I/O module. The SRM I/O module is a
passive device that provides the interfaces to make physical network and other
external connections to the SRM. Physical connectors on the SRM I/O module
include the following:
•One female DB-25 pin connector for connecting to an external alarm panel
•Two RJ48 T1 interfaces for BITS clock inputs
•One female DB-9 pin RS-232 connector (DTE) for connecting a console monitor
to the BSR 64000
•One female RJ-45 10BaseT Ethernet connector for connecting the BSR 64000 to
a management Ethernet network
Page 14
Module 4, Page 15
POS Resource Module
• Any slot other than 7&8
Contains:
• NVRAM
• FPGAs
• One for each set of POS
ports
• LEDs
• Fail
• Status
• Alarm
• Sonet Ports
• OK – Fail Port 0
• Ethernet Port
• Link – Fault
• Critical
The OC3 POS Resource Module provides one high speed OC3 SONET interfaces
and one 10/100BaseT interface for the BSR 64000. Each module contains a
hardware assisted packet forwarding and QoS engine that processes up to 3
million packets per second using a Motorola MPC8260 PowerQUICC II
processor with a bus speed of 66 MHz, a core speed of 166 MHz and a
communication edge speed of 133 MHz leading to 140 million instructions/
second.
A set of LEDs located on the front panel of the module provide a visual indication
of the status of the module itself as well as its individual ports. LED indicators
•Module LEDs: Fail, Status, Alarm
•Per port LEDs (SONET ports 0) Link, Fault
•Ethernet port LEDs: Link, Fault
Page 15
Module 4, Page 16
POS I/O Module (back)
ENET
Port 0
SONET
Port 0
Matched with each OC3 POS Resource Module installed in the system is an OC3
POS I/O Module. The I/O module is a passive module providing physical network
connections for the OC3 POS Resource Module. The I/O module provides the
following physical connectors:
•One active LC optical connector
•One active female RJ-45 10/100Base-T connector
Page 16
Module 4, Page 17
CMTS Resource Module
Contains:
• NVRAM
Integrated • FPGAs
Upconverter •PowerPC750
•Tuners
•Upconverter
Monitor Ports
Port 0 (R/F) Upstream

Tuners
Port 1 (I/F)
The DOCSIS 1:4 CMTS Resource Module provides the BSR 64000 with one
downstream and four upstream broadband HFC network interfaces. Combined
with its I/O module the DOCSIS 1:4 CMTS Resource Module handles all network
traffic transmitted onto the HFC cable plant and received from it using an Intel
PowerPC750 processor running at 400 MHz.
A set of LEDs located on the front panel of the module provide a visual indication
of the status of the module itself as well as its individual ports. LED indicators
•Module LEDs: Fail, Status, Alarm
•Downstream port LEDs (Port 0) Link, Fault
•Upstream port LEDs (Ports 0 through 3) Link, Fault
•Two Type F connectors accessible on the module front panel enable RF (radio
frequency) and IF (intermediate frequency) signal monitoring.
The 1:4 DOCSIS module contains the HFC network interface for the system. The
module supports all DOCSIS modulation modes, implements per-flow packet
classification, applies QoS policies and performs distributed packet forwarding.
Page 17
Module 4, Page 18
CMTS I/O Module (back)
Upstream
Port 0
Port 1
Port 2
Port 3
Downstream
Port 0
Associated with each DOCSIS 1:4 CMTS Module is an I/O module. The I/O
module is a passive device that provides the physical connectors required for the
resource module. The DOCSIS 1:4 I/O module provides the following physical
connectors:
•One Type F connector for Downstream Port 0
•Four Type F connectors for Downstream Ports 0 through 3
Page 18
Module 4, Page 19
Gigabit Ethernet Resource Module
Contains:
• NVRAM
• FPGAs
• Software Upgradeable
• LEDs
• Fail
• Status
• Alarm
•Optical Interface
• Link
•Fault
The Gigabit Ethernet Resource Module features Motorola’s FPGA based Packet
Forwarding/ processing architecture, which allows feature enhancement without
change of the hardware (software upgradeable FPGAs).
It also has the MPC8260 based controller which is responsible for normal
operation of the Gigabit Ethernet Module.
The board itself has the following memory structures:
•32 MB of on-board flash memory
•64 MB of SDRAM
•128 MB of Packet Buffer
Page 19
Module 4, Page 20
Gigabit Ethernet I/O Module (back)
Gigabit Ethernet I/O card Interface:

1000 Base-Lx , Single Mode 1310 nm,
Short Range
Page 20
Module 4, Page 21
Processors
Ô MPC8260
¾ Next generation of the MPC860 PowerQUICC(tm) device
¾ Provides higher performance in all areas of device operation,
» Including greater flexibility, extended capabilities, and higher integration
Ô PowerPC 750
¾ Low-power 32-bit implementations of the PowerPC Reduced Instruction Set
Computer (RISC) architecture
» Features dedicated L2 cache interface with on-chip L2 tags
» Fully JTAG-compliant.
The MPC8260 PowerQUICC II™ is the most advanced integrated communication

microprocessor ever designed for the telecommunications and networking
markets.
The MPC8260 PowerQUICC II can best be described as the next generation
MPC860 PowerQUICC, providing higher performance in all areas of device
operation, including greater flexibility, extended capabilities, and higher
integration.
The MPC8260 integrates two main components, the embedded G2 core and the
Communications Processor Module (CPM). This dual-processor architecture
consumes less power than traditional architectures because the CPM offloads
peripheral tasks from the embedded G2 core. The CPM simultaneously supports
three fast serial communications controllers (FCCs), two multichannel controllers
(MCCs), four serial communications controllers (SCCs), two serial management
controllers (SMCs), one serial peripheral interface (SPI) and one I2C interface.
The MPC750 and MPC740 processors are low-power 32-bit implementations of
the PowerPC Reduced Instruction Set Computer (RISC) architecture. The
MPC750 and the MPC740 processors differ only in that the MPC750 features a
dedicated L2 cache interface with on-chip L2 tags. Both are software-compatible
and bus-compatible with the MPC603e and MPC604e microprocessor families,
and the MPC740 is pin-compatible as well. MPC750/740 microprocessors are
fully JTAG-compliant.
Page 21
Module 4, Page 22
50A Variable Speed Fan Trays
Ô Fan Controller PCB (PCA-0028-03) can be swapped in for Non-Controller PCB (PCA-
0023-02)
Ô Non-Controller Trays (ASM-0002-XX) can be swapped out for Controller Trays (ASM-
0003-01)
Ô Integrated handle allows for easier removal
Ô Fan Guards (Top and Bottom of tray) allow for safe removal
Ô Designed as a FRU (Field Replaceable Unit)
Top Fan Tray communicates via I2C bus to Slot7 SRM and is Master when Slot7
is Active, while the Bottom Fan Tray communicates via I2C bus with Slot8 SRM
and is Master when Slot8 is Active.
The private I2C bus between Top Fan Tray and Bottom Fan Tray allows trays to
communicate with each other and stay synchronized via on-board
Microprocessor.
Page 22
Module 4, Page 23
Fan Control Curve
Control Curve
X
RPM 3360RPM, 45C (113F)
X
2688RPM, 37.5C (99.5F)
X
2016RPM, 30C (86F)
Temperature
ÔOn Board sensors (Highest Temp) on Fan Controllers and

Ô SRM sets the speed of the Fans based on Control Curve
The Fan Trays go to High Speed Mode when a failure occurs (i.e. Failed Fan,
Failed Tray, Failed I2C Communication) and an alarm is sent to the SRM.
Otherwise, they follow a linear control curve within the BSR64000s operational
environmental range.
Fan Tray LED Status Reporting on SRM:

There are two LED’s for each of the fan trays (total of Four LED’s) located on the
SRM’s front panel under the heading “FAN STATUS”. The Fan Trays will be
noted on the SRM as “TOP” and “BOT” for their respective locations within the
chassis.
Page 23
Module 4, Page 24
Rack Mounting
Ô To mount the BSR 64000 in an

equipment rack, perform the following
tasks:
1. Mark the location on the rack where you
intend to mount the BSR 64000
2. Mount and secure the BSR 64000 in its
equipment rack
Integral mounting brackets
1. Lift and hold the BSR 64000 at its intended position in the rack and align the
BSR 64000 mounting bracket holes with the mounting holes of the equipment
rack.
2. Secure the BSR 64000 in the equipment rack using eight mounting screws.
Tighten the screws using a screwdriver.
• For mid-mount installation, attach the mounting brackets (angle brackets) to

the sides of the BSR 64000 chassis and follow the same set of directions. It is
recommended that three people should perform the rack mounting procedure;
two to hold the BSR 64000 in position while the other secures it in the rack.
Page 24
Module 4, Page 25
Installing DC Power Source
Ô To provide DC power for the BSR 64000 perform the following

tasks:
1. Disable DC power source
2. Connect grounding cables to the BSR64000
3. Connect power cables to the BSR64000
Always connect the grounding cables first. The chassis has a set of two
grounding terminal posts located at the rear. These terminal posts are the primary
grounding point. To ground the BSR 64000:
•Locate the grounding terminal posts on the BSR 64000.
•Remove the KEP nuts and locking washers from the accessory bag.
•Place a grounding cable lead over the grounding terminal posts.
•Place KEP nuts and locking washers on each post, then tighten.
To connect power cables to each Power Entry Module:
•Locate the Power Entry Modules. Facing the rear of the BSR 64000
chassis, Power Entry Module A is on the left and B is on the right.
•Remove the four screws holding the clear plastic shield using a flathead
screwdriver, then remove the shield from the chassis
•Remove the nuts and washers from Module A’s terminal posts.
•Facing the terminal posts, place the positive cable lead on one of the
bottom posts of the pair labeled RTN. Replace the locking washer and
nut. Tighten the nut to secure the connection.
•Place the negative cable lead on the top post of the pair labeled -48VDC.
Replace the locking washer and nut. Tighten the nut to secure the
connection.
•Repeat Steps 2 through 6 for power unit B.
•Place the clear plastic shield over the terminal posts and secure it in
place by tightening the four screws with a flathead screwdriver.
Page 25
Module 4, Page 26
BSR64000 DC Connections
48VDC
48VDC
Return
Notes:
For NEBS compliance, each -48VDC lead should split to connect to each lug.
For full -48VDC redundant power sources, connect each source to both the A&B inputs separately.
Do not install hard bridges between power lugs.
NEBS stands for Network Equipment Building System and is a physical

requirement and testing standard written by Bellcore which the telephone
companies require their equipment suppliers to comply with. The requirements
ensure a product installed in a Central Office is safe, robust, and reliable. When
installing or working with DC power, follow these guidelines:
•The BSR 64000 is equipped with a DC terminal block. You must terminate the
DC input wiring on a DC source capable of supplying at least 60 A. A 90 A circuit
breaker is required at the -48 VDC facility power source. Be sure to connect the
grounding wire conduit to a solid earth ground. A closed loop ring is
recommended to terminate the ground conductor at the grounding terminal stud.
•Run two wires from the circuit breaker box to a source of -48 VDC. Use
appropriate gauge wire to handle up to 90 A. Ensure that the polarity of the DC
input wiring is correct. Under certain conditions, connections with reversed polarity
might trip the primary circuit breaker or damage the equipment.
•Because the BSR 64000 is a positive ground system, you must connect the
positive lead to the +RTN terminal, the negative lead to the –48V terminal, and the
earth ground to the chassis grounding points. Use a 3/8 in. (11 mm) nutdriver to
connect the leads to the terminal posts.
Page 26
Module 4, Page 27
Console Connection
Ô The BSR64000 uses Command Line

Interface (CLI) as a local user interface
Local connectivity is accomplished using a an RJ-45 to DB-9 adapter and a
terminal program on a p.c.
This interface allows for configuration of the BSR64000
DTE connector
•Allows the BSR to monitor CD (carrier detect) for a disconnect via a straight-through
cable
–The BSR64000 will disconnect the active session to prevent anyone else from gaining access
to the console port for security reasons
Over the course of this class, we will use a (crossed-over) db-9 adapter attached
to a (straight-through) cat-5 cable running from your p.c. to the console port of the
BSR64000. We will be using HyperTerminal as the terminal emulation program.
We will look at telnet access to the BSR64000 as well as the use of an SNMP
application to access configuration and performance information about the box.
The advantage to the local console access is that it is out-of-band. With SNMP
(or any remote management interface), there is some overhead associated
(network traffic).
In order to use HyperTerminal, ensure that your settings for console access are as
follows:
Bits per second- 9600
Data bits- 8
Parity- N
Stop bits- 1
Control Flow- none
Page 27
Module 4, Page 28
Show Chassis Status
ÔShow chassis status

¾ From the Privileged EXEC mode
» Show chassis status
This example shows the boot of CMTS (still in boot mode) with failed attempt to
contact SRM. This is an example of normal boot-up followed by the reboot of
CMTS modules. Typically, you will see one failure for all modules (with the
exception of the SRM), because these modules must wait for the SRM to indicate
that they are allowed to go active during the boot process.
Page 28
Module 4, Page 29
BSR1000 Boot Process
ÔPOST (power-on self test)

ÔFind software image
ÔFind and use startup-config file stored in nvram
¾ Provides operational parameters
» Device configurations
ÔDuring operation it uses its running-config.
¾ Changes to the running-config take effect immediately.
¾ Changes to the running-config will be lost unless saved to startup-
config.
There are three major operations involved in boot-up:

1. Hardware checking routines called power on self test (POST). This is a
series of diagnostic tests that run automatically whenever the device is turned
on. If the tests are successful, the BSR1000 boots itself. If the tests are
unsuccessful, the errors are reflected in the status lights on the front on the
CMTS.
2. Once the hardware passes the power-on tests, the BSR1000 performs it’s
startup routines. This is when the software image is found and initialized
(loaded and run).
3. Once the operating system is running, the devices try to find and apply the
configuration settings from the startup-config file.
The nvram file system contains the operating system and the startup configuration
for the BSR1000.
Care must be taken while making configuration changes to the BSR1000 during
its operational state. Changes in the operational parameters are immediate
and may cause the modems connected to the BSR1000 to re-synchronize to
the downstream or restart the registration process (which can be quite noisy
and cause problems).
Page 29
Module 4, Page 30
BSR1000 Boot-Up Console

Messages
VxWorks System Boot
Copyright 1984-1998 Wind River Systems, Inc.
CPU: RDN CMTS 740/750
Version: 5.4
BSP version: 1.1/0
Creation date: Oct 18 2000, 23:38:18
NVRAM:/ - Volume is OK
Press any key to stop auto-boot...
7
6
5
4
3
2
1
0
auto-booting...
VxWorks, from Wind River Systems, is a networked real-time operating system

designed to be used in a distributed environment. It runs on a wide variety of
hardware, including MC680x0, MC683xx, Intel i960, Intel i386, R3000, SPARC,
Fujitsu SPARClite, and TRON Gmicro, based systems.
It requires a host workstation for program development; supported host platforms
include Sun3, Sun4, HP9000, IBM RS-6000, DEC, SGI, and MIPS.
It does not run development systems software such as compiler, linker and editor
on the target machine. The development environment is based on cross-
development or remote-development method (typically run on a UNIX machine of
some sort (e.g. SUN's) to run the compilers and Debuggers).
The compiled application code can be downloaded to the target and runs as part
of the VxWorks image. During the development phase or thereafter, individual
object code can be downloaded dynamically to running target system. Finished
applications can be ROM'ed.
VxWorks was the name given the collection of software which ran on top of
various realtime kernels including VRTX and pSOS as well an earlier slower
version of WIND kernel. VxWorks no longer runs on other kernels; it now runs
exclusively on its own WIND kernel since the 5.0 release.
Page 30
Module 4, Page 31

Messages (cont.)
Start application from flash!
###################################################################
inflating standalone image success!
Starting at 0x100000...
CPU: RDN CMTS 740/750. Processor #0.
Memory Size: 0x8000000. BSP version 1.1/0.
WDB: Ready.
-> Starting Configuration of RDN CMTS
RDN CMTS Initializations Complete
Fpga memory base address: 80100000
UsRingBase: 0x5bbaa00, DsRingBase: 0x5bb9ff0
NVRAM:/ - Volume is OK
The application image (which includes a basic boot image) is inflated (it is stored
in a compressed format), tested and run. It will check, by default, at hardware
register 0x100000 for the start-up configuration.
Page 31
Module 4, Page 32

Messages (cont.)
BSR 1000R(tm) version 00.01.01.RA
Copyright (c) 2000 by RiverDelta Networks.
Compiled Thu Nov 16 16:26:31 EST 2000
MPC750 processor with 128MB memory.
Boot ROM: RDN CMTS Board BootRom 00.01.00.TA
CPU: MPC750
Memory Size: 128 MB
Board Program ID: RD
Format Version: 48
Assembly Type : 55
Part Number : PCA-0027-02
Serial Number : 0037B0025
Product Number:
Fabric Interface FPGA Version: 00000039
RDN>
The equivalent of a show version command is displayed, showing the application

and boot image as well as hardware version and serial number.
Page 32
Module 4, Page 33
BSR64000 Boot Process
ÔDiagnostic routines on all cards

ÔSRM loads from NVRAM or FTP/TFTP server
¾ POS and CMTS cards boot through diagnostics and wait for SRM
ÔSRM scans chassis for resource modules
¾ Activates Ethernet port
¾ Compares operational image of each module to NVRAM contents
» Loads any newer images it finds
¾ Images confirmed
» Final boot
ÔCMTS cards associated with POS card
ÔI/O ports activated
for the BSR64000.
its operational state. Changes in the operational parameters are immediate and
may cause the modems connected to the BSR64000 to re-synchronize to the
downstream or restart the registration process (which can be quite noisy and
cause problems) as well as disrupt network connectivity.
• Find and use startup-config file stored in nvram
•Provides operational parameters
•Device configurations
• During operation it uses its running-config.
•Changes to the running-config take effect immediately.
•Changes to the running-config will be lost unless saved to startup-
config.
Page 33
Module 4, Page 34
Boot-Up Console Messages
Interrupting system boot will leave you in the VxWorks real-time operating system.
Do not do this unless specifically instructed by a Global Technical Support
Engineer. If you should do this, type the “@” symbol to continue through the boot
process and load the start-up configuration.
Page 34
Module 4, Page 35
Boot-Up Console Messages (cont.)
When interrupted, the BSR can be configured to boot from an application image
located on an external ftp server. This is not necessary to do, since the same task
can be accomplished from the command line.
Page 35
Module 4, Page 36
ICP processes are enabled and resource cards are detected and booted.
Interfaces associated with each card are “created” or made active.
Page 36
Module 4, Page 37
The SRM controls the version of the application image run on all resource
modules. Should it detect that any of the cards are running and older (or newer)
version of code, it will update them so that all resource modules are running at the
same revision.
Page 37
Module 4, Page 38
Show Version
ÔVersion information
¾ show version [slot #]
» Can be shortened to sh ver
RDN> sh ver
…
…
The command could be reduced to the shortest recognizable version, or sh v.

This command displays the configuration of the system hardware, etc. We will
return to this command when we address troubleshooting, as it will provide much
of the information that RiverDelta Networks Global Technical Support will look for
when you call. Version information can provide clues to known bugs.
It is considered a best practice to keep a copy of the output from this command for
troubleshooting purposes.
Page 38
Module 4, Page 39
Show Version (cont.)
Slot Number
Boot Image Software Version
Application Image Software Version
Hardware Version
Mfg. Serial Number
By default a show version will display all slots. Several things can be found out
using this command, including the Part Number associated with firmware (cards),
the Image number (software version) and any FPGA version information (firmware
component versions).
Page 39
Module 4, Page 40
The chassis version is available from within the SRM Versions portion of this
output (it will begin with part number CHS-XXXX-XX).
Page 40
Module 4, Page 41
Page 41
Module 5, Page 0
Module 5
Unique Identification
Rev. 1.0 11/01
Page 0
Module 5, Page 1
Introduction
Introduction To move information efficiently, and to insure

that information is received by the intended
recipient, each manageable device must be
uniquely identified with an address. As there is
an addressing function for the Data link (Layer
2) and Network layers (Layer 3), we will look at
each address structure and how it is used.
Importance Understanding Layer 2 (MAC) and Layer 3 (IP)

address formats is essential when
implementing and troubleshooting networks.
This includes the HFC network, as cable
modems and cable modem termination
systems require both types of addresses for
proper operation and management.
Lesson Overview Topics covered include:

Layer 2 Addressing/ Layer 3 Addressing
Page 1
Module 5, Page 2
Objectives

ÔExplain the need for unique addressing in a
networked environment
ÔDescribe the importance and structure of IEEE
MAC addresses
ÔExplain the purpose and structure of IP
addressing
ÔDifferentiate between classes of IP addresses
¾Given an IP address, show which portion specifies
the network and which the host address
Page 2
Module 5, Page 3
Topic Order
Topic Order Datalink Layer Page 5

Unique Addressing Page 6
IEEE Committees Page 7
OSI Enhancements Page 9
Ethernet at Layers 1&2 Page 10
MAC address formats Page 12
Ethernet Frame Formats Page 14
MAC address operation Page 20
Network Layer Page 21
Understanding Binary Page 29
IP Address format Page 30
Masks Page 31
Address Resolution Page 36
Page 3
Module 5, Page 4
Topic Order (cont.)
Topic Order The Data Link Layer – Layer 2 Page 19

The Physical Layer – Layer 1 Page 20
Overhead Page 21
Encapsulation Page 22
How the Layers Work Together Page 23
DOD’s TCP/IP Protocol Suite Page 24
The Host-to-Host Model Page 25
Reference Architecture Page 28
DOCSIS Protocol Model Page 29
DOCSIS Protocol Stack Page 30
Layers and Devices Page 31
Devices in the HFC Page 32
Page 4
Module 5, Page 5
The Need for Unique Addressing
House
#3, Cool
!
4 3 2 1
Ethernet Segment Road, (Street)

Sulamita, MA 01879 (City, State)
It would be difficult, if not impossible to get mail where it was supposed to go, if
there wasn’t a method to uniquely identify the recipient. The post office uses a
street address, City , State, and zip code. There are similar addresses schemes
within the data communications world, which we will explore in more detail.
Page 5
Module 5, Page 6
Data Link Layer
OSI TCP/IP
Application 7
Session 5
Transport
Transport 4
(host-to-host)
Network 3 Internet
Data Link 2
Network Interface Ethernet, RF
Physical 1 Sonet
You should recall form module two, that communication is broken up into seven
distinct layers called the OSI model. When we talk about unique addressing, the
first place we look is layer two. Here is where we find the “house number’ of the
postal address. It is called the “MAC” address at layer two. Each layer of the
model is dependent on the layer above and below it. The are committees who
work on protocols and formats for each of the layers. Layer two is the concern of
the 802 committee as detailed on the next page.
Page 6
Module 5, Page 7
The IEEE 802 Committees
Ô802.1 Internetworking
Ô802.2 Logical Link Control
Ô802.3 CSMA/CD Ethernet
Ô802.5 Token Ring LAN
Ô802.6 Metropolitan Area Network
Ô802.7 Broadband
Ô802.8 Fiber-optic
Ô802.9 Integrated voice/data networks
Ô802.10 Network security
Ô802.11 Wireless networks
Ô802.12 Demand priority LANs (100 VG-AnyLAN)
The IEEE 802 Project was named for the year and month it began, 1980
February. It further defines the Physical and Data Link layers of the OSI model.
802.1 defines internetworking. It specifies the physical and data link standards
required for one network device to communicate with another network device on a
different LAN or WAN.
802.2 defines Logical Link Control (LLC). This identifies what upper layer protocol
is used inside the frame. It is used with 802.3, 802.4, 802.5, and 802.6 networks.
802.3 defines Ethernet and the Carrier Sense Multiple Access/Collision Detection
(CSMA/CD) protocol, baseband and broadband signaling, multiple media types,
physical topologies such as bus and star, and data rates.
802.4 defines Token Bus, a physical bus token passing topology that was
originally designed for factory automation.
Page 7
Module 5, Page 8
The IEEE 802 Committees (cont.)
Ô802.5 Token Ring LAN

Ô802.6 Metropolitan Area Network
Ô802.7 Broadband
Ô802.8 Fiber-optic
Ô802.9 Integrated voice/data networks
Ô802.10 Network security
Ô802.11 Wireless networks
Ô802.12 Demand priority LANs (100 VG-AnyLAN)
Ô802.14 Data over Cable
802.5 defines Token Ring. It is based on IBM’s original implementation of Token

Ring. It is a physical star, logical ring topology that operates at 4 or 16 Mbps over
a variety of media types.
802.6 defines Metropolitan Area Networks through a Distributed Queue Dual Bus
(DQDB) data transfer technology.
802.7 defines broadband communications.
802.8 defines fiber optic standards.
802.9 defines Isochronous Ethernet designed to integrate voice and data over the
same protocol.
802.10 defines security, in particular managing and distributing encryption key
information.
Page 8
Module 5, Page 9
OSI Enhancements
ÔLogical Link Control

¾ 802.2
¾ Defines Service Access Points
¾ Establish and maintain links
Ô Media Access Control
¾ Shared network access (access methods)
¾ Error checking
¾ Physical addressing
¾ 802.3
802.11 defines wireless transmission such as Spread Spectrum and infrared.

802.12 defines demand priority LANs. This 100 Mbps physical star topology first
started with Hewlett Packard (HP) as 100 VG-AnyLAN to support Ethernet and
Token Ring.
The IEEE divided the Data Link layer of the OSI model into two sublayers in the
802 project, LLC and MAC. The 802.2 LLC establishes and maintains links to
upper-layer protocols through Service Access Points (SAPs). MAC controls
media access on the wire, and is responsible for error checking and addressing.
SAP and SubNetwork Access Protocol (SNAP) are the two LLC frame types.
Source and destination SAPs (SSAPs and DSAPs) point to upper-layer protocols.
For example, a SAP of 06 hex is destined for IP, and a SAP of E0 hex is destined
for IPX.
Page 9
Module 5, Page 10
Ethernet at Layers 1 and 2
Network IP 802.1 Internetworking
802.2 LLC
Data Link
802.3 MAC
Ethernet II – DIX
100BaseFX
100BaseTX
(Type field)
100BaseT
10BaseT
10BaseF
10Base2
10Base5
Physical
The Physical and Data Link layers of the OSI model coordinate the basic
hardware and software communications in networking. For example, the Ethernet
Physical layer standards previously discussed have a Data Link counterpart for
communications to occur on the physical medium.
DIX Ethernet uses an EtherType field to point to the upper layer protocol. 08 00 is
IP.
IEEE Ethernet uses a length field and LLC SAPs to point to the upper layer
protocol. 06 is IP.
Page 10
Module 5, Page 11
Identifying Stations
B C
A
ÔEach host has a unique MAC address

ÔAddress that is built into network card
Media Access Control (MAC) addresses are hardware addresses. The term that is
used is “burned into the ROM.” The MAC addresses are normally not changed
once they are set into the ROM at the manufacturer. Generally speaking, it is
better to leave the assigned MAC addresses unless there is a compelling reason
to administratively change them. Regardless of the reason for changing a MAC
address, the administrator must ensure that all MAC addresses remain unique on
each local segment.
Page 11
Module 5, Page 12
IEEE MAC Address Format
6 bytes (48 bits)
Organizational Unique
Unique Identifier Number
3 bytes 3 bytes
(24 bits) (24 bits)
00-00-0c-11-22-33
What vendor? What machine?
The IEEE Media Access Control address is 6 bytes or 48 bits. 3 bytes identify the
vendor, and 3 bytes uniquely identify the card. Together they make up the MAC
address for the NIC which is commonly referred to as a hardware address,
Burned In Address (BIA), or Data Link address.
Vendors pay ($1250) to the IEEE for registering a block of addresses. It is not a
requirement to register a MAC address with IEEE, however, registered addresses
can be readily recognized by protocol analyzers and aid in standardization.
Every device on the LAN must have a unique MAC address. The first part (first 24
bits) is the Organizational Unique Identifier (OUI). The last 24 bits in the MAC
address is the vendor assigned serial number, and should be unique. Serial ports
do not have MAC addresses. Serial ports only have two ends and do not require a
MAC address to function properly. Duplicate MAC addresses on the local
segment will not properly resolve and WILL cause network problems. Consider
the example of house addresses on a particular street. If two or more houses had
the same address, the post office would never be able to deliver the mail.
Page 12
Module 5, Page 13
Some Common Vendor Codes
Ô 00-00-0c cisco
Ô 00-06-7c cisco
Ô 00-06-c1 cisco
Ô 00-10-1f cisco
Ô 00-00-1d Cabletron
Ô 00-00-52 Optical Data Systems
Ô 00-00-81 Synoptics
Ô 00-04-ac IBM
Ô 00-20-35 IBM mainframes, Etherjet printers
Ô 00-00-c0 SMC
Ô 00-20-85 3Com superstack
Ô 00-20-af 3com
For a more complete list see:

http://www.cavebear.com/CaveBear/Ethernet/vendor.html
Cisco presently owns 39 OUIs that are officially registered with the
IEEE. The highest registered value is E20C0F, which is registered
to Kingston Technologies. There are 899 OUIs listed as of March 9,
1999.
Page 13
Module 5, Page 14
Ethernet Frame Formats Overview
ÔEthernet II (DIX Ethernet)

¾ Uses Type Field to point to upper layer protocol
ÔIEEE 802.3 Ethernet
¾ Uses Length Field
¾ Does not use Type Field
¾ Uses 802.2 LLC SAPs to point to upper layer protocol
» SAP header
» SNAP header
A frame is a way of packaging data and different network types require different
frame formats. For example, Ethernet is different than Token Ring, which is
different than Frame Relay. Even within Ethernet, there are multiple frame
formats available. We will contrast and compare Ethernet II (DIX Ethernet) to
IEEE 802.3 Ethernet over the next couple of pages.
Page 14
Module 5, Page 15
Ethernet II (DIX) Frame Format
# Bytes 8 6 6 2 Variable 4
Preamble Dest add Source add TYPE Data FCS
MAC Address
Value > 0x5dc

or 1500 Decimal
The framing known as Ethernet II or DIX Ethernet is that used by the original 10
Mbps Ethernet and is still the main choice for transmitting IP frames across the
network. The IEEE 802.3 Ethernet variations specify a different frame format and
in fact there is more than one possible format within the IEEE 802.3 standard.
Regardless of the format in use, the basic Ethernet header is 14 bytes long. The
transmitting station will send a series of 64 bits consisting of alternating 0’s and
1’s. The last two bits will be consecutive 1’s. This signal is called the Preamble
and is not part of the actual Ethernet frame. It is used to signal to the receiving
stations(s) that a frame is coming and allows them to synchronize their receive
circuits to the incoming bit stream. The Destination and Source Addresses are
MAC addresses. The source is always a unicast address, whereas the
destination can be unicast, multicast, or broadcast.
The next field differs in use depending on whether the frame is an Ethernet II or an
802.3 frame. If Ethernet II framing is used, the next field is a type field. Since this
same field is used as a length field with IEEE 802.3 framing, it must be some
value that is not a valid packet length to be identified as an Ethernet II frame.
DIX Ethernet uses a 2-byte EtherType field to identify the upper layer data. 0800
(hex) is an IP datagram (IP version 4), and 0806 is an ARP datagram.
Page 15
Module 5, Page 16
Common EtherType Values
Ô 0800 IP
Ô 0806 ARP
Ô 8137, 8138 IPX (Novell)
Ô 81D6 LANtastic
Ô 81D7 LANtastic
Ô 86DD IPv6
Ô 0BAD Banyan
Ô 0BAF Banyan
Ô 6003 DecNet
Ethernet data packets range in size from 46 bytes to 1500 bytes. Note that in the
slide above, all of the type codes are greater than 1500 decimal (5dc hex). The
type codes are usually given in hex since most software that allows you to
examine a packet will display hex values. Since the maximum packet length for
Ethernet is 0x5dc (the leading 0x denotes this is a hexadecimal number), it is easy
to distinguish between Ethernet II frames and IEEE 802.3 frames. If the value of
the third field is some value greater than 1500 Decimal (5dc hex), it is an Ethernet
II frame and the field is a type field. The value in the type field identifies the layer 3
protocol that owns the data. If the value in this field is 1500 decimal or less, then
the field is a length field and the frame is an IEEE 802.3 frame. When the 14 bit
header and 4 byte FCS (frame Check Sequence) is added, we have a minimum
Ethernet frame size of 64 bytes and a maximum of 1518 bytes. Anything below
the minimum is considered a runt, and anything above the maximum is
considered a giant in Ethernet.
Reference www.cavebear.com/CaveBear/Ethernet/type.html for a more complete
list of EtherType codes.
Page 16
Module 5, Page 17
IEEE 802.3 Ethernet Frame Format
Preamble Dest add Source add Length Data FCS
MAC Address
Value ≤ 0x5dc
or 1500 Decimal
We have already stated that the third field is a length field if the framing is IEEE
802.3. The value must be between 46 and 1500. If the data is less than 64 bytes,
padding will be inserted achieve a minimum packet size. Remember also that we
stated earlier that there are different formats for framing within the IEEE 802.3
standard. Let’s take a look at a few of them.
Page 17
Module 5, Page 18

(cont.)
802.2 LLC 1 1 1 or 2 Variable

(SAP) DSAP SSAP Ctrl Data
header
MAC Layer — 802.3
The picture above is the IEEE 802.3 frame format with an 802.2 LLC header.
Reference Cisco and Novell web sites for more details on SAP numbers.
Unlike Ethernet II, IEEE 802.3 Ethernet uses the type header field as a length
field. This means that we must have some other means to identify the upper layer
protocol in use.
The 802.2 LLC (Logical Link Control) committee devised a scheme whereby a
part of the data field is borrowed and used to identify the upper layer protocol in
use. A one byte Destination SAP (Service Access Point) field and a one byte
Source SAP field are used to accomplish this. There is also a control field that has
limited use in today’s networks. The DSAP and SSAP fields fulfill the same
purpose for IEEE 802.3 frames as the type field does for Ethernet II frames. 06 is
the SAP for IP and E0 is for IPX.
To allow proprietary protocols in the 802.2 LLC Length field, IEEE defined the
SubNetwork Access Protocol (SNAP) frame format. This is kind of like putting the
type field back into the frame.
Page 18
Module 5, Page 19

(cont.)
1 1 1 or 2 3 2 Variable
802.2 LLC
Dest SAP Source SAP Ctrl OUI
(SNAP) Type Data
AA AA 03 ID
header
MAC Layer — 802.3
It would seem that the LLC committee has solved the problem of identifying upper
layer protocols in the 802.3 frame, however you may notice that the SAP fields are
only 1 byte (8 bits) long and 2 bits of the field are used for purposes other than
protocol identification. The remaining 6 bits are just not enough to handle the
needs of modern day networks, so another LLC scheme was devised. Additional
bytes were borrowed from the data field for what is called the Sub-Network
Access Protocol (SNAP). SNAP headers can be identified by confirming that there
is a length field in the header and that the fields normally used for DSAP and
SSAP contain the value AA. In the SNAP header, the field immediately following
the SSAP field is the control field, the next field is an OUI (Organizational Unique
Identifier) and the next is the type field. The OUI and type fields together identify
the upper layer protocol in use. OUIs are assigned to vendors in the same manner
as the OUIs in MAC addresses and are registered with IEEE.
Note that in both the LLC SAP and LLC SNAP schemes, the additional fields are
borrowed from the data packet. The basic frame header is still only 14 bytes long
and the maximum frame size is still 1,518 bytes including the FCS trailer.
Page 19
Module 5, Page 20
Ethernet MAC Address Operation
00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
B C
A
ÔTransmission from “A” to “C”

00-00-0e-33-33-33 00-00-0e-11-11-11 0800 Data
6 bytes 6 bytes 2 bytes Varies

DA SA
MAC addresses are generally confined to next-hop addresses. End-to-end

addressing is normally handled at layer 3, by IP addresses. The vendor code
specified above, 00000e is registered to Fujitsu. The EtherType field indicates that
an IPv4 datagram is expected at the Network layer.
Page 20
Module 5, Page 21

(cont.)
00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
B C
A
Ô All stations hear, but only station C pays attention

Ô Station B’s NIC will examine the Destination MAC
and immediately discard if there is no match
In real world operation, all stations on the local segment will see each frame. What
is significant is how early the frame gets discarded. For all stations other than C,
the frame is discarded when the destination MAC address is read.
Page 21
Module 5, Page 22

(cont.)
00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
B C
A
ÔStation C sees its own MAC, looks at the EtherType and hands
the packet to the appropriate upper-layer protocol (IPV4)
In each layer of the OSI reference model, there is a pointer to the next layer in the
protocol stack. In this instance, since the EtherType is 0800, the host knows to
expect IPv4 at the Network layer and is processed accordingly.
Page 22
Module 5, Page 23
A Special MAC Address
00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
B C
A
ÔMeans — “This packet is for all stations!”
FF-FF-FF-FF 00-00-0e-11-11-11 0800 Data
6 bytes 6 bytes 2 bytes Varies

DA SA
When a MAC address contains all F’s in the destination address, it is a broadcast.
All NICs receiving this special destination address pass the broadcast up the
protocol stack for processing. Broadcasts are not an uncommon occurrence on a
network.
It is a good idea to minimize broadcasts on networks because every station may
need to read each packet up to layer 4 before making a keep/discard decision.
Page 23
Module 5, Page 24
Broadcast Address — FF-FF-FF-

FF-FF-FF
00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
A B C
ÔNot discarded by the NIC

ÔAll stations process the packet
Broadcasts in general have a detrimental effect on the network if left uncontrolled.

Typically, many broadcasts on modern networks are unnecessary and can be
prevented by proper protocol conservation and network design. This is especially
true in Windows networking.
Layer 2 broadcasts are identified by a series of F’s, for example FF-FF-FF-FF-FF-
FF.
Page 24
Module 5, Page 25
Network Layer
OSI TCP/IP
Application 7
Session 5
Transport
Transport 4
(host-to-host)
Network 3 Internet IP,
IP,ICMP,
ICMP,ARP
ARP
Data Link 2
Network Interface
Physical 1
In module two, we spoke about the function of Layer 3 (network

Addressing). This is the City and State and street part of the postal
address presented at the beginning of this module.
Page 25
Module 5, Page 26
What is Internet Protocol?
•What is internet Protocol

•De facto standard
•Most widely used protocol in the world
•HISTORY
•Started 1969
•Adopted 1980 official internet Std.
•Why use IP?

•Most popular Layer 3 Standard
•Gain access to the internet
In the next few slides, we are going to explore the Internet Protocol
(IP)
Page 26
Module 5, Page 27
How does it stack up?
Application Telnet FTP HTTP TFTP DHCP SNMP
Presentation
Session
Transport Transmission Control Protocol User Datagram Protocol
Network ICMP IP ARP/RARP
Datalink
Any Physical and Datalaink Layer Protocols such as
Physical Ethernet, PPP, SONET, RF Broadband
OSI Model TCP/IP Stack
Page 27
Module 5, Page 28
IP Datagram Format
Type
Preamble Dest add Source add Data FCS
0800
MAC Address
# Bits 0 4 8 16 31
Version Header Type of
Length Service Total Length
Identification 0|1|2 Fragment

IP Flags Offset
Header Time to Live Protocol Header Checksum
Source IP Address
Destination IP Address
IP Options (if any) Padding
DATA
There are lots of things to look at inside an IP packet, but the most
significant thing for our purposes is the destination and Source IP
addresses. We know that it is an IP packet, by the type code in the
header.
Page 28
Module 5, Page 29
Understanding Binary
Base2
27 26 25 24 23 22 21 20
128 64 32 16 8 4 2 1
1 1 1 1 1 1 1 1 = 255
0 1 1 0 0 1 0 0 = 100
Before we go diving into the world of IP addresses, lets take a quick

review of the binary numbering system. An IP address is made up
of 32 bits, or four bytes of information. An IP address is
represented to us in a form called decimal dotted notation. That just
makes it easier for us to understand. In actual practice, each device
that deals with the address looks at all 32 bits, and doesn’t see
“128.128.128.1”, but sees
“10000000100000001000000000000001”. It’s all math!
Binary is Base2. Write out the place values.
27 26 25 24 23 22 21 20
128 64 32 16 8 4 2 1
0 1 1 0 0 1 0 0 = 100
Page 29
Module 5, Page 30
IP Address Format
32 bits
172 16 141 129

10101100 00010000 10001101 10000001
8 bits 8 bits 8 bits 8 bits
Ô32 = 4.2 billion addresses for IPv4

Ô0-255 (256 bit positions)
Ô4 Octets (bytes)
ÔDotted decimal notation/Binary notation
ÔClass B (How do we know?)
As previously stated, IP addresses are logical addresses at layer 3 of the OSI

model. They are 32-bits or 4-octets in length. Every device on an IP network that
wants to communicate needs a unique IP address. This includes hosts, router
interfaces, servers, printers, etc. Our focus is on IPv4 although IPv6 is in use on
the Internet. IPv6 is 128 bits and is hex whereas IPv4 deals with 32 dotted
decimal numbers.
IP addresses are hierarchical in nature. The network address uniquely identifies
each network (street) and every device on that network shares that portion of the
address. Each host (house) needs a unique node address on that network. In the
class B example above, 172.16.0.0 is the network, and 0.0.141.129 is the host.
The default subnet mask of 255.255.0.0 determines the network host division.
Page 30
Module 5, Page 31
Default Masks
Class A N H H H /8
255 0 0 0
Class B N N H H /16
255 255 0 0
Class C N N N H /24
255 255 255 0
N = Network
H = Host
The default mask acts like a network tape measure to identify the network portion
of the address. The term “mask” is somewhat ambiguous; tape measure is more
precise. We can actually measure off thenetwork by specifying (measuring) how
many bits are required for the network. The remaining bits will always be allocated
for host addresses. Think of the network portion of the address as the street you
live on with many houses. Each house is equivalent to a host.
Class A addresses use a default mask of 255.0.0.0 or
11111111.00000000.00000000.00000000 in binary, or /8 in bit-notation.
Class B addresses use the default subnet mask of 255.255.0.0 or
Class C addresses use the default subnet mask of 255.255.255.0 or
Page 31
Module 5, Page 32
Default Mask Example
Address:
172 16 141 129 /16
Mask:
11111111.11111111 00000000.00000000
255 255 0 0
Network:
172 16 0 0
The address on the top row denotes a class B address. Since the network
number for class B addresses includes the first two octets, we say that the default
mask is 16 bits.
In decimal notation, this is expressed as 255.255.0.0.
This is expressed in binary as 11111111.11111111.00000000.00000000.
In short-hand bit-count notation, it is also expressed as /16 after the IP address.
Each expression above means the same thing. They all signify that the network
number is in the first two octets (first 16 bits) and every number to the right of the
first two octets is a host value.
Page 32
Module 5, Page 33
Class Origin (Let’s Calculate!)
27 26 25 24 23 22 21 20 IP Address Class is determined

By the value of the first 3 bits in
128 64 32 16 8 4 2 1 The first byte of the IP Address
CLASS A 00000001 1
01111111 127 (loop back)
CLASS B 10000000 128

10111111 191
CLASS C 11000000 192

11011111 223
The first byte of the IP address determines the class. To be a little more specific
there are particular bits that denotes the class. The first bit is the most significant
for class A, the second bit for class B, and the third bit position for class C. An
easy way to remember this is A-B-C … easy as 1-2-3. (Thank you Jackson 5).
There are also class D and E addresses. Class D and E addresses are not
assignable to hosts. Class D addresses start at 224.0.0.0 and are used for
multicasting. Class E addresses start at 240.0.0.0 and are reserved for
experimental purposes.
Page 33
Module 5, Page 34
Networks and Hosts
• Class A 0nnnnnnn.h.h.h 255.0.0.0
27-2=126 networks 224-2=16,777,214 hosts
• Class B 10nnnnnn.8n.h.h 255.255.0.0
214 = 16,384 networks 216-2=65,534 hosts
• Class C 110nnnnn.8n.8n.h 255.255.255.0
221 = 2,097,152 networks 28-2= 254 hosts
Classes were designed with numbers of networks and hosts in mind. Class A
networks are found in extremely large organizations with the largest number of
node requirements such as Xerox or IBM. Class C is the other extreme. It is
reserved for organizations with smaller numbers of hosts. Class B is in the middle.
There are many methods for calculating binary arithmetic including charts, tables,
memorization, or calculators. Use what works best for you.
If you are using Windows there is a calculator program that has a scientific view
that allows conversions among decimal, binary, and hexadecimal.
Page 34
Module 5, Page 35
The Need for Address Resolution
192.168.10.1 192.168.10.2 192.168.10.3

00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
A B C
192.168.10.0 Network
•Each Network Interface Card (NIC) has a unique MAC address.
•The Ethernet segment shares the same Network Address
•Each Host Interface (Computer running the IP Protocol) has a Unique
IP address.
•The MAC address of the receiving station is needed to move the frame
across the Ethernet segment. Mapping must occur between the IP and
MAC addresses.
Before and IP packet can be delivered to a specific station, it’s MAC

address must be known. A protocol called ARP (Type 806) is used
to find out the MAC address of the intended destination host.
Remember, routing decisions are made on IP addresses, but the
MAC address must be known on a specific segment so the “frame”
can be addressed correctly.
Page 35
Module 5, Page 36
Address Resolution Protocol

(ARP)
PURPOSE
•Obtain the MAC address for the Destination IP host
•MAC Frame Type code (806)
MECHANISM
•ARP Request – Source sends to broadcast address
•ARP Reply – Destination returns it’s MAC address
Remember, before we can move an IP packet from one device to

another, the MAC address of that device must be inserted in the
datalink layer frame. The ARP protocol provides us with a
mechanism to get the needed information, and it is done
automatically. The two mechanisms are ARP request (I need your
MAC address) and ARP response (Here is my MAC address).
Page 36
Module 5, Page 37
ARP Request
192.168.10.1 192.168.10.2 192.168.10.3

00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
A B C
My IP address
I’ll ignore that, I’ll send an ARP
Broadcast
Not my IP address response
192.168.10.0 Network
What is the
MAC address
of
192.168.10.3?
•In this example, station A needed to know what MAC address to put in the frame
for station C. Station A knew the IP address, so sent out an ARP request to get
the address needed to send the frame on it’s way.
Page 37
Module 5, Page 38
ARP Reply
192.168.10.1 192.168.10.2 192.168.10.3

00-00-0e-11-11-11 00-00-0e-22-22-22 00-00-0e-33-33-33
A B C
Unicast to 192.168.10.1
192.168.10.0 Network
My MAC address is
Thanks! 00-00-0e-33-33-33
Now I can send I learned your address
the frame from your broadcast.
Thanks!
• Here, station a receives the ARP response from station C, and can now send the
frame to station Cs MAC address.
Page 38
Module 5, Page 39
Host Addresses
172.16.2.2 10.1.1.1
10.6.24.2
E1
172.16.3.10 E0 10.250.8.11
172.16.2.1
172.16.12.12 10.180.30.118
Routing Table
172.16 . 12 . 12 Network Interface
Network Host 172.16.0.0 E0
10.0.0.0 E1
Remember, there are two levels of addressing that apply to a communicating

device. The MAC address is used in the frame at layer two, and the IP address at
layer three. Remember, there is a Network and host part to every layer three IP
Address. Routers use the “Network” portion of the IP address to make forwarding
decisions. In the diagram above, there are two different networks. Network
172.16.0.0 and 10.0.0.0. Those networks are associated with the “wires” that are
attached to each port of the router. The router builds a a routing table based on
the “Network” part of the IP address.
Page 39
Module 5, Page 40
Determining Available Host

Addresses
Network Host
172 16 0 0
N
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
10101100 00010000 00000000 00000000 1
00000000 00000001 2
00000000 00000011 3
...
...
...
11111111 11111101 65534
11111111 11111110 65535
11111111 11111111 65536
– 2
2N – 2 = 216 – 2 = 65534 65534
Now that we know there is a network and host part to an IP address, let’s review
how we determine how many hosts addresses are available. Remember, there is
a default mask for each class of address (A,B,and C), so those bits that are not
part of the network are host bits. You can not use a host address of all “1’s”,
because that is considered part of the network broadcast address we will discuss
later. You can not use a host address of all “0’s, because zero in the host position
would indicate “this network”. We will look at how we figure all these things out in
the next few slides.
Page 40
Module 5, Page 41
IP Address Classes Exercise
Address Class Network Host
10.2.1.1
128.63.2.100
201.222.5.64
192.6.141.2
130.113.64.16
256.241.201.10
Lets take a minute, and figure out the network class, what the network number is,
and what the host is based on the exercise above. Don’t turn the page and peek!
;>)
Page 41
Module 5, Page 42
IP Address Classes Exercise

Answers
Address Class Network Host
10.2.1.1 A 10.0.0.0 0.2.1.1
128.63.2.100 B 128.63.0.0 0.0.2.100
201.222.5.64 C 201.222.5.0 0.0.0.64
192.6.141.2 C 192.6.141.0 0.0.0.2
130.113.64.16 B 130.113.0.0 0.0.64.16
256.241.201.10 Nonexistent
So, how did you do?
Page 42
Module 5, Page 43
Addressing Without Subnets
172.16.0.1 172.16.0.2 172.16.0.3 172.16.255.253 172.16.255.254
…...
172.16.0.0
ÔNetwork 172.16.0.0
With a single address as depicted on the slide, it would be impossible to have

65,534 host on one wire. It would also be undesireable, even if you could do that,
because of the competition for the ethernet bandwidth. It would be what is called
a “flat’ domain, with no routing. So, to avoid buying new addresses from IANA,
you have to find a way to break up your address space into multiple networks.
There is a mechanism to do just that. It is called subnetting, and we will address it
in the next few slides.
Page 43
Module 5, Page 44
Addressing with Subnets
172.16.3.0
172.16.4.0
172.16.1.0 172.16.2.0
ÔNetwork 172.16.0.0
This figure describes network structure when subnets are used.
The host bits of an IP address can be subdivided into a subnetwork section and a
host section. The subnetwork section in this example is the full third octet.
A subnetted address space is like a highway with exits. A network device uses a
subnet mask to determine what part of the IP address is used for the network, the
subnet, and the host ID. A subnet mask is a 32-bit value containing a number of
one bits for the network and subnet ID, and a number of zero bits for the host ID.
Given its own IP address and subnet mask, a device can determine if an IP
packet is destined for:
1) a device on its own subnet
2) a device on a different subnet on its own network
3) a device on a different network.
A device can determine what class of address the device has been assigned from
its own IP address. The subnet mask then tells the device where the boundary is
between the subnet ID and the host ID. I know this sounds a little confusing, so
let’s take a look at it in detail in the next few pages.
Page 44
Module 5, Page 45
Subnet Addressing
172.16.2.200 172.16.3.5
172.16.3.1
E1
172.16.2.2 E0 172.16.3.100
172.16.2.1
172.16.2.160 172.16.3.150
New Routing Table

Network Host 172.16.0.0 E0
172.16.0.0 E1
In this figure, notice that we are using the same network number on both sides of
the router. In reality, this won’t work as is, because, the router would believe that
network 172.16.0.0 was on both interfaces, and is not allowed. This is where
subnetting comes in. How could we differentiate the network on the left from the
network on the right so that the router could make a forwarding decision between
different networks?
Page 45
Module 5, Page 46
Subnet Addressing
172.16.2.200 172.16.3.5
172.16.3.1
E1
172.16.2.2 E0 172.16.3.100
172.16.2.1
172.16.2.160 172.16.3.150
New Routing Table

Network Subnet Host 172.16.2.0 E0
172.16.3.0 E1
By turning on more bits in the mask (extending the mask further than the default),
we then reserve some bits as network information and can use these bits to
describe subnetworks. By examining the routing table above, we can see that
the third octet was used for subnetting. Let’s take a closer look at how this was
done.
Page 46
Module 5, Page 47
Subnet Mask
Network Host
IP
Address
172 16 0 0
Network Host
Default
Subnet
Mask
255 255 0 0
11111111 11111111 00000000 00000000
• Also written as “/16,” where 16 represents the number of 1s
in the mask
Network Subnet Host
8-Bit
Subnet 255 255 255 0
Mask
• Also written as “/24,” where 24 represents the number of
1s in the mask
The following are the rules for IP addressing:

An address is 32 bits, divided into three components:
First octet rule bits (class of service (A,B,C,etc)
Network bits (path selection bits)
Node bits
The first octet rule states that the most significant bit pattern in the first
octet determines the class of the address.
Path selection bits cannot be all ones or zeros.
Certain addresses are reserved. RFC 1918 defines some of those.
Prefix or mask one bits are path selection significant; zero bits are host
bits and therefore not significant.
Use the logical AND to combine the address and mask bits to get the
subnet address.
Page 47
Module 5, Page 48
Decimal Equivalents of Bit

Patterns
128 64 32 16 8 4 2 1
0 0 0 0 0 0 0 0 = 0
1 0 0 0 0 0 0 0 = 128
1 1 0 0 0 0 0 0 = 192
1 1 1 0 0 0 0 0 = 224
1 1 1 1 0 0 0 0 = 240
1 1 1 1 1 0 0 0 = 248
1 1 1 1 1 1 0 0 = 252
1 1 1 1 1 1 1 0 = 254
1 1 1 1 1 1 1 1 = 255
Here is a table which will help you in determining values for subnet masks, and
also will help you understand which networks/subnetworks you can use as a result
of subnetting.
Page 48
Module 5, Page 49
Subnet Mask Without Subnets
Network Host
172.16.2.160 10101100 00010000 00000010 10100000
255.255.0.0 11111111 11111111 00000000 00000000
10101100 00010000 00000000 00000000
Network 172 16 0 0
Number
ÔSubnets not in use—the default
This figure illustrates how the “logical AND” function works. Although you could
use a calculator to do this, it is good to understand how networking devices
determine the network and host parts of an IP address. In this example, no
subnet mask is being used.
Page 49
Module 5, Page 50
Subnet Mask with Subnets
Network Subnet Host
172.16.2.160 10101100 00010000 00000010 10100000
255.255.255.0 11111111 11111111 11111111 00000000
10101100 00010000 00000010 00000000
128
192
224
240
248
252
254
255
Network
Number 172 16 2 0
• Network number extended by eight bits
This example makes a Class B address space look like a collection of Class C
address spaces.
Now the logical AND allows us to extract the subnet number as well as the
assigned network number. Remember from earlier discussions, where there is a
mask of all 1s, the value above it falls through as the same value. Where the
mask is 0’s, the value falls through as zero.
Page 50
Module 5, Page 51
Subnet Mask with Subnets (cont.)
Network Subnet Host
172.16.2.160 10101100 00010000 00000010 10100000
255.255.255.192 11111111 11111111 11111111 11000000
10101100 00010000 00000010 10000000
128
192
224
240
248
252
254
255
128
192
224
240
248
252
254
255
Network
Number 172 16 2 128
• Network number extended by ten bits
In this figure, we have extended the mask even further to include 2 bits of the next
octet.
Page 51
Module 5, Page 52
Subnet Mask Exercise
Address Subnet Mask Class Subnet
172.16.2.10 255.255.255.0
10.6.24.20 255.255.240.0
10.30.36.12 255.255.255.0
Based on the masks provided, complete the above exercise. Don’t turn the page
to check your answers until we review them! ;>)
Page 52
Module 5, Page 53
Subnet Mask Exercise Answers
Address Subnet Mask Class Subnet
172.16.2.10 255.255.255.0 B 172.16.2.0
10.6.24.20 255.255.240.0 A 10.6.16.0
10.30.36.12 255.255.255.0 A 10.30.36.0
How did you do?
Page 53
Module 5, Page 54
Broadcast Addresses
172.16.3.0
172.16.4.0
172.16.1.0
172.16.3.255 172.16.2.0
(Directed Broadcast)
255.255.255.255
(Local Network Broadcast)
X
172.16.255.255
(All Subnets Broadcast)
This figure explains shows how Network layer broadcast work. There are
basically three forms of Networl layer broadcast addresses:
Directed broadcast – to another subnet
Local broadcast – this subnet only
All subnets broadcast – all subnets
The following RFCs provide more information about broadcasts:

RFC 919, Broadcasting Internet Datagrams
RFC 922, Broadcasting IP Datagrams in the Presence of Subnets
Page 54
Module 5, Page 55
Addressing Summary Example
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 Mask
Subnet 4
Broadcast
First
Last
Lets take a look at how we can break this address down into hosts, broadcast
address, subnet, first and last host numbers:
Step 1 –Look at the 32 bit address
Page 55
Module 5, Page 56
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2
Subnet
Broadcast
First
Last
Step 2 – add the mask
Page 56
Module 5, Page 57
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2
Subnet
Broadcast
First
Last 7
Step 3 – Determine from the mask the number of host bits
Page 57
Module 5, Page 58
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2
10000000 Subnet 4
Broadcast
First
Last
Step 4 –Determine the subnetwork bits
Page 58
Module 5, Page 59
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2
10000000 Subnet 4
10111111 Broadcast
5
First 6
Last
Step 5 – By converting the host bits to all 1s, you have the subnet broadcast
address
Page 59
Module 5, Page 60
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2
10000000 Subnet 4
10111111 Broadcast
5
10000001 First 6
Last
Step 6 – the first host is determined by setting the least significant bit of the host
portion of the IP address
Page 60
Module 5, Page 61
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2
10000000 Subnet 4
10111111 Broadcast
5
10000001 First 6
10111110 Last 7
Step 7 – The last host is determined by setting all the host bits to one, minus the
last bit. What happens when the host bits are all 1s?
Page 61
Module 5, Page 62
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2

8
10101100 00010000 00000010 10000000 Subnet 4
10101100 00010000 00000010 10111111 Broadcast

5
10101100 00010000 00000010 10000001 First 6
10101100 00010000 00000010 10111110 Last 7
Step 8 –Again, do the math in binary, then determine the decimal equivilents. You
can use the handy chart from earlier in the module to help you out.
Page 62
Module 5, Page 63
172 16 2 160
172.16.2.160 10101100 00010000 00000010 10100000 Host 1
255.255.255.192 11111111 11111111 11111111 11000000 Mask 2

9 8
172.16.2.128 10101100 00010000 00000010 10000000 Subnet 4
172.16.2.191 10101100 00010000 00000010 10111111 Broadcast

5
172.16.2.129 10101100 00010000 00000010 10000001 First 6
172.16.2.190 10101100 00010000 00000010 10111110 Last 7
Step 9 –Convert your binary answers to decimal, and you should come up with
the above answers!
Page 63
Module 5, Page 64
Class B Subnet Example
IP Host Address: 172.16.2.121

Subnet Mask: 255.255.255.0
Network Network Subnet Host
172.16.2.121: 10101100 00010000 00000010 01111001

255.255.255.0: 11111111 11111111 11111111 00000000
Subnet: 10101100 00010000 00000010 00000000

Broadcast: 10101100 00010000 00000010 11111111
ÔSubnet Address = 172.16.2.0

ÔHost Addresses = 172.16.2.1–172.16.2.254
ÔBroadcast Address = 172.16.2.255
ÔEight Bits of Subnetting
Let’s review with the simple example above
Page 64
Module 5, Page 65
Subnet Planning
20
20 Subnets
Subnets
55 Hosts
Hosts per
per Subnet
Subnet
Class C Address:
Class C Address:
192.168.5.0
192.168.5.0
192.168.5.16
Other
Subnets
192.168.5.32 192.168.5.48
This example takes a little more forethought. What mask would we use to
accomplish the task above?
ÔHow many bits of subetting

•What are the range of host addresses
Page 65
Module 5, Page 66
Class C Subnet Planning Example
IP Host Address: 192.168.5.121

Subnet Mask: 255.255.255.248
Network Network Network Subnet Host
192.168.5.121: 11000000 10101000 00000101 01111001

255.255.255.248: 11111111 11111111 11111111 11111000
Subnet: 11000000 10101000 00000101 01111000

Broadcast: 11000000 10101000 00000101 01111111
ÔHost Addresses = 192.168.5.121–192.168.5.126

ÔBroadcast Address = 192.168.5.127
ÔFive Bits of Subnetting
ÔSubnet Address = 192.168.5.120
Here are the answers for the exercise.
Page 66
Module 5, Page 67
Broadcast Addresses Exercise
Address Subnet Mask Class Subnet Broadcast
201.222.10.60 255.255.255.248
15.16.193.6 255.255.248.0
128.16.32.13 255.255.255.252
153.50.6.27 255.255.255.128
Let’s do a couple more practice exercises. Answer these exercises before

proceeding to the next page.
Page 67
Module 5, Page 68
Broadcast Addresses Exercise

Answers
Address Subnet Mask Class Subnet Broadcast
201.222.10.60 255.255.255.248 C 201.222.10.56 201.222.10.63

15.16.193.6 255.255.248.0 A 15.16.192.0 15.16.199.255
128.16.32.13 255.255.255.252 B 128.16.32.12 128.16.32.15
153.50.6.27 255.255.255.128 B 153.50.6.0 153.50.6.127
These are the answers. If you are still having difficulty, review the previous
examples, and if you are still confused, speak with the instructor.
Page 68
Module 6, Page 0
Module 6
Initial Configuration
Rev. 1.0 11/01
Page
Copyright ©2002 0
Motorola
Module 6, Page 1
Introduction
Introduction In a previous module, we looked at the physical

installation and components of the BSR1000/64000,
as well as the boot process. We will now look at the
software components involved in a deployment.
Importance There are some basic configuration tasks that must

be accomplished before a BSR1000/64000 can be
deployed in a CATV headend environment.
Lesson Overview We will look at system setup, including the network

identification, management parameters and security.
Page 1
Module 6, Page 2
Objectives
Objectives Upon successful completion of this module you

ÔNavigate the CLI (Command Line Interface)
structure
ÔBegin the basic configuration needed to deploy
the BSR1000/64000, including assigning
¾Network Identity
¾Management Access via SNMP
¾System Security
ÔUse the context-sensitive help in the command
line to aid in configuration tasks
ÔSave configuration updates to non-volatile
memory
ÔReset the BSR1000/64000 configuration to factory
default
ÔMonitor configuration updates as reflected in
system operation
Page 2
Module 6, Page 3
Topic Order
Topic Order Overview Page 5

Console Connection Page 8
Command Line Interface Page 9
User-Based Access Page 11
Monitoring and Managing User Access Page 12
Command Line Interface Page 13
EXEC Hierarchy Page 15
Command Line Help Page 16
Advanced Editing Commands Page 17
Output Modifiers Page 18
Resetting to Factory Default
Configuration Page 19
Running-Configuration Page 20
Initial Configuration Page 23
Page 3
Module 6, Page 4
Topic Order (cont.)
Topic Order Remote Access Page 28

Saving Running Configuration Page 34
Troubleshooting Page 35
Page 4
Module 6, Page 5
Overview
ÔThe BSR1000/64000 acts a DOCSIS qualified Cable Modem

Termination System (CMTS)
It acts as a connection between r/f and wide area networks
The BSR1000/64000 was developed in accordance with the DOCSIS

specifications that define CMTS data interfaces, r/f interfaces and operations
support systems interfaces (OSSI).
The intended service will allow transparent bi-directional transfer of Internet
Protocol (IP) traffic, between the cable system headend and customer locations,
over a hybrid fiber/coax (HFC) cable television system. The transmission path
over the cable system is realized at the headend by a CMTS, and at each
customer location by a cable modem. At the headend (or hub), the interface to
the data-over-cable system is called the Cable Modem Termination System -
Network-Side Interface (CMTS-NSI) and is specified in [MCNS3]. At the customer
locations, the interface is called the cable-modem-to-customer-premise-
equipment interface (CMCI) and is specified in [MCNS4]. The intent is for the
MCNS operators to transparently transfer IP traffic between these interfaces,
including but not limited to datagrams, DHCP, ICMP, and IP Group addressing
(broadcast and multicast).
Page 5
Module 6, Page 6
Overview (cont.)
Ô BSR64000 can act as an aggregator for BSR1000s

Providing connectivity between hubs and the headend
The BSR 64000 from Motorola provides broadband carriers with a competitive
edge in defining, deploying, and managing broadband services. This carrier-class
switching and service provisioning solution allows cable operators to rapidly
introduce differentiated data, voice, and multimedia services for both corporate
and residential subscribers.
The BSR 64000 is fully based on Data Over Cable Service Interface Specification
(DOCSIS) 1.0, DOCSIS 1.1, and PacketCable 1.0 standards. It is a highly
integrated, carrier-class, multiservice IP switch/router with an integrated Cable
Modem Termination System (CMTS) that offers the highest density and lowest
price-per-port in the industry for Hybrid Fiber Coax (HFC) connectivity. The BSR
64000 can support up to 26 downstream transmitters and up to 104 upstream
receivers in a single, space-saving chassis.
The system includes flexible interfaces for SONET and Ethernet connectivity, and
it eliminates the need for discrete CMTS equipment, up converters, aggregation
switches, and routers – allowing for it’s use in aggregating of traffic from regional
hubs to a headend.
Page 6
Module 6, Page 7
Overview (cont.)
ÔThe Command Line Interface (CLI) is Cisco-compatible

Allows for ease-of-use
Interoperability with legacy infrastructure
Supports full scripting capability
» ASCII-formatted command files can be
uploaded, downloaded, and executed
Connection referred to as an EXEC session
We will examine the command line interface as a means of configuring the

BSR1000/64000. Because this interface is Cisco-like, it should prove to be
familiar to many in the networking industry.
Page 7
Module 6, Page 8
Console Connection
Ô The BSR1000/64000 uses Command Line

Interface (CLI) as a local user interface
Local connectivity is accomplished using a an RJ-45 to DB-9 adapter and a
terminal program on a p.c.
Remote connectivity to the BSR64000 is handled through telnet or SNMP
applications
This interface allows for configuration of the BSR64000
DTE connector
•Allows the BSR to monitor CD (carrier detect) for a disconnect via a straight-through
cable
–The BSR1000/64000 will disconnect the active session to prevent anyone else from gaining
access to the console port for security reasons
Over the course of this class, we will use a (crossed-over) db-9 adapter attached
to a (straight-through) cat-5 cable running from your p.c. to the console port of the
BSR1000/64000. We will be using HyperTerminal as the terminal emulation
program.
We will look at telnet access to the BSR1000/64000 as well as the use of an
SNMP application to access configuration and performance information about the
box.
The advantage to the local console access is that it is out-of-band. With SNMP
(or any remote management interface), there is some overhead associated
(network traffic).
In order to use HyperTerminal, ensure that your settings for console access are as
follows:
Bits per second- 9600
Data bits- 8
Parity- N
Stop bits- 1
Control Flow- none
Page 8
Module 6, Page 9
Command Line Interface
ÔTwo main EXEC modes

Both have different prompts
First is User Exec
» Allows Limited access to BSR
• No real configuration
• Command prompt is hostname>
RDN> ?
cable
clear
cmts
…
To enter commands in this user interface, you can either type or paste them within
the console command modes.
Each mode (level) has a distinct prompt for commands.
There is a hierarchy of commands in the command-mode structure. This
hierarchy is dependant on the EXEC access levels. There are two different
access levels. We will start by looking at the User EXEC mode.
User EXEC allows for access to all but configuration commands. The ENTER key
tells the BSR1000/64000 to read and execute any commands entered.
When a User EXEC session is started, the BSR64000 will display hostname> as
it’s prompt. The greater than symbol (>) indicates that you are in User EXEC
mode.
The list of commands available can be accessed by entering a question mark (?).
This is what is called context sensitive help.
Page 9
Module 6, Page 10
Command Line Interface (cont.)
ÔSecond main EXEC mode

Privileged (also called Enabled)
» Detailed look at BSR configuration
» Configuration and debugging
» Necessary in order to enter further configuration modes
» Entered by specifying enable command in User EXEC
• Command prompt is hostname#
RDN> en
RDN# ?
copy
debug
delete
…
Configuration and management commands require that you be in privileged EXEC

mode. To switch to privileged EXEC, type the enable command (or en for short).
If an enable EXEC password has been configured, you will be prompted for it
here.
When the correct password is supplied, the prompt will change to host#. Using
the context sensitive help (typing the question mark <?>) will reveal a number of
new commands that are available. The <tab> key can be used to complete long
commands provided enough characters are keyed in for it to be recognized.
To save any modifications made in the configuration mode the copy command
must be used. Any modifications made through the CLI are immediately active on
the system, because they are written to the running-config. In order to ensure
that those changes are permanent (will survive through next boot), you must copy
the running-config to the startup-config. The shortened version of this command it
copy run start.
Once again, anytime you copy running-config to startup-config, it is considered
best practice to copy a version of the latest configuration into a text file for
archiving.
To return to User EXEC, enter disable.
Page 10
Module 6, Page 11
User-Based Access
• Access based on username

Access-level dependant on user-privileges assigned
• Entered by specifying login <username> command in User EXEC
– Command prompt is hostname#
– Exited by specifying disable
RDN> login ispuser

Password:
RDN#
…
Page 11
Module 6, Page 12
Monitoring and Managing User

Access
Ô show users
¾ shows who is currently on the system
BSR64000# show users

Active connections
user group/privilege origin session
-----------------------------------------------------------------------------------------------------
enabled-user owner/rw console 0*
dao-ro isp/ro 198.112.190.57 1
dao-rw isp/rw 198.112.190.57 2
*unauthenticated* none/ro 198.112.190.201 3
…
Ô logout
¾ owner command to logout other users
BSR64000# logout session-id 2

…
As a user with owner rights, you are able to logout other users as follows:
BSR64000#logout session-id 2
This would be the first step to take upon discovering unauthorized access to the
BSR.
Page 12
Module 6, Page 13
ÔGlobal Configuration EXEC mode

Allows for global configuration
» Entered by specifying configure command in Privileged EXEC
mode
• Command prompt is hostname(config)#
RDN> en
RDN# config
RDN(config)#
…
Global Configuration commands apply to features that affect the entire system.
These commands apply to system features and enable routing functions.
To enter Global Configuration mode, enter the configure command from Privileged
EXEC mode. The prompt changes to hostname(config)#. To return to Privileged
EXEC mode, enter the end or exit command or press Ctrl-Z.
Page 13
Module 6, Page 14
ÔInterface Configuration EXEC mode

Allows for interface-level configuration
Takes precedence over global configurations
» Entered by specifying interface <interface-type> <interface
identifier number> command in Global Config EXEC mode
• Command prompt is hostname(config-if)#
• Includes POS, cable, Ethernet and loopback interfaces
RDN> en
RDN# config
RDN(config)# interface pos 5/0
RDN(config-if)#
…
RDN(config)# interface loopback 1

RDN(config-if)#
…
You may enable features on a per-interface basis. Interface Configuration

commands modify the operation of an interface such as an Ethernet port.
Interface Configuration commands always follow a Global Configuration
command, which defines the interface type. From Global Configuration mode,
enter Interface Configuration mode by entering any configure command, such as
the following:
hostname(config)#configure cable 3/0
The prompt changes to hostname(config-if)#, To exit Interface Configuration mode
and return to Global Configuration mode, enter the exit command. To exit
configuration mode and return to Privileged EXEC mode, use the end command
or the exit command or type Ctrl-Z.
Page 14
Module 6, Page 15
EXEC Hierarchy
User EXEC
hostname> enable
Privileged EXEC
logout | exit | ^z
hostname# config
Global Config EXEC

hostname(config)# hostname(config)# interface
router [rip | ospf | bgp] [pos | ethernet | cable |
loopback] <slot/port>
Router Config EXEC

hostname(config-rip | ospf | bgp)#
Interface Config EXEC
hostname(config-if)#
The exit command will bring you up one level. Control-z (^z) will bring you up to
Privileged EXEC from any lower configuration level. The disable command will
bring you from the Enable EXEC to User EXEC. The logout command will bring
you up one level.
Page 15
Module 6, Page 16
Command Line Help
Ô Context Sensitive Help

¾ Enter ?
» Provides a list of commands
» Completes a partial command
» When typed after a command, provides arguments for that command
RDN> ?
cable Cable commands
clear Reset functions
RDN> en?
enable
RDN>en_
RDN>no ?
cable Cable commands
debug Turn on debug information
Two types of context-sensitive help are available from the command line interface.
Entering ? Will return a list of commands available. Also, entering a character
sequence followed by the question mark (e.g., en?) will provide a list of
commands that begin with those characters.
Entering the question mark after a command (with a space in between) will
provide a command syntax. Specifically, when entered in place of an argument,
the BSR64000 will provide a list of available command options (with <cr>
specifying no options, or carriage return). The <tab> key can be used to complete
long commands provided enough characters are keyed in for it to be recognized.
Page 16
Module 6, Page 17
Advanced Editing Commands
Ô Advanced Editing Commands

<ctrl> + key
» Provides shortcuts to functions
<ctrl> + p
<ctrl> + b <ctrl> + f
<ctrl> + n
<control-p> will recall commands from the history buffer and <control-n> allows
you to cycle forward through previously entered commands. <control-f> moves
one character forward; <control-b> moves one character backwards. Other
combinations include:
<control-a> - beginning of line
<control-e> - end of line
<control-z> - exit to Privileged EXEC from any Config EXEC
Page 17
Module 6, Page 18
Output Modifiers
Ô Output Modifiers act as filter to show commands

| (vertical bar) is the start of your filter
Filter type
» begin – suppresses each output line until matching search string
» exclude – suppresses all lines that match search string
» include – display only those lines that include search string
Search string
» A word or double quote string that is used by the filter
BSR64000# show running-config | begin interface

interface ethernet 7/0
no shutdown
ip directed-broadcast
…
You may use output modifiers to filter the length of the output from most show
commands. Currently, you may only provide a single output modifier per show
command.
Page 18
Module 6, Page 19
Resetting to Factory Default

Configuration
ÔEnsuring a known state for initial configuration

delete nvram:startup-config
» Deletes startup-config from non-volatile memory
» Does not affect (active) running-config
• Must power-cycle BSR64000 to complete process
Since the startup-configuration is stored in NVRAM, it is possible to delete it from

the command line interface. The BSR64000 will be returned to a factory default
configuration when this is done. It is important to remember that the BSR64000
needs to be power-cycled before the running-configuration will be purged and the
factory defaults are loaded.
Page 19
Module 6, Page 20
Running-Configuration
Ô During operation, the BSR uses its running-config

Provides operational parameters
» Device configurations
Changes to the running-config take effect immediately.
Changes to the running-config will be lost unless saved to startup-config
for the BSR64000.
its operational state. Changes in the operational parameters are immediate and
may cause the modems connected to the BSR64000 to re-synchronize to the
downstream or restart the registration process (which can be quite noisy and
cause problems).
Page 20
Module 6, Page 21
Running-Configuration (cont.)
Ô Configuration file can be located in places other than NVRAM

Initial configuration is through the console terminal
Telnet session
SNMP application
Can by downloaded to the BSR64000 from a TFTP (Trivial File Transfer
Protocol) server on the network
Upon initial boot of the BSR64000, the running-configuration will be more or less
empty of operational parameters (other than some defaults). It will be
configurable via the console port.
Once TCP/IP and security have been configured on the BSR64000, it can be
configured remotely either through a telnet session, SNMP application or via a
plain text file available on a TFTP server.
Page 21
Module 6, Page 22
Running-Configuration (cont.)
Ô Available from Privilege EXEC

show running-config
» Can be shortened to sh run
show startup-config
» Can be shortened to sh start
RDN# sh run
!
! Last configuration change at THU MAY 25 07:19:03 2000
!
hostname RDN
!
!
no service password-encryption
!
…
All commands can be shortened to their smallest recognizable versions. For

example, show running-config could be further shortened to sh run and sh
startup-config to sh st. The output from either of these commands can be
captured in a text file (either through the utilities in the terminal emulation program
or by copying them from the screen) and archived. With a brand new system, we
should see only some default settings in the startup configuration. It is always a
good idea to check your configuration after making any changes to ensure their
accuracy.
It is considered a best practice to keep an archive of your last two or three
configurations for disaster recovery, since any of these can be loaded onto a
factory-defaulted system.
Page 22
Module 6, Page 23
Ô Basic configuration tasks:

Set network identity
» Hostname
» Interface I.P. addresses
System security
» Define system login
» Restrict access to Privileged EXEC
» Enabling automatic password encryption
System management
Setting SNMP parameters
» Configuring loopback I.P. address
Save running configuration
Basic configuration tasks include specifying a hostname, assigning I.P.

information and establishing security parameters. This allows for remote access
for further configuration through a telnet session or SNMP-based application.
Remember that any modifications made through the CLI are immediately active
on the system, because they are written to the running-config. In order to
ensure that those changes are permanent (will survive through next boot), you
must copy the running-config to the startup-config. The shortened version of this
command it copy run start.
Page 23
Module 6, Page 24
Initial Configuration (cont.)
Ô Set network identity

Assign hostname
From the Global Configuration EXEC mode
» hostname <new-hostname>
• Should be unique
RDN> en
RDN# config
RDN(config)# hostname newhostname
newhostname(config)#
The hostname command can be used at any time to alter the BSR’s name.
Because you may have to access more than one BSR on the system remotely, it
is important to ensure that it’s hostname unique across your network. The prompt
will provide remote-access users an easy way to determine which device they are
accessing.
After you complete this command, the CLI prompt changes to the new hostname.
Page 24
Module 6, Page 25
Ô Set network identity

Assign interface I.P. address
From the Interface Configuration EXEC mode
» ip address <ip-address> <subnet-mask>
• Unique to each interface
newhostname> en
newhostname # config
newhostname(config)# int ethernet 5/0
newhostname(config-if)# ip address 192.168.100.69 255.255.255.0
newhostname(config-if)# exit
Assigning IP information involves assigning an IP address and subnet mask to the

BSR64000. From Privileged EXEC mode, enter Global Configuration mode (with
the configure command). Enter Interface Configuration mode, and enter the
interface to which you want to assign IP information (with the interface <interface
type> <interface number> command). Enter the IP address and subnet mask
(with the ip address <ip-address> <subnet-mask> command).
When finished, return to Global Configuration mode (using the exit command).
Return to Privileged EXEC mode (using the end command). This allows you to
ensure that the information was entered correctly by displaying the running
configuration (using the show running-config command).
Page 25
Module 6, Page 26
Ô Configure security
Set Privileged EXEC password
» enable password <password>
• By default unencrypted
• Can be read as plain text in configuration files
newhostname> en
newhostname# enable password newpassword
newhostname#
There is, by default, no password protecting access to the privileged EXEC mode.
It is important to assign one to restrict configuration access to the BSR64000. It is
important to remember that the password is unencrypted in the running and start-
up configurations. Anyone with access to online or offline versions of those files
could check for the password.
Page 26
Module 6, Page 27
Set remote access password
» [no] password <password>
• By default unencrypted
• Can be read as plain text in configuration files
• Must be configured in order for remote (non-console) access
– Enable password must also be set
• Starts telnetd (telnet daemon)
Stopped when password removed
Adding or removing the telnet password has no effect on existing
connections
newhostname> en
newhostname# password newpassword
newhostname#
Nobody will be able to remotely access the BSR’s configuration if

this password is not set. Further, in order to be able to telnet to the
BSR, an I.P. address must be set up for at least one available
interface.
Page 27
Module 6, Page 28
Remote Access
Ô session-timeout configures how long a session can remain idle

¾ console <0-30> – configures that timeout for the console (default 0)
¾ telnet <0-30> – configures that timeout for each telnet session (default 5)
Ô access-class - restricts telnet access based on a standard access-list
BSR64000(config)# access-class 10 in
BSR64000(config)# ^z
BSR64000#show running-config | include access-list
access-list 10 permit 10.10.10.0 0.0.0.255
access-list 10 deny any
BSR64000#
Page 28
Module 6, Page 29
Enabling automatic password encryption
» service-password encryption
• Disabled by default
• Encrypts all existing passwords
• Automatically encrypts all new passwords
newhostname> en
newhostname# service-password encryption
newhostname#
This adds a level of security. Even someone with physical access to the
BSR64000’s configuration files would not be able to ascertain it’s passwords.
Removing service-password encryption will leave previously encrypted passwords
as they are. To have unencrypted passwords in your configuration, you would
have to remove the encrypted password with the no password <password>
command.
Page 29
Module 6, Page 30
Ô Configure for system management

Assign loopback I.P. address (only on the BSR64000)
Up to 16 loopback addresses
From the Loopback Interface Configuration EXEC mode
» ip address <ip-address> <subnet-mask>
newhostname> en
newhostname(config)# int loopback 1
newhostname(config-if)# ip address 200.24.3.1 255.255.255.0
newhostname(config-if)# exit
Assigning loopback IP information is important to system management via any

SNMP-based application. It is dangerous to allow an NMS (Network Management
System) to discover the BSR64000 using an interface address. Anytime that
interface is down, the NMS would not be able to poll for data. Anytime the i.p.
address were to change for a discovered interface, a great deal of work would
have to be done on the NMS to map historical information to the newly gathered
information.
From Privileged EXEC mode, enter Global Configuration mode (with the
configure command). Enter Interface Configuration mode, and enter the
interface to which you want to assign IP information (with the interface loopback
<interface number between 1-16> command). Enter the IP address and subnet
mask (with the ip address <ip-address> <subnet-mask> command).
Page 30
Module 6, Page 31

Creating a view
Allows restriction of access to specific mibs
» snmp-server view <view name> <access level>
newhostname> en
newhostname(config)# snmp-server view classroom 1.3.6 included
newhostname(config)# exit
newhostname#
Creating a view allows for restriction of the mibs available to an NMS (user). This
is done by associating community strings with view names, as we will see in the
next slide.
Page 31
Module 6, Page 32

Assigning community strings
Associating community strings with views
» snmp-server community <community string> <read [ro] or
read/write [rw] view <previously configured view name>
newhostname> en
newhostname(config)# snmp-server community river ro view classroom
newhostname(config)# snmp-server community delta rw view classroom
newhostname#
Read-only community strings allow for access to mib data from an NMS with the
restrictions imposed by the associated view. Read-write community strings allow
for configuration changes via an SNMP-based application. Multiple views and
community strings may be configured and associated allowing a division of
configuration and informational access.
Remember that community strings are case sensitive.
Page 32
Module 6, Page 33

Enabling traps
• snmp-server enable traps
Specifying NMS to receive traps
• snmp-server host <i.p. address of NMS>
newhostname> en
newhostname(config)# snmp-server enable traps
newhostname(config)# snmp-server host 192.168.4.9
newhostname#
Traps are defined in the BSR’s SNMP agent to be sent based on pre-defined
events, such as interface going down or reboots. An SNMP-based application
must be configured to receive and process these trapped events. Instructions for
this sort of configuration should be available in your NMS vendor’s documentation
as it varies from vendor to vendor.
Page 33
Module 6, Page 34
Saving Running Configuration
Ô Saving Running Configuration

CLI-based changes are to the running-configuration
» Would be lost on next reboot
copy running-config startup-config
» Can be shortened to copy run start
newhostname> en
newhostname# copy run start
newhostname#
To save any modifications made in the configuration mode the copy command
must be used. Any modifications made through the CLI are immediately active on
the system, because they are written to the running-config. In order to ensure
that those changes are permanent (will survive through next boot), you must copy
the running-config to the startup-config. The shortened version of this command it
copy run start.
Once again, anytime you copy running-config to startup-config, it is considered
best practice to copy a version of the latest configuration into a text file for
archiving.
Page 34
Module 6, Page 35
Troubleshooting
Ô Version information
show version
» Can be shortened to sh ver
RDN> sh ver
…
Once again, the command could be reduced to the shortest recognizable version,
or sh v. This command displays the configuration of the system hardware, etc.
We will return to this command when we address troubleshooting, as it will
provide much of the information that Motorola Networks Global Technical Support
will look for when you call. Version information can provide clues to known bugs.
It is considered a best practice to keep a copy of the output from this command for
troubleshooting purposes.
Page 35
Module 6, Page 36
Troubleshooting (cont.)
Ô Interface configuration and statistics

show interfaces
» Can be shortened to sh int
Can isolate specific interfaces
» Specify interface as argument
• sh int eth 0/0
» Must specify port number (i.e., 0/0 is slot 0, port 0)
RDN> sh int eth 0/0

Ethernet 0/0 is up, line protocol is up
Hardware address is 0:29:b4:69:4:df
Internet address is 10.1.3.1/24
Half-duplex, 100 mb/s, 100BaseTX/FX
ARP Type: Arpa, ARP Timeout 01:00:00
Last input never, output 00:34:10
…
This command will show configuration information (i.e., MAC address, i.p.
address, interface speed, etc.), as well as functional status (i.e., packet
information, line status, etc.). We will go into greater detail about this command
when we explore troubleshooting interfaces.
Page 36
Module 6, Page 37
Troubleshooting (cont.)
Ô Technical Support-Related Information

show tech
RDN# sh tech
…
This will output a long stream of information including memory

usage, semaphores, etc. that Motorola Global Technical Support
may ask for when working an issue with you. It is recommended
that you capture this output into a text file.
Page 37
Module 7, Page 0
Module 7
CMTS Configuration
Rev. 1.0 11/01
Page 0
Module 7, Page 1
Introduction
Introduction The BSR, by design, delivers transparent IP

traffic from the CM-CI (CPE) through the
CMTS-NSI. This is accomplished through
configuration of the CMTS r/f interfaces, as well
as the network side interfaces.
Importance In order to provide IP traffic from CPE to

CMTS-NSI, we need to configure the CMTS on
the BSR.
Lesson Overview We will discuss configuring the CMTS

interfaces on the BSR. We will examine all of
the commonly altered variably configured
parameters.
Page 1
Module 7, Page 2
Objectives

ÔConfigure the CMTS interfaces on the BSR
ÔExplain the impact of varying configurable
parameters, such as interleave-depth and channel-
width
Page 2
Module 7, Page 3
Topic Order
Topic Order One-Way Services Page 5

Proprietary Systems Page 6
DOCSIS-Compliant Systems Page 7
Typical Cable Frequency Plan Page 8
BSR Overview Page 9
Integrated CMTS/Router Design Page 10
Reference Architecture Page 14
RFI Technical Details Page 15
DOCSIS-OSI Mapping Page 18
DOCSIS Protocol Signaling Page 19
Downstream Page 20
Page 3
Module 7, Page 4
Topic Order (cont.)
Topic Order Downstream Modulation Rates Page 22

Forward Error Correction (FEC) Page 23
Interleaving Page 24
Configuring Downstream Parameters Page 25
Upstream Page 28
Upstream Signal Profile Page 29
Upstream Channel Descriptors Page 30
Maps Page 31
Data Transfer Page 32
Upstream Power vs. Location Page 33
Configuring Upstream Parameters Page 34
Upstream Modulation Rates Page 39
Performance Page 40
Page 4
Module 7, Page 5
One-Way Services
Phone
Internet Network
Customer Home
Analog
and/or
Digital
A/V
Broadcast
(i.e. TV) Cable
Plant
Cable Company (Coax)

Early systems were point-to-multipoint, feeding signals from a headend site out to
the subscribers. Amplifier cascades could run from 25-45 actives deep on a trunk
run of the Tree and Branch configuration. Attempts at returning signals from the
subscriber to the headend (multipoint-to-point) failed. This was because of the
large amount of amplifiers funneling noise and distortions back to the headend.
Poorly shielded cable used in the house drop system added interfering signals
into the mix.
possible. As early as 1957 experimentation with Pay Per View services on cable
systems took place In Bartlesville, OK.
Today systems use a hybrid of fiber and coax (HFC) with short cascades of 1 to 4
amplifiers and minimal power supplies. This paved the way for offering digital
Page 5
Module 7, Page 6
Proprietary Systems
Phone
Internet Network
Customer Home
Analog
and/or
Digital
A/V
Broadcast Proprietary
(i.e. TV) Cable Cable Modem
Proprietary Plant
Cable Modem
Head End Combiner
Cable Company (Coax)
To provide consumers with higher speed internet access, some manufacturers

introduced proprietary Cable Broadband solutions, which consisted of packaged
combinations of their Cable Modems with companion headend equipment. These
proprietary cable modem solutions provided no interoperability among different
manufacturers’ Cable Modems and headend components. Many cable systems
weren’t yet capable of upstream communications. For these “one-way” cable
systems, hybrid proprietary Cable Modem solutions were created. These used
CATV coax cable for downstream (from the headend to the customer)
communication, but Telephony Returns upstream (from the customer to the
headend) communication, to provide internet access services to homes and
businesses.
Institutional networks were built by Multiple System Operators (MSOs) at the
request of the cities and towns to connect public buildings together (typically this
was required as a part of the MSO’s franchise agreement). These systems were
separate from the subscriber systems and were originally used for two-way video.
The early cable modems ran at 300 to 9600 baud.
In 1994, 5 MSOs participated in a CableLabs sponsored two-way plant
characterization test. This test proved it was possible to transport data across
HFC subscriber plant relatively error free. When Continental Cablevision
launched High Speed Data services in 1996 using LanCity modems the cost per
modem was $650. Today the costs are significantly less. A typical DOCSIS
modem retails at around $120.
Page 6
Module 7, Page 7
DOCSIS-Compliant Systems
Provisioning Phone
Internet Services Network
Customer Premises
Gateways
Servers Analog
and/or
Digital
A/V
Broadcast DOCSIS-Compliant
(i.e. TV) HFC Cable Modem
Plant
Combiner
Cable Company
The first two Cable Modem models built to the first round of DOCSIS
specifications (DOCSIS 1.0) became CableLabs Certified in March of 1999. The
first CMTS (DOCSIS-compliant Cable Modem Termination System) became
CableLabs Qualified soon thereafter.
The cable operator is now positioned to provide standardized cable modem
internet access nationwide, using CableLabs Certified DOCSIS CMs and
CableLabs Certified DOCSIS CMTSs.
In a CableLabs DOCSIS compliant cable modem system, any vendors’ CMTS
should be interoperable with any vendors’ CMs.
Page 7
Module 7, Page 8
Typical Cable Frequency Plan

“UPSTREAM” “DOWNSTREAM”
5 - 42 MHz 54 - 750/860 MHz CM Downstream Channels

6 MHz spacing
CM Upstream Channels
~6 MHz BW
5-42 MHz range
88-860 MHz range (300 - 750 typ.)
QPSK or 16 QAM
64 or 256 QAM
Analog A/V Channels Digital A/V Channels

6 MHz spacing 6 MHz spacing
~6 MHz BW ~6 MHz BW
54-750/860 MHz range
64 or 256 QAM
In the upstream direction, one or more channels of upstream bandwidth can be

allocated anywhere in the 5 - 42 MHz range that does not interfere with other US
signals that may be carried on the cable system, to transport DOCSIS Cable
Modem bursts to the CMTS. An upstream channel’s occupied bandwidth
depends on how fast data is being sent on that channel (the “symbol rate.”) This
data rate parameter is one of the many upstream channel parameters that are
configured into the CMTS, then continuously “advertised” by the CMTS to its
Cable Modems using the CMTS broadcast downstream UCDs.
In the downstream direction, a DOCSIS channel is 6 MHz wide. It can be placed
in any 6 MHz slot in the downstream spectrum (88 to 860 MHz), which is in the
cable systems downstream bandpass and does not interfere with other
downstream cable system signals (in place of a standard 6 MHz bandwidth analog
or digital TV signal that might otherwise be placed there). One common
downstream spectrum allocation places analog TV channels at the lowest
downstream frequencies, then digital TV channels, then one or more DOCSIS
channels at the highest downstream frequencies on the system.
Page 8
Module 7, Page 9
BSR Overview
ÔThe BSR64000 and BSR1000 are a DOCSIS compliant Cable

Modem Termination System (CMTS)
¾ Developed against DOCSIS standards
¾ It acts as a connection between r/f and wide area networks
ÔPrincipal Function of the DOCSIS Cable Modem System Is to
Transmit Internet Protocol (IP) Packets Transparently Between
the Headend and the Subscriber Location
Cable
Wide-Area Modem
Network Termination
System Cable
CMTS
(CMTS) Network Cable Modem CM Customer Premises
Network Side
Interface HFC (CM) Equipment Interface
Customer
Premises
Equipment
Transparent IP Traffic Through the System
The BSR64000 and BSR1000 were developed in accordance with the DOCSIS
specifications that define CMTS data interfaces, r/f interfaces and operations
support systems interfaces (OSSI).
The intended service will allow transparent bi-directional transfer of Internet
Protocol (IP) traffic, between the cable system headend and customer locations,
over a hybrid fiber/coax (HFC) cable television system. The transmission path
over the cable system is realized at the headend by a CMTS, and at each
customer location by a cable modem. At the headend (or hub), the interface to
the data-over-cable system is called the Cable Modem Termination System -
Network-Side Interface (CMTS-NSI) and is specified in [MCNS3]. At the customer
locations, the interface is called the cable-modem-to-customer-premise-
equipment interface (CMCI) and is specified in [MCNS4]. The intent is for the
MCNS operators to transparently transfer IP traffic between these interfaces,
including but not limited to datagrams, DHCP, ICMP, and IP Group addressing
(broadcast and multicast).
Page 9
Module 7, Page 10
Integrated CMTS/Router Design
Internet
IP Internet
IP Internet
Service Provider
Backbone Service Provider
Service Provider
Backbone DOCSIS HFC Networks
DOCSIS
HFC Networks
Edge Router Aggregation
(GSR12000) Switch Integrated Router/CMTS
(C5000)
CMTS’s
(UBR7246)
ÔOld Provisioning Model ÔNew Provisioning Model

¾ Multiple platforms ¾ Single integrated platform
¾ Complex service provisioning and ¾ Simple/intuitive service
management provisioning and management
¾ Segmented QoS ¾ End-to-end QoS guarantees
¾ Lower reliability, higher cost ¾ Higher reliability, lower cost
The BSR64000 provides for routing and IF to RF up and down conversion in a

single box. This means a move from a multiple box (multiple points of failure) to a
single box with redundancy capabilities. Not only that, but, the addition of carrier-
class routing within the same box provides for less hops between devices,
therefore less latency for data traffic.
Page 10
Module 7, Page 11
ÔBasic configuration tasks:

¾ Configure RF Interfaces
» IP Helper Address
» Cable Helper Address
• Cable Modem
• CPE
» Downstream Parameters
• Interleave Depth
• Modulation
• Unshut Interface
» Upstream Parameters
• Channel-Width
• Frequency
• Power-Level
• Modulation Profile
• Unshut Interface
¾ Save running configuration
Basic configuration tasks in include configuring your helper addresses and setting
up your up and downstream parameters, including your modulation-type,
interleave depth and channel-width, all of which will affect your throughput.
Page 11
Module 7, Page 12
Introduction to DOCSIS
Ô Data Over Cable Service Interface Specifications

¾ System-level specification against reference architecture
» DOCSIS 1.0
• Baseline Privacy Interface Specification (BPI)
• Cable Modem Telephony Return Interface Specification
• Acceptance Test Plan 1.0
» DOCSIS 1.1
• Baseline Privacy Plus Interface Specification (BPI+)
• DOCSIS 1.1 CMCI Acceptance Test Plan
• DOCSIS 1.1 RFI Acceptance Test Plan
Standards effort in IEEE 802.14 were stalled and Major MSOs wanted to lower
their costs through retail cable modems and interoperability of multiple vendors’
equipment. DOCSIS 1.0 is effectively complete. Product is available and being
installed in hundreds of CATV systems. DOCSIS 1.1 cable modems are available
and CMTS are currently being tested. This will add technology that makes
possible Quality of Service (QoS) guarantees needed for newer services like
DOCSIS-based telephony over cable. Another CableLabs effort -- PacketCable --
is merging its efforts with DOCSIS to realize this.
DOCSIS is really a family of documents. These are available from the public area
of the CableLabs cable modem website at
http://www.cablemodem.com/specifications.html. DOCSIS defines a
Reference Architecture for an overall data communication system optimized to
connect consumer PCs to the internet using the DOCSIS-compliant Cable
Modems and DOCSIS-compliant Cable Modem Termination Systems. These
specifications really started as just RFI and grew into family of specifications that
have been accepted by the MCNS and IEEE.
In Summary, DOCSIS Defines:
•Interfaces
•Data (CMCI, CMTS-NSI)
•Radio frequency (RFI)
•Telephony return (CMTRI)
•Security (BPI, BPI+)
•Management & Business Support (OSSI)
•System Operation
Page 12
Module 7, Page 13
Reference Architecture
DISTRIBUTION HUB OR HEADEND

DS TV signals
LOCAL
SERVER
Cable FACILITY OTx
COMBINER
Modem
Termination CMTS Cable Modem
System Termination FIBER
Network Side CABLE DS = >54 MHz
MOD System
TERMINATION PLANT
NETWORK
Interface Downstream
(CMTS-NSI) RF Interface F.O.
LAN
NODE
DEMOD Cable Modem
SPLITTER
Internet Backbone
to RF Interface CM CPE
ORx
OSS COAX
Interface Baseline Privacy CABLE
Network
Cable
(OSSI) Interface (BPI) Modem PLANT
OPERATIONS SECURITY & Termination
SUPPORT ACCESS US signals System
US = <42 MHz
SERVER CONTROLLER Upstream
RF Interface
Cable Modem to
REMOTE SERVER FACILITY CPE Interface
(CMCI)
The CPE - Customer Premise Equipment is (today) primarily the consumer’s PC.
CPE connects to the Cable Modem via an Ethernet or USB port. The Cable
Modem connects to the 2-way cable plant coax via a standard CATV drop and
jack.
A two-way coax cable transports data to/from the HFC optical node that feeds the
home, using traditional 2-way coax cable technology. This technology transports
signals higher than 54 MHz in the downstream direction, and signals lower than
42 MHz in the upstream direction, by RFI specification.
data upstream, hence the use of optical splitters and combiners. At the headend,
the CMTS downstream output is combined with other downstream cable signals
(such as analog and digital video services), while the CMTS upstream input is
picked from the upstream cable spectrum.
On the CMTS network side interface, a LAN in the headend connects the CMTS
to the Internet and a number of support servers required by DOCSIS specification
(i.e., DHCP,ToD and TFPT), plus (typically) data caching servers and provisioning
systems. The support servers are often physically located at another Regional
Data Center, that may be tens or hundreds of miles from the headend. Caching
servers are co-located with the CMTS at the cable headend.
Page 13
Module 7, Page 14
DOCSIS Yesterday
Ô DOCSIS 1.0
¾ Product Interoperability across available CMTS’s
¾ 64 and 256 QAM modulation (downstream) formats
¾ 6-MHz occupied spectrum coexists with all other signals on the cable plant
¾ Variable-depth interleaver supports both latency-sensitive and -insensitive data.
¾ The features in the upstream direction are as follows:
» Flexible and programmable CM under control of the CMTS
» Frequency agility
» Time division multiple access
» QPSK and 16 QAM modulation formats
» Support of both fixed-frame and variable-length PDU formats
» Multiple symbol rates
» Programmable Reed-Solomon block coding
» Programmable preambles
Page 14
Module 7, Page 15
DOCSIS Today
Ô DOCSIS 1.1
¾ Packet Classification, based on fields in the Ethernet, IP, and UDP/TCP headers,
into a Service Flow
¾ Service Flow association with a DOCSIS Service Identifier
¾ QoS MIBs
¾ Fragmentation
» Enhanced MAC scheduling / Unsolicited grants
» Real-time polling / Non-real time polling
¾ Payload Header Suppression (for increased bandwidth efficiency, particularly in
the case of relatively small Voice-over-IP [VoIP] packets)
¾ Priority Queuing (e.g. Weighted Fair Queuing) at the CMTS
¾ BPI+ (Base Line Privacy - Plus)
» Adds digital-certificate based CM authentication to the key management
protocol
¾ IGMP (Internet Group Management Protocol) Management
Page 15
Module 7, Page 16
DOCSIS Tomorrow
Ô DOCSIS 2.0
¾ RFI Issued (step before closed)
¾ Increased upstream channel capacity
» Symmetrical services
¾ Better noise immunity (increased robustness)
¾ Backward compatibility with DOCSIS 1.0 and 1.1
» Coexistence in the same u/s and d/s with deployed modems
Ô Potentially more robust
¾ Greater bandwidth usage
Page 16
Module 7, Page 17
DOCSIS Summary
Version Highlights Benefits/Services

DOCSIS 1.0 Ô Widespread Certification Ô High speed data
(5 Mbps u/s) Ô Standard Specification Ô Internet access
DOCSIS 1.1 Ô QoS Ô Tiered service

(10 Mbps u/s) ÔGreater Bandwidth Ô Double u/s capacity
Efficiency Mechanisms Ô Lower operational costs
Ô Pre-Equalization
Ô Enhanced Operational
Systems Support Interfaces
ÔEnhanced Security
DOCSIS 2.0 “Adv PHY” Ô S-CDMA proposal Ô Symmetric Services
(30 Mbps u/s) Ô Advanced TDMA proposal ÔGreater Upstream Capacity
Page 17
Module 7, Page 18
RFI Technical Details
ÔRF Interface Specifications

¾ Network layer
» IP only
¾ Data Link Layer
» LLC sublayer conforms to Ethernet
» Link security sublayer
» MAC sublayer supports varaible length PDUs
¾ PHY
Page 18
Module 7, Page 19
RFI Technical Details (cont.)
ÔPHY Features [Downstream]

¾ Based on North American Video Transmission Specs
» J83 Annex B
¾ 64/256 QAM
¾ Combination of MPEG- Framing, Reed-Solomon FEC & Trellis
Encoding
¾ Variable depth interleaving
Page 19
Module 7, Page 20
RFI Technical Details (cont.)
ÔPHY Features [Upstream]

¾ QPSK & 16QAM
¾ Carrier width dependant on baud rate
¾ Multiple symbol rates (160KBaud to 2.56MBaud)
¾ TDMA
¾ Programmable Reed-Solomon block coding
¾ Programmable preambles
¾ Support of variable & fixed frame PDU formats
The upstream Physical Media Dependent (PMD) sublayer uses a FDMA/TDMA

burst modulation format, which provides five symbol rates and two modulation
formats (QPSK and 16QAM). The modulation format includes pulse shaping for
spectral efficiency, is carrier-frequency agile, and has selectable output power
level. The PMD sublayer format includes a variable-length modulated burst with
precise timing beginning at boundaries spaced at integer multiples of 6.25 msec
apart (which is 16 symbols at the highest data rate). Each burst supports a flexible
modulation, symbol rate, preamble, randomization of the payload, and
programmable FEC encoding. All of the upstream transmission parameters
associated with burst transmission outputs from the CM are configurable by the
CMTS via MAC messaging.
The PMD sublayer can support a near-continuous mode of transmission, wherein
ramp-down of one burst MAY overlap the ramp-up of the following burst, so that
the transmitted envelope is never zero. The system timing of the TDMA
transmissions from the various CMs MUST provide that the center of the last
symbol of one burst and the center of the first symbol of the preamble of an
immediately following burst are separated by at least the duration of five symbols.
The guard time MUST be greater than or equal to the duration of five symbols
plus the maximum timing error.
The upstream modulator is the part of the cable modem which interfaces with the
cable network. The modulator contains the actual electrical-level modulation
function and the digital signal-processing function; the latter provides the FEC,
preamble prepend, symbol mapping, and other processing steps. The DOCSIS
specification was written with the idea of buffering the bursts in the signal
processing portion, and with the signal processing portion (1) accepting the
information stream a burst at a time, (2) processing this stream into a complete
burst of symbols for the modulator, and (3) feeding the properly-timed bursted
symbol stream to a memory less modulator at the exact burst transmit time.
Page 20
Module 7, Page 21
DOCSIS Protocol Signaling
Ô Frames and Timing

¾ Upstream Time Sharing (TDMA)
¾ Time allocation MAP from head end (every 4 ms)
¾ Upstream time allocated for Cable Modems in mini slots
(Mini-slot = 8 ticks, Tick = 6.25 usec)
¾ Shared time slots for Maintenance & Requests (e.g. for new modems
with no SID to come online)
¾ MPEG Frames
188 Bytes, 4 Byte header
¾ Synchronous Transmission
Clock Synch messages from head end (613 per second)
One source per downstream
Multiple sources per upstream requiring time sharing
Cable Modems identified by 16 bit Service ID (SID)
The upstream utilizes TDMA (time division multiple access) which means all of
the opportunities for a cable modem to transmit are allocated by time.
A MAP (Media Access Protocol) is sent every 4 mille seconds providing
information for the modems.
Cable modems are directed to use mini slots for upstream transmissions, these
mini slots sent from the CMTS.
Page 21
Module 7, Page 22
DOCSIS-OSI Mapping
OSI Stack Data-Over-Cable Interface Specifications
Cable Modem Termination System Cable Modem (Ethernet CMCI)

Security Security
DHCP TFTP SNMP Mgmt DHCP TFTP SNMP Mgmt
Higher Layers Applications DOCSIS Applications Applications
Control
Transport TCP/UDP Messages TCP/UDP TCP/UDP
IP
Network IP IP
Forwarding 802.2/DIX LLC 802.2/DIX LLC 802.1d

Bridging 802.2/DIX
Data LLC
Data Link Link Security Link Security
Link Layer
802.3/DIX
DOCSIS MAC DOCSIS MAC MAC
U/S TDMA (minislots) U/S TDMA (minislots)

Physical PHY Layer D/S TDM (MPEG) D/S TDM (MPEG) 802.3
10 Base-T
Cable PMD Cable PMD
CMTS-NSI Cable Transmission Network CMCI

Interface to/from WAN Interface
Management only to/from Customer
Data Premises Equipment
Data forwarding through the CMTS may be transparent bridging or may use
network layer forwarding (routing, IP switching). Data forwarding through the
cable modem is link layer transparent bridging. Forwarding of IP traffic must be
supported. DOCSIS was designed against as a CSMA shared medium with the
RFI defining layers one and two only. Layer three is handled by standard IP
technologies by definition.
Page 22
Module 7, Page 23
DOCSIS/PacketCable TM
Architectures
Application Services
IP-Based Voice/Video Interactive
Telephony
Service TV
Conferencing
PACKETCABLE
PacketCable Protocols
Internet Protocol
DOCSIS
Media Access Control
Physical Media Dependant (PMD) Layer
Page 23
Module 7, Page 24
Downstream
• Up to 38 Mbps
+50 to +61 dBmV
• 88 - 860 MHz
• 6 MHz BW
COMBINER
CMTS
• 64 or 256 QAM OTx DS
• TDM, ADM
F.O.
• MPEG-2 format NODE
• Received by all CMs CM

on node CPE
COAX
• Continuous stream of CABLE
PLANT
data, MAPs, UCDs, & FIBER
Sync CABLE
PLANT -15 to +15 dBmV
• Should be “10dB
down”
In the downstream direction, a DOCSIS data carrier may be placed anywhere 88 -

860 MHz and will have a fixed 6 MHz bandwidth of always present and
measurable. One Cable Modem could have more than one downstream signal
appear on its input, but only one downstream is that Cable Modems home
frequency at any one time. Typically, this one downstream is used to serve all the
Cable Modems attached to a sizable quantity of optical nodes. At any instant in
time, the data modulated onto the CMTS downstream output is destined for only
one Cable Modem, one CPE, or one predefined group of Cable Modems or CPEs.
That is, downstream data is Time Division Multiplexed (TDM) on the downstream
data carrier. DOCSIS can deliver a DS raw data rate as high as 38 Mbps, though
no single Cable Modem could forward all that data to a PC. Data is modulated
onto a downstream DOCSIS carrier using 64 or 256 QAM. It is also framed in
MPEG-2 format, so it is possible for a cable operator to TDM digital TV and
DOCSIS Cable Modem data on the same DS. In actual practice, the CMTS
outputs a 41 - 47 MHz IF to it’s internal upconverter which upconvertes the signal
to the downstream DOCSIS frequency. In addition to internet data destined for
Cable Modems or CPEs, the CMTS must also generate and broadcast
downstream to all its Cable Modems a continuous stream of support data packets
that control Cable Modem upstream transmission: SYNCs, UCDs and MAPs.
DOCSIS recommends downstream CMTS signals be 6 - 10 dB below, or “10 dB
down from”, the level of adjacent analog TV carrier.
Page 24
Module 7, Page 25
Downstream (cont.)
Ô The DOCSIS Specification Uses a Modulation and Coding

Scheme Defined by ITU J.83 Annex-b, for the Downstream:
¾ Modulation Type: 64-QAM or 256-QAM
¾ Maximum Data Rate: 27 Mbps at 64-QAM, 38 Mbps at 256-QAM
¾ Symbol Rates: 5.057 Msym/s at 64-QAM, 5.361 Msym/s at 256-QAM
¾ Bandwidth: 6 MHz channel
¾ Frequency Range: 88 - 860 MHz (30 kHz min. step size)
¾ Transport Protocol: MPEG-2
¾ Forward Error Correction (FEC) encoding: outer Reed-Solomon and
inner Trellis code
¾ 1E-8 BER
¾ Carrier to noise ratio:
• 23.5 dB for 64-QAM
• 30 dB for 256-QAM
The downstream can use either 64 or 256 QAM modulation which will produce a
rate of 27 to 38 Mbps providing the CNR is between 23.5 to 30 dB.
Page 25
Module 7, Page 26
Downstream Modulation Rates
Ô64 QAM
¾ 6 bits per symbol
» 6 bits * 5.056941Msym/sec = 30.3416 Mb/s rate
• -10% with FEC on
» BER 10-8@ 23.5C/N
Ô256 QAM
¾ 8 bits per symbol
» 8 bits * 5.36Msym/sec = 43Mb/s rate
• -10% with FEC on
» BER10-8 @ 30dB C/N
Page 26
Module 7, Page 27
Forward Error Correction (FEC)
ÔReed Solomon
ÔEnables receiver to detect and fix errors
¾ Reduces need for retransmits
» Retransmits add to latency
• Round trip of NACK and retransmit of bit stream
ÔSegmenting of data stream into fixed-size blocks
¾ Addition of redundant bits
» Called Hamming codes
» Calculation done at receiver to detect errored bits
• Fixed when possible
» Works well when bits spread
• If too many errors close together, becomes difficult to detect errored bits
Page 27
Module 7, Page 28
Interleaving
ÔInterleaving
ÔReordering of bits
¾ Errors due to impulse noise spread over time
Burst Error
Transmission
1 2 3 4 5 6 7 8 9
Order
Interleave
1 2 3 4 5 6 7 8 9
Buffer
Apply FEC to
reorder stream 1 7 5 3 6 2 9 8 4
In the interleaving process, the bits are placed in an interleave buffer and
reordered. A burst error that would normally affect a group of adjacent bits will
now be affecting an out-of-order group. The erred bits are effectively spread out.
Forward error correction is applied in the interleave buffer.
By concatenating interleaving with Hamming code error correction, erred bits can
be reduced by factors of up to 104.
The cost here is in latency. The entire bit stream will experience a delay as the
bits are put in and taken out of the interleave buffer. This is called serialization
delay. Also, Reed Solomon calculations must be performed. The amount of
latency increases with the interleave depth, as does the robustness of the data
stream.
Page 28
Module 7, Page 29
Configuring Downstream
Parameters
ÔConfigure Downstream Parameters

¾ Interleave Depth
¾ From the Interface Configuration EXEC mode
cable downstream <port> interleave-depth <interleave depth>
RDN> en
RDN# config
RDN(config)# int cable 5/0
RDN(config-if)# cable downstream 0 interleave-depth 16
Page 29
Module 7, Page 30
Parameters (cont.)

¾ Modulation
cable downstream <port> modulation <QAM-type [64 or 256]>
RDN> en
RDN# config
RDN(config-if)# cable downstream 0 modulation 64
Page 30
Module 7, Page 31
Parameters (cont.)

¾ Unshut port
no cable downstream <port> shutdown
RDN> en
RDN# config
RDN(config-if)# no cable downstream 0 shut
Page 31
Module 7, Page 32
Upstream
Up To 10 Mbps FIBER
CABLE
5 - 42 MHz PLANT
CMTS
0.2 - 3.2 MHz BW US
QPSK or 16 QAM
FDM/TDM/ADM F.O.
NODE
CM
SPLITTER
CPE
ORx COAX
CABLE
PLANT
-16 to +26 dBmV

(Typical)
+8 to +58 dBmV
Upstream power is negotiated

between the CM and CMTS
DOCSIS requires that a Cable Modem be capable of transmit digital information at

a raw data rate up to 10 Mbps, modulated on a upstream carrier placed
somewhere between 5 and 42 MHz. The Cable Modem can use QPSK or 16
QAM modulation and occupy a bandwidth of .2, .4, .8, 1.6 or 3.2 MHz. Upstream
frequency, bandwidth and many other upstream transmission parameters are
dictated to Cable Modems by the CMTS in the UCDs.
If Cable Modems using the same upstream channel were allowed to transmit
upstream at any time they wanted, their transmissions would often collide.
Therefore, Cable Modem upstream transmissions take place in short bursts (a
digitally modulated carrier is not continually present like it is for a DOCSIS
downstream). Ideally, at any one instant, only one Cable Modem would be
allowed to generate a burst. This means that bursts from different Cable Modems
should never collide.
The CMTS uses MAPs to control Cable Modem use of upstream bandwidth.
Before sending a large chunk of data upstream, a Cable Modem must first send a
tiny packet requesting permission. Then, via an entry into the next MAP, the
CMTS informs the Cable Modem of its upstream bandwidth reservation (when it
may send its data upstream).
DOCSIS says a Cable Modem must be capable of setting its upstream burst
carrier power to anywhere +8 to +58 dBmV. This is a wide range (a 58 dBmV
signal has 105 times the power of an 8 dBmV signal).
Page 32
Module 7, Page 33
Upstream Signal Profile
The downstream signal contains the following information for the modems:
•Upstream channel descriptors
•Mapping of mini-slots
•Request region (contention)
•Transmit opportunity (reserved)
•Maintenance area (contention and reserved)
•Synchronization
Page 33
Module 7, Page 34
Upstream Channel Descriptors
ÔUCDs
¾ Describe general upstream channel characteristics:
» Center frequency
» Channel width
¾ Describes each burst type:
» Initial maintenance
» Request
» Request/data
» Periodic maintenance
» Short data
» Long data
The UCD contains data that describes the characteristics of a particular upstream
channel. Within a UCD, the mini-slot size, upstream channel id, and downstream
channel id is specified. Also there are physical layer characteristics describing
each type of burst transmission ie; symbol rate, frequency, preamble length, and
burst descriptor. Burst descriptors further describe each information element
which include the amount of FEC that will be used, maximum burst size, and
guard time size.
The upstream bandwidth allocation map (MAP) contains information for the cable
modem to know where and when to transmit. MAP’s contain information elements
that are described by interval usage codes (IUC). There are six basic IUC’s.
They are Request, REQ/Data, Initial Maintenance, Station Maintenance, Short
Data Grant, and Long Data Grant. In essence there are six basic types of bursts
the cable modem will transmit to the CMTS. Each IUC as a burst descriptor
explaining it. The burst descriptor is located in the UCD.
Page 34
Module 7, Page 35
Maps
ÔDescribe the upstream timeline

ÔWhich modem can transmit and when
ÔMap elements specify:
¾ Burst type
¾ Which modem (may be broadcast)
¾ When
Time
Maps describe the upstream timeline into the future and can be adjusted for plant
delay. Map elements specify:
•Burst type
•Request
•Maintenance
•Transmit data
•Which modem
•By SID
•Where on the time line
The upstream channel is modeled as a string of many slots. The CMTS must
generate the time and reference for identifying the slots. It must also control
access to the slots by the cable modems. For example, it may grant some
number of contiguous slots to a CM for it to transmit a data PDU.
The upstream bandwidth allocation map (MAP) notifies all modems the timing
nature of the upstream transmission opportunities of the near future. It is a MAC
management message that contains information elements. Each information
element defines the allowed usage for a range of mini slots. Maps also contain
acknowledgments for data transmitted during contention intervals.
Maps vary in length. Under light load, maps are small to allow for the best
possible access latency. Under heavy load, maps get longer as transmission
grants consume a larger portion of the upstream bandwidth. The minimum map
size is determined by (a) the minimum number of contention slots that are usefully
considered at one time, and (b) transmission overhead on the downstream
channel. The maximum map size is determined mainly by how far into the future
the modems will want to track their transmission opportunities (limited in the
DOCSIS specs to 4096 mini-slots).
Page 35
Module 7, Page 36
Data Transfer
ÔUpstream
¾ Contention Data:
» Wait for Req/Data in Map, then transmit.
¾ On Request:
» Transmit a request, wait for grant
¾ Scheduled:
» Wait for a grant
Since additional services such as digital video may be provided on the same
physical layer as the Data-Over-Cable, a DOCSIS modem implements a
downstream transmission convergence sublayer between the physical layer and
the MAC sublayer. This defines the downstream bitstream as a continuous series
of 188-byte MPEG which is a 4-byte MPEG header followed by 184-bytes of
payload. The MPEG transmission convergence sublayer filters all MPEG frames
except those containing MAC payloads. It further strips the MPEG header and
any MPEG stuffing bytes prior to presenting the assembled MAC frame to the
MAC sublayer.
The modem’s MAC sublayer monitors the downstream channel for data to be
forwarded. If configured to do so, it will accept the MAC frames; otherwise, it
discards the frame. A filter/forward/process decision is made and the frame is
either discarded, forwarded to the Ethernet and/or processed by the modem itself.
It monitors its reception rate to keep it within the maximum downstream data rate
for the class of service. Broadcast packets are not included in the reception rate.
While monitoring the downstream channel for data and MAC Management
frames, the modem records indications of the quality of the signal it receives for
management reporting.
Page 36
Module 7, Page 37
Upstream Power vs. Location
Downstream
29-8 20-4 14-8 11-8
Upstream
Å Higher Lower Æ
Due to attenuation
In a properly adjusted and maintained cable plant, the settled upstream level from
different CMs along a cable plant distribution line will naturally vary. This is a
result of several factors:
•CM/CMTS level adjustment - During the Cable Modem boot process, the CMTS
and a Cable Modem talk with one another to get the upstream transmit power
level adjusted so that upstream bursts arrive at the CMTS receiver at the CMTS
target level. Once a CM has completed this ranging process, its actual upstream
transmit level has settled at somewhere between 8 and 58 dBmV (by DOCSIS
RFI specification).
•Upstream plant balancing - Cable system upstream amplifiers are typically
adjusted to provide unity gain input-to-input in the return direction. Attenuation is
very low at DOCSIS upstream frequencies compared with the cable plant’s
downstream design frequency.
Downstream plant balancing - Cable plants are designed with high tap values
near a DS amplifier output (upstream amplifier inputs) and lower tap values further
downstream before the next amplifier. This is done to keep the downstream
output levels at all tap ports about the same.
The settled upstream transmit levels of Cable Modems installed along an average
coax cable distribution line will normally be on the higher end of the 8 - 58 dBmV
allowed range for Cable Modems close to a upstream amplifier input and on the
lower end of the range for Cable Modems close to a upstream amplifier output.
The 50 dB US transmit level range DOCSIS Cable Modems can provide is quite
large. It is good engineering practice to try to reserve 5 - 10 dB of the top and
bottom of this range as an operating margin.
Page 37
Module 7, Page 38
Configuring Upstream Parameters
ÔConfigure Upstream Parameters

¾ Channel Width
cable upstream <port> channel-width <channel-width>
RDN> en
RDN# config
RDN(config-if)# cable upstream 0 channel-width 1600000
Page 38
Module 7, Page 39

(cont.)

¾ Frequency
cable upstream <port> frequency <frequency>
RDN> en
RDN# config
RDN(config-if)# cable upstream 0 frequency 32000000
Page 39
Module 7, Page 40

(cont.)

¾ Received Power Level
cable upstream <port> power-level <power-level>
RDN> en
RDN# config
RDN(config-if)# cable upstream 0 power-level 0
Page 40
Module 7, Page 41

(cont.)

¾ Pre-configured Modulation Profile
cable upstream <port> modulation-profile <modulation-profile>
• Only two pre-defined (1 or 2)
RDN> en
RDN# config
RDN(config-if)# cable upstream 0 modulation-profile 2
Page 41
Module 7, Page 42

(cont.)

¾ Unshut port
no cable upstream <port> shutdown
RDN> en
RDN# config
RDN(config-if)# no cable upstream 0 shut
Page 42
Module 7, Page 43
Upstream Modulation Rates
ÔChannel Parameters
¾ Single Center Frequency
» Expressed in MHz
¾ Single Carrier with multiple widths - width dependant on baud rate
¾ Preamble length
Symbol Rate as Channel Width in QPSK Speed in bits per 16QAM speed in bits
symbols per second Kilohertz to Megahertz second* per second**
160.000 sym/sec 200 KHz 320,000 bps 640,000 bps
320,000 sym/sec 400 KHz 640,000 bps 1,280,000 bps
640,000 sym/sec 800 KHz 1,280,000 bps 2,560,000 bps
1,280,000 sym/sec 1.6 MHz 2,560,000 bps 5,120,000 bps
2,560,000 sym/sec 3.2 MHz 5,120,000 bps 10,240,000 bps
* QPSK modulates 2 bits per symbol

**QAM 16 modulates 4 bits per symbol
The upstream modulator provides QPSK at 160, 320, 640, 1,280, and 2,560
ksym/sec, and 16QAM at 160, 320, 640, 1,280, and 2,560 ksym/sec. This variety
of modulation rates, and flexibility in setting upstream carrier frequencies, permits
operators to position carriers in gaps in the pattern of narrowband ingress, as
discussed later in the course
The upstream symbol rate MUST be fixed for each upstream frequency.
Page 43
Module 7, Page 44
Performance
ÔClean Environment
¾ Short preamble (less latency)
¾ Less FEC
¾ Higher Symbol rates
¾ 16 QAM
ÔNot-So Clean Environment

¾ Longer Preambles
¾ More FEC
¾ Lower Symbol Rate
¾ QPSK
¾ Slower
Page 44
Module 8, Page 0
Module 8
CM/CMTS Communication
Rev. 1.0 11/01
Page 0
Module 8, Page 1
Introduction
Introduction The BSR, by design, delivers transparent IP

traffic from the CM-CI (CPE) through the
CMTS-NSI. This is accomplished through
configuration of the CMTS r/f interfaces, as well
as the network side interfaces.
Importance In order to provide IP traffic from CPE to

CMTS-NSI, we need to configure the CMTS on
the BSR.
Lesson Overview We will discuss configuring the CMTS

interfaces on the BSR. We will examine all of
the commonly altered variably configured
parameters.
Page 1
Module 8, Page 2
Objectives

ÔDescribe the DOCSIS-specified ranging and
registration process
ÔSuccessfully range and register a cable modem
on the BSR1000/64000
Page 2
Module 8, Page 3
Topic Order
Topic Order CMTS/CM Interaction Page 5

Viewing Ranging and Registration Page 6
Cable Modem Boot Process Page 7
Downstream Tuning Page 8
CM Downstream Signal Profile Page 9
Maps Page 10
Ranging Page 11
Upstream Power vs. Location Page 12
Initial Ranging Summary Page 13
Connection Page 14
Configuring Helper Parameters Page 15
Dynamic IP Configuration Page 17
Introduction to DHCP Page 18
DHCP Scopes Page 20
DHCP Server Page 21
Page 3
Module 8, Page 4
Topic Order (cont.)
Topic Order IP Address Allocation in DHCP Page 22

Connection: DHCP Summary Page 29
Debugging Connection Process Page 30
DOCSIS ToD Server Requirements Page 31
Time of Day (ToD) Page 32
DOCSIS TFTP Server Requirements Page 33
TFTP Process Page 34
Configuration Page 35
Registration Page 36
Registration: MICs Page 39
Debugging Ranging and Registration Page 40
Baseline Privacy Page 41
BPI Service Goals Page 42
Security Association Page 44
Periodic Maintenance Page 46
Page 4
Module 8, Page 5

¾ Configure DOCSIS-Specified OSS
DHCP
TFTP (CM Configuration File)
ToD
¾ Ensure Ranging and Registration
Tune
Range
Connectivity
Configuration
Registration
It is also necessary to configure operational support services outside of the
CMTS, including DHCP, ToD and TFPT servers. The last step is to ensure that
all cable modems have successfully ranged and registered.
Page 5
Module 8, Page 6
CMTS/CM Interaction
Tuning
ÔScan for downstream channel and sync with the
CMTS
Ranging ÔObtain the transmit parameters (from UCD
message)
ÔPerform ranging
Connection ÔEstablish IP connectivity
ÔEstablish time of day
Configuration ÔTransfer operational parameters
ÔPerform registration
Registration
Maintenance
Page 6
Module 8, Page 7
Viewing Ranging and Registration
Ô Show cable modem command

show cable modem [<mac> | <i.p. address> hosts]
• Can specify detail | offline | registered | unregistered | summary
as arguments
RDN> en
RDN# sh cable modem
cm->mac: 0030.ebff.033
Interface Upstream Prim Connect Timing Rec Ip Address Mac Address
IfIndex Sid State Offset Power
Cable 0/0 4 1 online(pk) 1239 109 10.200.220.2 0030.ebff.f03
cm->mac: 0050.f112.2563
Cable 0/0 4 2 online(pt) 1228 116 10.200.220.3 0050.f112.2563
Total cable modems reg: 2
Total cable modems other state: 0
Page 7
Module 8, Page 8
Cable Modem Boot Process
Tuning
Ranging
Connection
CM Listens
Configuration for CMTS
downstream
transmission
Registration
ÔCM searches for a downstream data channel
ÔSynchronize with QAM
Maintenance
ÔSynchronize with FEC and MPEG
Cable modems search the downstream in 6 MHz intervals for both Standard and
HRC channel plans. The spectrum scanned is from 90 -860 MHz. The modem
stores the downstream frequency of a DOCSIS signal in flash if it completes
registration. Modems can take upwards of 1/3 second for each downstream.
Modem registration can take a very long time if multiple DOCSIS downstream
signals are present. Once a modem has locked onto a signal it will find UCDs and
mapping information. From the mapping information it knows when it will be able
to Transmit an initial ranging request. The initial ranging request is transmitted in
the contention region of the upstream signal and may collide with other signals.
The CMTS will reply with frequency, timing and level adjustments. The CM stores
the ranged power level in flash.
Page 8
Module 8, Page 9
Downstream Tuning
Analog Digital TV
5 @ 1 sec 2 sec DOCSIS
7 8 9 10 11 44 101 108
STD/IRC Channels >6-134
HRC Channels >6-134
When a Cable Modem boots, it begins tuning at 93 MHz, the lowest standard
channel plan DOCSIS CMTS downstream channel. It tunes to all the DOCSIS
standard channel frequencies in 6 MHz steps, then all the HRC channel
frequencies in 6 MHz steps. If it doesn’t detect a downstream carrier, it will cycle,
and start all over again.
When the Cable Modem encounters a digital downstream channel, it studies it
closely to decide if it is a DOCSIS channel . If it finds no DOCSIS SYNCs
broadcast on the digital channel, it resumes scanning.
The Cable Modem will scan across analog TV channels at a rate of about 5 per
second. Digital TV channels take about 2 seconds per channel.
Page 9
Module 8, Page 10
CM Downstream Signal Profile
The downstream signal contains the following information for the modems:
Upstream channel descriptors
Mapping of mini-slots
Request region (contention)
Transmit opportunity (reserved)
Maintenance area (contention and reserved)
Synchronization
Page 10
Module 8, Page 11
MAPS
MAPS Describe the upstream timeline into the future. Can be adjusted for plant
delay. Map elements specify:
Burst type
Request
Maintenance
Transmit data
Which modem
By SID
Where on the time line
Page 11
Module 8, Page 12
Ranging
Tuning
•Adjusts for location in cable plant

Ranging
•Power levels
Connection
Configuration
Registration
Maintenance
UCD’s describe general upstream channel characteristics:

Center frequency
Width
Describes each burst type:
Initial maintenance
Request
Request/data
Periodic maintenance
Short data
Long data
Burst descriptors:
Modulation scheme for each burst type
FEC
Preamble length
Differential encoding
• The power level of the upstream signal should be high enough to achieve a
workable margin above noise and interference.
• Uniform power loading for all signals on the upstream per unit bandwidth is
calculated
Page 12
Module 8, Page 13
Upstream Power vs. Location
Downstream
29-8 20-4 14-8 11-8
Upstream
Å Higher Lower Æ
Due to attenuation
In a properly adjusted and maintained cable plant, the settled upstream level from
different CMs along a cable plant distribution line will naturally vary. This is a
result of several factors:
•CM/CMTS level adjustment
During the Cable Modem boot process, the CMTS and a Cable Modem talk with
one another to get the upstream transmit power level adjusted so that upstream
bursts arrive at the CMTS receiver at the CMTS target level. Once a CM has
completed this ranging process, its actual upstream transmit level has settled at
somewhere between 8 and 58 dBmV (by DOCSIS RFI specification).
Upstream plant balancing. Cable system upstream amplifiers are typically
adjusted to provide unity gain input-to-input in the return direction. Attenuation is
very low at DOCSIS upstream frequencies compared with the cable plant’s
downstream design frequency.
Downstream plant balancing. Cable plants are designed with high tap values near
a DS amplifier output (upstream amplifier inputs) and lower tap values further
downstream before the next amplifier. This is done to keep the downstream
output levels at all tap ports about the same.
The settled upstream transmit levels of Cable Modems installed along an average
coax cable distribution line will normally be on the higher end of the 8 - 58 dBmV
allowed range for Cable Modems close to a upstream amplifier input and on the
lower end of the range for Cable Modems close to a upstream amplifier output.
The 50 dB US transmit level range DOCSIS Cable Modems can provide is quite
large. It is good engineering practice to try to reserve 5 - 10 dB of the top and
bottom of this range as an operating margin.
Page 13
Module 8, Page 14
Initial Ranging Summary
CMTS
Tuning
UCD MAP Rng-Rsp
Ranging
Connection
Configuration
Registration Rng-Req
Maintenance CM
The allocator transmits maps which describe the allowed use of the upstream
bandwidth for some period into the future. The maps may contain:
•Descriptions of contention areas in which modems may request time in the future
for upstream data transmission (Request IEs)
•Descriptions of contention areas in which modems may transmit either small data
packets or requests for future grants (Request/Data IEs)
•Grants of time to a particular Service ID to transmit data upstream. (Short and
Long Data Grants)
•Maintenance requests to particular modems for such activities as periodic loop
delay calculation. (Station Maintenance)
•Acknowledgment of a transmission request without actually granting the request.
(null Data Grants)
•Contention maintenance areas to allow power-up of new modems. (Initial
Maintenance IEs)
Page 14
Module 8, Page 15
Connection
Tuning Ô After the modem has successfully

ranged it must register with the
CMTS for network connectivity.
Ranging Ô DHCP is used to provide for the
following:
¾ IP address
Connection ¾ Lease time
¾ Gateway address
¾ File server and file name
Configuration
¾ Time of day server and offset
Registration
Maintenance
The modem makes a DHCP request. The next step varies between modems.
Typically, the modem makes a TOD request. If the TOD server information is
missing in the DHCP response, or, the ToD server is unreachable, some modems
continue the registration process, while others reboot. The modem then makes a
TFTP request for the MD5 (configuration) file specified in the DHCP response.
The MD5 file contains information for:
Network access
Bandwidth
Priority
SNMP sets
Filters
Community strings
MIC’s (Message Integrity Checks)
The CM MIC is used for file integrity
The CMTS MIC is used for theft of service prevention
Shared Secret
Page 15
Module 8, Page 16
Configuring Helper Parameters
ÔConfigure Helper Parameters

¾ Cable Helper Address
» Tells BSR64000 where to forward UDP traffic (i.e., TFPT
requests for configuration files during Cable Modem Boot)
» cable helper-address <ip address> cable modem | host
RDN> en
RDN# config
RDN(config-if)# cable helper-address 192.168.100.100 cable modem
RDN(config-if)# cable helper-address 192.168.100.100 host
This will only forward UDP traffic from cable modems if that parameter is used or
from CPE if the host parameter is used.
Page 16
Module 8, Page 17
Configuring Helper Parameters

(cont.)
ÔConfigure Helper Parameters

¾ IP Helper Address
» Tells BSR64000 where to forward broadcasts from the CMTS
interfaces
» ip helper-address <ip address>
RDN> en
RDN# config
RDN(config-if)# ip helper-address 192.168.100.100
IP helper addressing is a form of static addressing that cause the router to forward
UDP broadcast packets as an IP unicast to the IP helper address location. To
configure helper addressing, you must specify the ip helper-address command
on every interface on every router that receives a broadcast that needs to be
forwarded. The helper address is specified for UDP broadcast packets destined
to the following port numbers by default:
• Trivial File Transfer Protocol (TFTP) (port 69)
• Domain Naming System (port 53)
• Time service (port 37)
• NetBIOS Name Server (port 137)
• NetBIOS Datagram Server (port 138)
• Boot Protocol (BOOTP) client and server datagrams (ports 67 and 68)
The router places the IP address of the interface which received the broadcast in
the "giaddr" field (meaning Gateway IP Address). This will be the address to
which the DHCP server will send the DHCP response.
Page 17
Module 8, Page 18
Dynamic I.P. Configuration
ÔManually configuring IP parameters for all workstations on a

network can be unwieldy and error-prone
¾ Dynamic IP configuration removes the drawbacks of manual
configuration
¾ Typically follows client-server architecture
» Server manages available IP addresses
• Allows for centralized administration of all IP addresses
» Clients request IP addressing services from server
• Includes IP address, default gateway, etc.
As networks and internetworks grow in size, manually configuring networking

parameters becomes unwieldy. Additionally, as hosts move between networks,
updating these parameters becomes a full time job.
Dynamic configuration protocols are designed as client-server technologies. This
is what allows for centralized control. The majority of processing occurs on the
server-side, with the clients requesting services.
Page 18
Module 8, Page 19
Introduction to DHCP
ÔDynamic Host Configuration Protocol (DHCP) commonly used

¾ Open standard
» Successor to bootstrap protocol (bootp)
• Based on RFC 951
» Developed by the Dynamic Host Configuration working group
(DHC WG) of the Internet Engineering Task Force (IETF)
• Based on RFCs 1533, 1534, 1541 and 1542
¾ Used to assign IP address to cable modems and customer
premises equipment (cpe)
In the cable modem world, hosts are added and removed daily – and are not
physically available to administrators. As such, an automated method of
configuring addressing schemes was necessary. Dynamic Host Configuration
Protocol, based on RFCs 1533, 1534, 1541 and 1542 was developed by the
Dynamic Host configuration working group of the Internet Engineering Task Force
and was subsequently adapted as a part of DOCSIS.
DHCP is based on BOOTP and maintains some backward compatibility.
Page 19
Module 8, Page 20
Introduction to DHCP (cont.)
ÔDHCP (Dynamic Host Configuration Protocol)

¾ Enables automatic assignment of IP address to a host’s TCP/IP
stack software
» Assigns IP address dynamically from pre-defined range of
numbers
• Pre-defined range called a scope
¾ Clients configured to use DHCP don’t need to have a statically
assigned IP address
¾ Runs over UDP on ports 67 and 68
The Dynamic Host Configuration Protocol (DHCP) provides a mechanism through

which hosts using TCP/IP can obtain protocol configuration parameters
automatically through a server on the network.
DHCP is a client-server mechanism, in which the DHCP client (host on the
network) contacts a DHCP server for configuration parameters. The DHCP server
is typically centrally located and operated by the network administrator. Because
the server is run by a network administrator, DHCP clients can be reliably and
dynamically configured with parameters appropriate to the current network
architecture.
The most important configuration parameter carried by DHCP is the IP address. A
computer must be initially assigned a specific IP address that is appropriate to the
network to which the computer is attached, and that is not assigned to any other
computer on that network. If a computer moves to a new network, it must be
assigned a new IP address for that new network. DHCP can be used to manage
these assignments automatically.
Page 20
Module 8, Page 21
DHCP Scopes
ÔDHCP Scopes
¾ IP address pool a client leases from
¾ Managed by DHCP server
» May manage multiple scopes
¾ May contain additional parameters
» Options
A scope is a data structure managed by a DHCP server that contains a pool of IP

addresses and options to send to requesting clients. A DHCP server may
manage multiple scopes.
Page 21
Module 8, Page 22
DHCP Server
ÔDHCP Server
¾ Assigns IP addresses to clients
Addresses are “leased” to clients for a period of time
IP addresses can be reserved for specific clients
DHCP Clients may be filtered based on their MAC addresses
Server
Address may be assigned from different scopes based on
extended options
Cable modems will be assigned an IP address from a scope (usually a non

routable pool of addresses) that is different from the scope for the customer
premise equipment (routable IP addresses).
Page 22
Module 8, Page 23
I.P. Address Allocation in DHCP
ÔClient-side view of IP address allocation process
Initialization
Selection Rebinding
Request Renewing
Binding
The DHCP process, from a client’s perspective occurs in six phases:

•Initialization
•Selection
•Request
•Binding
•Renewing
•Rebinding
Page 23
Module 8, Page 24

(cont.)
ÔInitialization
¾ DHCPDISCOVER
Message sent
¾ Broadcast
¾ Contains MAC and
hostname
DHCPDISCOVER
Broadcast MAC address
To 255.255.255.255 to
BootP UDP port 67
Initialization
Selection Selection
Request Renewing
Binding
When a DHCP client starts up, it’s TCP/IP stack initializes with an IP address of
0.0.0.0 (null IP address). With TCP/IP enabled, the DHCP client broadcasts a
DHCPDISCOVER message on the local subnet to find information about DHCP
servers. The server port for these broadcasts is the BootP UDP port 67 and the
destination address is 255.255.255.255.
With this DHCPDISCOVER message, the client sends the only information it has
that uniquely identifies it – it’s Media Access Control (MAC) address and
hostname.
Page 24
Module 8, Page 25

(cont.)
ÔSelection
¾ DHCPOFFER
¾ Sent Broadcast DHCPOFFER
Server MAC and IP address
¾ Server MAC and IP Lease with IP address,
Subnet mask and duration
¾ Client IP, Subnet
mask, Lease duration
Initialization
Selection Selection
Request Renewing
Binding
Any DHCP server that receives the DHCPDISCOVER message and can provide
a lease sends a DHCPOFFER message to the client on BootP UDP client port 68.
Because the client is still in it’s initialized state (and has no IP address), this
message is broadcast. The DHCP server sends an IP address offer to the client,
including the subnet mask and lease duration. The server includes it’s MAC and
IP address in the message to facilitate further communication with the DHCP
client.
If no server responds with a DHCP offer message, the DHCP client will retry
several times. If it continues to not receive a response, it goes into a sleep state
for some amount of time.
Page 25
Module 8, Page 26

(cont.)
ÔRequest
¾ DHCPREQUEST
¾ Sent to server IP
address
¾ Requests options
» Configuration file
DHCPREQUEST
» Default Gateway Server IP address
Request for options
Initialization
Selection Selection
Request Renewing
Binding
When a client receives DHCPOFFER messages from servers on the network, it

selects one. Typically, hosts are configured to choose the first server that
responded to it with a DHCPOFFER message. The client sends a
DHCPREQUEST message. The minimum information in a DHCPREQUEST
message is the server’s IP address. The client can request other information (or
options), such as default gateway, configuration file, etc.
Page 26
Module 8, Page 27

(cont.)
ÔBinding
¾ DHCPACK
¾ DHCP lease
DHCPACK
information sent Lease information
¾ Requested options Options requested
sent
¾ Sent broadcast
Initialization
Selection Selection
Request Renewing
Binding
When a DHCP server receives a DHCPREQUEST message, it responds with a

DHCPACK (DHCP acknowledgment) message. This message contains lease
information (IP address, subnet mask, lease duration) that was negotiated
previously as well as other parameters (options) that were requested by the client.
Because the client’s TCP/IP stack is still not initialized, the DHCPACK message is
sent via a broadcast.
As soon as the client receives this lease, it finishes initializing it’s TCP/IP stack.
Page 27
Module 8, Page 28

(cont.)
ÔRenewing
¾ DHCPREQUEST
¾ Sent directly to server
DHCPACK
¾ DHCPACK from server Lease information
Options requested
¾ Lease renewal
¾ Updates to options DHCPREQUEST
Server IP address
Request for options
Initialization
Selection Selection
Request Renewing
Binding
Since DHCP leases are configured to expire after some amount of time, clients
need a means of renewing their leases. A DHCP client enters two states during
renewing a lease:
•Renewing – When the lease is 50% expire
•Rebinding – When the lease is 87.5% expires and renewing has failed.
The client attempts to renew it’s lease from the server that originally provided it.
The client sends a DHCPREQUEST message directly to the DHCP servers
asking to renew it’s current lease.
The DHCP server will respond with a DHCPACK message if it can renew the
lease. The message will include any updated parameters (options).
Page 28
Module 8, Page 29

(cont.)
ÔRebinding
¾ DHCPREQUEST
¾ Broadcast
DHCPACK
¾ Listens for DHCPACK Lease information
Options requested
¾ Renewal information
DHCPREQUEST
Server IP address
Request for options
Initialization
Selection Selection
Request Renewing
Binding
If for some reason the client can’t communicate with the DHCP server it originally
obtained it’s lease from, the client enters the rebinding state at 87.5% lease
expired. In this state, the client will contact any DHCP server by broadcasting it’s
DHCPREQUEST message with it’s current lease information.
If any server can renew the lease, it will send a DHCPACK message. If, for some
reason, a DHCPNACK message (DHCP not acknowledged) is received, the client
will discontinue using the current IP address immediately. It will then re-enter it’s
initializing state and try to acquire a new lease.
Page 29
Module 8, Page 30
Connection: DHCP Summary

DHCP Server
Tuning
Offer Response
Ranging
Connection
CMTS
Configuration
Discover Request IP Address

Registration
Gateway
TFTP Server
Maintenance CM
Config File Name
The following fields are typically expected in the DHCP response returned to the
CM:
The IP address to be used by the CM (yiaddr).
The IP address of the TFTP server for use in the next phase of the bootstrap
process (siaddr).
If the DHCP server is on a different network (requiring a relay agent), then the IP
address of the relay agent (giaddr). Note: this may differ from the IP address of
the first hop router.
The name of the CM configuration file to be read from the TFTP server by the CM
(file).
The subnet mask to be used by the CM (Subnet Mask, option 1).
The time offset of the CM from Universal Coordinated Time (UTC) (Time Offset,
option 2). This is used by the CM to calculate the local time for use in time-
stamping error logs.
A list of addresses of one or more routers to be used for forwarding CM-originated
IP traffic (Router Option, option 3). The CM is not required to use more than one
router IP address for forwarding.
A list of [RFC-868] time-servers from which the current time may be obtained
(Time Server Option, option 4).
A list of SYSLOG servers to which logging information may be sent (Log Server
Option, option 7)
The CM MUST configure itself based on the DHCP response.
Page 30
Module 8, Page 31
Debugging Connection Process
ÔDebugging DHCP/TFTP-Related Behaviors

» debug ip udp [dhcp]
• Allows for watching source/destination of DHCP/UDP messages
RDN> en
RDN# debug ip udp dhcp
UDP DHCP Debugging is turned on
…
Page 31
Module 8, Page 32
DOCSIS ToD Server Requirements
ÔToD Server
¾ Internet Time Protocol (ITP)
RFC 868
¾ UDP and TCP requests honored on port 37

ToD
Server
¾ 32-bit value defining the number of seconds since 00:00

(midnight January 1, 1900 GMT)
This protocol provides a site-independent, machine readable date and time. The
Time service sends back to the originating source the time in seconds since
midnight on January first 1900.
This protocol may be used either above the Transmission Control Protocol (TCP)
or above the User Datagram Protocol (UDP).
The Time of Day server provides a 32 bit value which contains the correct time (in
seconds) since 1/1/1900 at midnight.
Page 32
Module 8, Page 33
Time of Day (ToD)
Ô Configure system time of day server location

Specify Time of Day (ToD) server location
» time-of-day server <ip-address> <time-zone>
• Verify with show time-of-day server command
newhostname> en
newhostname# time-of-day server 192.168.115.69
newhostname#
All ToD requests from cable modems will be forwarded to the

address indicated. Most modern modems will continue operations
even if they do not find a Time of Day server, but, it is considered
best practice to ensure that one is available anyway.
Page 33
Module 8, Page 34
DOCSIS TFTP Server

Requirements
ÔTFTP Server
¾ Trivial File Transfer Protocol
(RFC 1350)
TFTP
Server ¾ UDP port 69
¾ Small and easy to implement
¾ Read and write to and from remote servers
The TFTP server contains the configuration file for the cable modem.
TFTP is a simple protocol (RFC1350) to transfer files, and therefore was named
the Trivial File Transfer Protocol or TFTP. It has been implemented on top of the
Internet User Datagram protocol (UDP or Datagram).
UDP port 69 is used for the transfer process. Port 69 is a software port in the
TFTP protocol.
Page 34
Module 8, Page 35
TFTP Process
¾ The following settings MUST be included in the configuration file:

Network Access Configuration Setting
¾ The following settings are optional:

Downstream Frequency
Upstream Channel ID
Class of Service
Vendor ID
Baseline Privacy
TFTP Software Upgrade filename
Server SNMP Write-Access Control
SNMP MIB Object
Software Server IP Address
CPE Ethernet MAC Address
Maximum Number of CPE’s (32 Max)
Pad Configuration
SNMP IP Address (if applicable)
Telephone Settings (if applicable)
Vendor-Specific Configuration (if applicable)
Only one parameter is needed for DOCSIS 1.0 and that is the Network setting.
Other parameters ( downstream, upstream, class of service) are needed for
Service Class provisioning.
SNMP parameters if SNMP will be used.
CPE MAC address for enhanced security.
Page 35
Module 8, Page 36
Configuration
Tuning ÔAfter the modem has successfully

acquired an IP address, it must be
given some basic configuration
information
Ranging
ÔThe configuration file name and
location provided during the DHCP
Connection process is used to make the TFTP
request
Configuration
Registration
Maintenance
The modem makes a DHCP request. The next step varies between modems.
Typically, the modem makes a TOD request. If the TOD server information is
missing in the DHCP response, or, the ToD server is unreachable, some modems
continue the registration process, while others reboot. The modem then makes a
TFTP request for the MD5 (configuration) file specified in the DHCP response.
The MD5 file contains information for:
Network access
Bandwidth
Priority
SNMP sets
Filters
Community strings
MIC’s (Message Integrity Checks)
The CM MIC is used for file integrity
The CMTS MIC is used for theft of service prevention
Shared Secret
Page 36
Module 8, Page 37
Registration
Tuning
Ranging
Connection
Configuration
ÔThe modem only comes online

Registration after it has registered with the
CMTS
Maintenance ¾ Reports that all configuration parameters
received and applied
At this point of time a modem can join in the network and transmit PDUs.
Page 37
Module 8, Page 38
Registration (cont.)
ÔCM generates a Registration Request (REG-REQ)

ÔIncludes configuration parameters received from TFTP
configuration file:
¾ Downstream frequency, Upstream channel ID
¾ Network access configuration settings
¾ Class of Service
CMTS
¾ Modem Capabilities
¾ Modem IP address
REG-REQ
HFC
Cable Modem
The modem will now try to register with the CMTS via a Registration Request.
Page 38
Module 8, Page 39
Registration (cont.)
ÔCMTS
¾ Checks CM’s MAC address and authentication signature
¾ Assigns a SID
¾ Provides bandwidth for CM requested Class of Service
¾ Modifies forwarding table to allow full user data if the modem
requested Network Access
¾ Sends REG-RSP to CM CMTS
» CM can pass unencrypted data
REG-RSP
HFC
Cable Modem
The CMTS will respond with a SID for the modem.

The modem is now online.
CM can pass unencrypted data only if the SID is defined as unencrypted.
CM MIC is a digest which ensures that the data sent from the provisioning server
were not modified en route.
CMTS MIC is a digest used to authenticate the provisioning server to the CMTS
during registration. It is taken over a number of fields one of which is a shared
secret between the CMTS and the provisioning server.
The authentication string is a shared secret between the provisioning server
(Which creates the configuration files) and the CMTS. It allows the CMTS to
authenticate the CM provisioning.
Page 39
Module 8, Page 40
Registration: MICs
Tuning
Things the
Ranging CMTS needs
to know about
Connection
CMTS MIC
Configuration
Registration
CM MIC
Filters, e.g.
Maintenance
The CMTS MIC allows the CMTS (and, by extension, the cable operator) to verify
that the operational parameters and class of service that the CM is asking for
were provisioned by an authorized TFTP server. Because the CMTS MIC is
calculated with a secret shared between the CMTS and the provisioning server
(TFTP file creation software), the modem or subscriber is unable to forge this
digital signature. The secret key is provisioned by the operator at both the
provisioning server and the CMTS.
The CM MIC is simply a digest over all of the parameters in the TFTP file that the
CMTS would not be interested in. Because the TFTP file could run to thousands
of bytes, it is easier to send up this 16-byte digest than to send the thousands of
bytes needed to recalculate the CMTS MIC. The purpose of the CM MIC is to
guarantee the integrity of things like filters. Without it, one could mount a denial-of-
service attack by putting a TFTP server on the network that copied the CMTS
MIC-needed fields exactly but tacked on a bogus filter-all SNMP object.
The shared secret referenced above is an MD5 file. MD5 is the result of an
algorithm that takes as input a message of arbitrary length, and produces as
output a 128-bit “fingerprint” or “message digest” of the input.
The MD5 algorithm is intended for digital signature applications, where a large file
must be “compressed” in a secure manner.
Page 40
Module 8, Page 41
Debugging Ranging and

Registration
ÔDebugging DHCP/TFTP-Related Behaviors

» debug ip udp [dhcp]
• Allows for watching source/destination of DHCP/UDP messages
RDN> en
RDN# debug ip udp dhcp
UDP DHCP Debugging is turned on
…
Page 41
Module 8, Page 42
Baseline Privacy Interface (BPI)
ÔFollows modem registration

ÔProvides user data privacy by encrypting traffic flows, upstream and
downstream
ÔProvides cable operators basic protection from theft of service
ÔMechanisms for:
¾ authentication: CM to CMTS and CMTS to CM
¾ key distribution: traffic keys and lifetimes
¾ data encryption applied to Sid's
Baseline Privacy uses the Cipher Block Chaining (CBC) mode of the US Data
Encryption Standard (DES) algorithm [FIPS-46, FIPS-46-1, FIPS-74, FIPS-81] to
encrypt the Packet PDU field in both upstream and downstream RF MAC Packet
Data PDUs. The MCNS MAC headers of these Packet Data PDUs are not
encrypted. The payloads, as well as headers, of MAC management messages are
sent in the clear to facilitate registration, ranging, and normal operation of the
MAC sublayer.
The MCNS Radio Frequency Interface Specification [MCNS1] defines a Service
ID (SID) as a mapping between CM and CMTS for the purposes of upstream
bandwidth allocation and class-of-service management. When Baseline Privacy is
in operation, the SID also identifies a particular security association. The SID, in
combination with other components of the extended header element, identifies to
a modem the keying material required to decrypt the MAC PDU's Packet Data
field.
Baseline Privacy’s key management protocol runs between CM and CMTS; CMs
use the protocol to obtain authorization and traffic keying material (pertaining to a
particular SID) from the CMTS, and to support periodic reauthorization and key
refresh. The key management protocol uses the RSA [RSA, RSA1] public-key
encryption algorithm, and the Electronic Codebook (ECB) mode of DES [FIPS-81]
to secure key exchanges between CM and CMTS. CMs have RSA private/public
key pairs factory-installed in flash memory. Once a key pair is generated, it is
retained for the operational life of the CM.
A SID's keying material (DES key and CBC Initialization Vector) has a limited
lifetime. When the CMTS delivers SID keying material to a CM, it also provides
the CM with that material's remaining lifetime. It is the responsibility of the CM to
request new keying material from the CMTS before the set of keying material the
CM currently has expires at the CMTS.
Page 42
Module 8, Page 43
Baseline Privacy Interface Plus

(BPI+)
Ô BPI+ security goals are twofold:
¾ provide cable modem users with data privacy across the cable network, and
¾ provide MSOs with service protection
Ô BPI+ provides a level of data privacy across the shared medium cable
network equal to or better than that provided by dedicated line network
access services (analog modems or digital subscriber lines)
Ô The protected RF MAC data communications services fall into three
categories:
¾ Best-effort, high-speed, IP data services;
¾ QoS (e.g., constant bit rate) data services; and
¾ IP multicast group services.
Ô The earlier BPI specification had “weak” service protection
¾ Key management protocol did not authenticate CMs
¾ BPI+ strengthens this service protection by adding digital-certificate based CM
authentication to its key exchange protocol
Ô BPI+, as an extension of BPI, is entirely backwards and forwards
compatible with the previous specification
¾ BPI+ devices will interoperate in networks with CMs or CMTSs implementing
BPI, and vice versa
Baseline Privacy Plus (BPI+) provides cable modem users with data privacy
across the cable network. It does this by encrypting traffic flows between CM and
CMTS.
In addition, BPI+ provides cable operators with strong protection from theft of
service. The protected DOCSIS MAC data communications services fall into three
categories:
• best-effort, high-speed, IP data services;
• QoS (e.g., constant bit rate) data services; and
• IP multicast group services.
Under BPI+, the CMTS protects against unauthorized access to these data
transport services by enforcing encryption of the associated traffic flows across
the cable network. BPI+ employs an authenticated client/server key management
protocol in which the CMTS, the server, controls distribution of keying material to
client CMs.
The original BPI specification had “weak” service protection because the
underlying key management protocol did not authenticate CMs. BPI+ strengthens
this service protection by adding digital-certificate based CM authentication to its
key management protocol.
Page 43
Module 8, Page 44
BPI+ Architecture
ÔBaseline Privacy Plus has two component protocols:

¾ An encapsulation protocol for encrypting packet data across the cable
network. This protocol defines
» the frame format for carrying encrypted packet data within DOCSIS
MAC frames,
» a set of supported cryptographic suites, i.e., pairings of data
encryption and authentication algorithms, and
» the rules for applying those algorithms to a DOCSIS MAC frame’s
packet data.
¾ A key management protocol (Baseline Privacy Key Management, or
“BPKM”) providing the secure distribution of keying data from CMTS to
CMs
» Through this key management protocol, CM and CMTS
synchronize keying data; in addition, the CMTS uses the protocol
to enforce conditional access to network services
Page 44
Module 8, Page 45
Packet Data Encryption (BPI+)
ÔBPI+ encryption services are defined as a set of extended services

within the DOCSIS MAC sublayer
¾ Packet Header information specific to BPI+ is placed in a Baseline
Privacy Extended Header element within the MAC Extended Header.
ÔBPI+ supports a single packet date encryption algorithm
¾ Cipher Block Chaining (CBC) mode of the US Data Encryption
Standard (DES) algorithm
» BPI+ does not pair DES CBC with any packet data authentication
algorithm
BPI+ encrypts a DOCSIS MAC Frame’s packet data; the DOCSIS MAC Frame’s
Header is not encrypted. DOCSIS MAC management messages MUST be sent in
the clear to facilitate registration, ranging, and normal operation of the DOCSIS
MAC sublayer.
Page 45
Module 8, Page 46
BPI Service Goals
Internet
ÔProvides CM users
CMTS with data privacy
across the RF network
ÔProvides MSOs with

basic protection from
theft of service
Baseline Privacy provides cable modem users with data privacy across the RF
network. It provides user data privacy by encrypting traffic flows between CM and
CMTS. Baseline Privacy provides basic protection of service by insuring that a
modem, uniquely identified by its 48-bit IEEE MAC address, can only obtain
keying material for services it is authorized to access.
Page 46
Module 8, Page 47
BPI Service Goals (cont.)
Ô Does not provide data privacy beyond

CMTS Internet the RF network
Ô While providing privacy from the teenage

hacker next door, will not thwart the FBI
or NSA
Ô Does not provide MSOs with strong

protection from theft of service by a
sophisticated hacker
Page 47
Module 8, Page 48
Security Association
ÔIf CM is configured for Baseline Privacy in the modem TFTP

configuration file:
¾ CM sends Authorization Request
» Public key, MAC address, and SID’s
¾ CMTS responds with an Authorization Response
» Authorization Key (encrypted KEK)
» Key Sequence number and Lifetimes
CMTS
» List of SID’s
• for each requested Class of Service
AUTH-REQ
HFC
Cable Modem AUTH-RSP
Baseline Privacy initialization begins with the CM sending the CMTS and
authorization request, containing data identifying the CM (e.g., MAC address), the
CM's RSA public key, and a list of zero or more assigned unicast SID’s that have
been configured to run Baseline Privacy (Zero is the modem is only configured to
run privacy on multicast SID’s).
If the CMTS determines the requesting CM is authorized for these services, the
CMTS responds with an authorization reply containing a list of SID’s (both unicast
and multicast) that the CM is permitted to run Baseline Privacy on. The reply also
includes an authorization key from which CM and CMTS derive the keys needed
to secure a CM's subsequent requests for per-SID traffic encryption keys, and the
CMTS response to these requests. The authorization key is encrypted with the
receiving cable modem’s public key.
Page 48
Module 8, Page 49
Security Association (cont.)
ÔCM requests Key Request for each SID

ÔCMTS responds with DES encrypted TEK for each SID
ÔCM can now pass encrypted data
CMTS
KEY-REQ
HFC
Cable Modem
TEK
After successfully completing authorization with the CMTS, the cable modem
sends key requests to the CMTS, requesting traffic encryption keys to use with
each of its Baseline Privacy SID’s. A CM's traffic key requests are authenticated
using a keyed hash (the HMAC algorithm); the message authentication key is
derived from the authorization key obtained during the earlier authorization
exchange. The CMTS responds with key replies, containing the traffic encryption
keys: the keys are DES encrypted with a key encryption key derived from the
authorization key. Like the Key Requests, Key Replies are authenticated with a
keyed hash, where the message authentication key is derived from the
authorization key.
It is the responsibility of individual cable modems to insure the keys they are using
match those the CMTS is using. Cable modems do this by tracking when a
particular SID’s key is scheduled to expire and issuing a new key request for the
latest key prior to that expiration time.
Page 49
Module 8, Page 50
Periodic Maintenance
Tuning ÔPeriodic ranging

¾ Periodic loop delay, power, equalization
¾ At least every 30 seconds
Ranging
Connection CMTS
Configuration
Rng-Req
Registration
HFC
Maintenance Cable Modem

Rng-Rsp
Periodic ranging occurs at least every 30 seconds to reset:

Delay
Power level
Frequency
Equalization
Periodic ranging can also be used to:
Change UCD’s on-the fly
UCC Request/Response to move a modem to a different upstream
frequency
Page 50
Module 9, Page 0
Module 9
Path Determination
Rev. 1.0 11/01
Page 0
Module 9, Page 1
Introduction
Introduction In a previous module, we explored basic

configuration tasks, including network addressing.
We will now look at how data passed to an interface
is shared with other internetworked devices.
Importance You will need a basic understanding of routing in

order to connect networked devices. This is what will
allow you to enable both basic and advanced
network-based services.
Lesson Overview We will quickly review the OSI model, and contrast
switching and routing. We discuss the basics of
routing, including path selection and forwarding
mechanisms. In addition, we will see why and how
we create static routes, and their benefits and
shortcomings. We will explore dynamic routing
protocols in Modules 10 and 11.
Page 1
Module 9, Page 2
Objectives

ÔExplain basic routing concepts
ÔCreate and verify static routes
ÔCreate and verify default routes
Page 2
Module 9, Page 3
Topic Order
Topic Order OSI Model Page 5

Bridges/Switches and Routers Page 7
Three Switch/Bridge Functions Page 8
How Switches/Bridges Learn
Host Locations Page 9
How Switches/Bridges Filter Frames Page 12
Broadcast and Multicast Frames Page 13
Routers Operate at Layers 1, 2 and 3 Page 14
What is Routing? Page 15
Routing and Routed Protocols Page 17
Static Routes Page 18
Configuring Static Routes Page 19
Static Route Example Page 20
Page 3
Module 9, Page 4
Topic Order (cont.)
Topic Order Default Static Route Page 21

Verifying Routes Page 22
Where do the Routes Go? Page 23
How do I look at the Routing Table? Page 24
Static Routing Summary Page 25
Page 4
Module 9, Page 5
OSI Model
Application
Presentation
Session
Telnet, FTP, TFTP…..

TCP/UDP Transport
ICMP RIP, OSPF

IP (Packet) Network H Source IP Destination IP………
ATM, PPP,Ethernet
MAC Addresses Mac
Datalink Pre Src Mac Dest Mac H Source IP Destination IP
And Frames LLC
Transmission Ethernet, Sonet

And timing Physical 101010101010101010101010101010101010101010
Page 5
Module 9, Page 6
OSI Model (cont.)
Application
Presentation
Session
TCP/UDP Transport
Router
IP Network Packet Packet Network

Switch
MAC Addresses Datalink Mac Ethernet Ethernet
And Frames LLC Frame Datalink Frame Datalink
Transmission
Physical Physical Physical
And timing 10101010 10101010
Ultimately, devices make their forwarding decisions at a layer that provides for
some form of unique identification. You will remember from a previous module
that MAC address allows for unique identification of physical interfaces. It is a
fixed address (burned in by manufacturer, although some devices allow for
altering of MAC addressed), which makes it useful for local unique identification
(and therefore local forwarding decisions), but not remote path determination
decisions. This is because, as networks grow in size and are segmented
(logically and physically), administrator-assigned (and designed) addressing
schemes become important. That is where Layer 3 (or IP) addresses become
important. IP addresses, because they are assigned by the administrator, can be
altered to fit changing topologies and to connect diverse networks. As such, they
are obvious choices for remote (or logical) path determination decisions.
We label a device, such as a switch, bridge or cable modem, a Layer 2 device
because it makes it’s forwarding decisions based on MAC addresses. Routers,
on the other hand, use IP addresses to make their logical path, or forwarding,
decisions – which is why we call them Layer 3 devices.
Page 6
Module 9, Page 7
Bridges/Switches and Routers

Network 1
• Switches (Connect Hosts)
• They extend broadcast domains.
• One domain per VLAN (1 Network)
• Each VLAN is a separate broadcast
domain. A router is required to move
traffic from one VLAN to another-based
on the Network number.
Router
• Routers (Connect Networks)

• Each interface on a router must be
different network number.
• Broadcasts are blocked (by default)
• Use static routes, or dynamic protocols
like RIP, OSPF or BGP
Network 1 Network 2 Network 3
Remembering that switches act at layer 2, it is easy to remember that they are
designed, intrinsically, to connect hosts. That means that we can attach a number
of hosts to a switch, and, although each port will be considered a different collision
domain, they are still within the same broadcast domain (or network segment).
Routers, on the other hand, are intended to connect networks. They can connect
different kinds of networks, geographically disparate networks, or just logically
separate networks. As such, they both divide collision and broadcast domains by
default.
Page 7
Module 9, Page 8
Three Switch/Bridge Functions
MAC Address Table
A B
Forwarding
Forwarding Table
Table
0260.8c01.1111 0260.8c01.3333
E0 E1
C D
E2 E3
0260.8c01.2222 0260.8c01.4444
¾ Address learning
¾ Forward/Filter decision
¾ Loop Avoidance
Switches and bridges have more in common than it may seem at first glance.
Their primary functions are to learn the layer 2 addresses of the devices within
their network segment (that they connect). Based on this knowledge of
connected, local addresses, it is their job to make forwarding (and in many cases
filtering) decisions at layer 2 (based on MAC address).
Another principal function at layer 2 is to avoid network loops.
Page 8
Module 9, Page 9
How Switches/Bridges Learn Host

Locations
MAC Address Table
A B
Forwarding
Forwarding Table
Table
0260.8c01.1111 0260.8c01.3333
E0 E1
C D
E2 E3
0260.8c01.2222 0260.8c01.4444
¾ Initial MAC address table is empty
When switches and bridges are initially powered on, they do not know about the
topology of the networked devices that they are connecting. Their first job will be
to learn of these devices in order to service their data forwarding needs.
Page 9
Module 9, Page 10

Locations (cont.)
MAC Address Table
E0: 0004.ac01.1111
A B
B
0004.ac01.1111
0004.ac01.3333
E0 E1
C D
D
E2 E3
0004.ac01.2222 0004.ac01.4444
¾ Station A sends a frame to station C.

¾ Switch caches the station A MAC address to port E0 by
learning the source address of data frames.
¾ The frame from station A to station C is flooded out to all
ports except port E0 (unknown unicasts are flooded).
When a device needs to send data to another device within the same segment,
the switch or bridge that connects them will remember the MAC address of the
sending device and associate it, internally, with the port that received the data.
Assuming that switch or bridge doesn’t know the location of the destination
device, it will flood the message out all ports other than the receiving port.
Page 10
Module 9, Page 11

Locations (cont.)
MAC Address Table
E0: 0004.ac01.1111
A E3: 0004.ac01.4444 B
0004.ac01.1111 0004.ac01.3333
E0 E1
C D
E2 E3
0004.ac01.2222 0004.ac01.4444
¾ Station D sends a frame to station C.

¾ Switch caches the station D MAC address to port E3 by learning
the source address of data frames.
¾ The frame from station D to station C is flooded out to all ports
except port E3 (unknown unicasts are flooded).
As more messages are received on each of the switches or bridges ports, it builds
this table of known MAC addresses and associated ports until a time when
messages from devices are to stations whose MAC address is know – at which
time it is no longer necessary to flood messages out all ports. The switch or
bridge will continue to update and maintain this table of know addresses and
locations as a primary function.
Page 11
Module 9, Page 12
How Switches/Bridges Filter

Frames
MAC Address Table

E0: 0004.ac01.1111
A E2: 0004.ac01.2222 B
E1: 0004.ac01.3333
E3: 0004.ac01.4444
0004.ac01.1111 0004.ac01.3333
E0 E1
C X D
E2
XE3
0004.ac01.2222 0004.ac01.4444
ÔStation A sends a frame to station C.

ÔDestination is known; frame is not flooded.
Here we see the culmination of the learning operations of the switch or bridge. It
knows the location of the destination host, which allows it to forward the data to
only the port on which it participates. It is essentially filtering on the non-
participating ports from the perspective of a network segment. Image the same
scenario with a hub (layer 1 device) connecting each of these hosts. If Station A
sent a message intended for only Station C, all devices would hear that message
(since hubs perpetuate shared media). This means that each of the network
interfaces at the hosts would be forced to filter messages not intended for them,
which is less efficient.
Page 12
Module 9, Page 13
Broadcast and Multicast Frames
MAC Address Table

A E0: 0004.ac01.1111 B
E2: 0004.ac01.2222
E1: 0004.ac01.3333
E3: 0004.ac01.4444
0004.ac01.1111 0004.ac01.3333
E0 E1
C D
E2 E3
0004.ac01.2222 0004.ac01.4444
¾ Station D sends a broadcast or multicast frame

¾ Broadcast and multicast frames are flooded to all ports
other than the originating port
Broadcast and multicast frames, because they are intended for all hosts within a
local network segment, are forwarded out all interfaces on switches. This means
that it is the receiving hosts’ task to examine the destination address and decide
whether the message is intended for it or not.
Page 13
Module 9, Page 14
Routers Operate at Layers 1, 2 & 3
OSI
Application 7
Presentation 6
Session 5
Transport 4
Network 3 IP ADDRESS
for Each Interface Router
Data Link 2 MAC ADDRESS

cm
Cmts
cm
Physical 1
Saying that routers are Layer 3 devices can be a bit misleading. Although it
makes it logical path determination decisions at layer 3, it must still forward data,
on local segments, according to its layer 2 address, and put data onto physical
media (layer 1).
Page 14
Module 9, Page 15
What is Routing?
.2 .3 ¾ To route packets a router

must know the following:
192.168.130.0 .1 ¾ Network Destination address
¾ Where is it going?
¾ Routes to remote networks
¾ Router maintains a routing table
¾ Do I know about the destination
network?
¾ Next Hop Network (or
Interface)
.1 ¾ Where do I send it? (Which
interface or next hop address)
192.168.120.0
.2 ¾ Filter
¾ Drop or forward traffic based on
policy
Destination addresses
Sources it can learn from
Possible routes
Best route
Maintain and verify routing information
Drop or forward traffic based on policy (access-lists (filters))
Page 15
Module 9, Page 16
What is Routing? (cont.)
Router
Network 1 Network 3
Network 2
¾ Routers are aware of all directly connected networks

¾ Routers must learn about networks that are not directly
connected
Routing is the process by which an item gets from one location to another.
Examples of items that get routed:
•Trains
•Mail
•Telephone Calls
•Traffic (A router moves data traffic)
Page 16
Module 9, Page 17
Routing and Routed Protocols

Ô Routed Protocol
¾ IP, IPX, APPLETALK
» User data packets that are
sent from source host to
destination host.
Routing Protocol
RIP, OSPF, or BGP
• Protocols that allow routers to
exchange information about networks.
• Routers depend on static routes or a
dynamic routing protocol to learn about
remote networks. (Not directly
connected). Routing protocols are
carried in routed protocols
Routed protocols are nothing more than protocols that actually pass user data. By
contrast, routing protocols are used to share data between routers in order to help
determine information about networks. This information is used to determine
paths through connected and remote networks, in order to pass routed (or
routable) protocols with user data.
Page 17
Module 9, Page 18
Static Routes
Stub Network
200.20.20.0 200.10.10.0
A B
B
192.168.1.2 192.168.1.1
¾Static Routes are unidirectional

¾Static routes must be entered in both routers to allow bi-directional
communications to occur
For intercommunication, static routes must be configured in both directions. Static

routes are often used to route traffic to a stub network or other network where only
a single route to that network exists.
Page 18
Module 9, Page 19
Configuring Static Routes
• Command to add a static route

192.168.120.0
• ip route [destination network]
[mask] [next hop address]
192.168.3.2 [administrative distance]
• The administrative distance is a
192.168.3.1 number between 0 and 255 that
represents the trustworthiness of the
routing information source
192.168.5.1
192.168.5.2
Router (Config)# ip route 192.168.120.0 255.255.255.0
192.168.3.2
Router (Config)# ip route 192.168.130.0 255.255.255.0
192.168.130.0 192.168.5.2
Page 19
Module 9, Page 20
Static Route Example
Stub Network Stub Network
200.20.20.0 200.10.10.0
A B
B
192.168.1.2 192.168.1.1
Router (Config)# ip route 200.10.10.0 255.255.255.0 192.168.1.1

Router (Config)#
¾This is a unidirectional route

¾You must have a route configured in the opposite direction.
Page 20
Module 9, Page 21
Default Static Route
Stub Network
144.27.0.0
Other Networks 200.10.10.0
192.168.1.1 192.168.1.2
205.7.7.0
Router (Config)# ip route 0.0.0.0 0.0.0.0 192.168.1.1

Router (Config)#
¾This route allows the stub network to reach all known networks beyond router A
With an address and subnet mask of 0.0.0.0 0.0.0.0 in the ip route statement,
packets for any network not listed in the routing table will be sent to the next hop,
192.168.1.1.
Page 21
Module 9, Page 22
Verifying Routes
192.168.120.2
• Ping
192.168.120.1 Ping [destination address]
192.168.3.2
• Traceroute
Traceroute [destination address]
192.168.3.1
192.168.5.1
Router# ping 192.168.120.2
192.168.5.2 Type escape sequence to abort
Sending 5, 100-byte ICMP Echos to 192.168.120.2, timeout is
2 seconds:
192.168.130.0 !!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max =
34/45/77
The ping command sends an ICMP echo request to the specified destination
address. This gives us information on reachability (out of five attempts) as
minimum, maximum and average round trip times.
Traceroute is a utility that records the route taken by packets from the BSR64000
and a specified destination host. It also calculates and displays the amount of time
each hop took. The original traceroute was a Unix utility first included in early
BSD distributions.
When you enter the traceroute command, the utility initiates the sending of a
packet (using the Internet Control Message Protocol or ICMP), including in the
packet a time limit value (known as the "time to live" (TTL) that is designed to be
exceeded by the first router that receives it, which will return a Time Exceeded
message. This enables traceroute to determine the time required for the hop to
the first router.
Increasing the time limit value, it resends the packet so that it will reach the
second router in the path to the destination, which returns another Time Exceeded
message, and so forth. Traceroute determines when the packet has reached the
destination by including a port number that is outside the normal range. When it's
received, a Port Unreachable message is returned, enabling traceroute to
measure the time length of the final hop. As the tracerouting progresses, the
records are displayed for you hop by hop.
Page 22
Module 9, Page 23
Where do the Routes Go?
•Routes stored in Routing Table

•The Routing Table displays the best routes
•Determined from entries in a database
•A Routing Table contains:
•Metrics
•Hop count/Cost/Distance
•Dependant on routing protocol
•Directly Connected Networks
•Static Routes
•Routes to remote networks
•Dynamic Entries learned from a routing protocol
Page 23
Module 9, Page 24
How do I look at the Routing

Table?
¾ Show the IP Routing Table
show ip route [connected or OSPF or RIP or Static, or summary]
» Connected shows directly connected networks only (interfaces)
» OSPF or RIP will show only OSPF or RIP learned routes
» Summary will list the number of different type routes in the table (1 RIP,
1 Static, 2 directly connected)
Codes: C - connected, S - static, R - RIP, B - BGP

O - OSPF, IA - OSPF inter area, N1 - OSPF NSSA external type 1
N2 - OSPF NSSA external type 2, E1 - OSPF external type 1
E2 - OSPF external type 2, * - candidate default
Gateway of last resort is not set

C 192.168.120.0/24 is directly connected, Ethernet 0/0
C 192.168.130.0/24 is directly connected, pos 5/0
S 192.168.120.0/24 [1/0] via 192.168.130.10, Ethernet 0/0
R 192.168.130.0/24 [120/3] via 192.168.130.10, Ethernet 0/1
Page 24
Module 9, Page 25
Static Routing Summary
• Static routing involves

manually adding routes in
each router’s routing tables
• Advantages
• This reduces processing
overhead in routers
• Reduces bandwidth used
between routers (no dynamic
exchanges)
• Disadvantages
• Time-consuming
• Addition of routers means
updates to all existing router’s
routing tables
• If a static path breaks, traffic
will not reach the destination
Page 25
Module 10, Page 0
Module 10
Routing Information Protocol (RIP)
Rev. 1.0 11/01
Page 0
Module 10, Page 1
Introduction
Introduction In a previous module, we explored static

routes and the steps necessary to configure
them. In this lesson will will explore
dynamic routing using the RIP routing
protocol.
Importance You will need a basic understanding of
routing in order to connect networked
devices. This is what will allow you to
enable both basic and advanced network-
based services.
Lesson Overview In this module, we will explore general
routing protocol mechanisms, and basic
operation. We concentrate on RIP, which is
a distance vector routing protocol. We will
discuss RIP operation, and how to
implement and monitor for proper RIP
operation.
Page 1
Module 10, Page 2
Objectives

ÔExplain basic dynamic routing concepts
ÔExplain the purpose of an Interior Gateway
Routing Protocol
ÔExplain the basic operation of RIP
ÔExplain the impact of split horizon when routing
ÔImplement a basic configuration using RIP
¾Start RIP process
¾Specify networks to advertise
¾Optionally redistribute external routes
¾Optionally specify RIP version
¾Optionally specify passive interface(s)
ÔMonitor operation of RIP
Page 2
Module 10, Page 3
Topic Order
Topic Order What is a routing Protocol? Page 5
Interior/Exterior Routing Protocols Page 6
Interior Routing Protocol Types Page 8
Distance Vector Routing Protocols Page 9
Starting the RIP Process Page 11
Specifying Networks to Advertise Page 12
Discovering Routes with RIP Page 13
Maintaining Routing Information Page 16
Administrative Distance:
Ranking Routes Page 19
Examining Routing Information Page 20
Examine the RIP Database Page 21
Examine the Routing Table Page 22
Routing Loops Page 23
Page 3
Module 10, Page 4
Topic Order (cont.)

Topic Order Symptom: Counting to Infinity Page 24
Solution: Defining a Maximum Page 28
Solution: Split Horizon Page 29
Configuring Split-Horizon Page 30
Route Poisoning Page 31
Poison Reverse Page 32
Version 1 and 2 Differences Page 33
Specifying RIP Version Page 34
Configuring a Passive Interface Page 35
Route Redistribution Page 36
Debugging RIP Page 37
Stopping the RIP process Page 38
RIP Configuration Example Page 39
RIP Limitations Page 40
Page 4
Module 10, Page 5
What is a Routing Protocol?
10.120.2.0
200.20.20.0
Ô Routing protocols are used between

routers to determine paths and maintain
routing tables.
Ô Once the path is determined a router Network Destination Exit 200.10.10.0
can route a routed protocol. Protocol Network Interface
Connected 10.120.2.0 E0
RIP 200.20.20.0 P3
RIP 200.10.10.0 P5
Routing protocol: RIP, or OSPF
Routed Protocol: IP
If network 10.120.2.0 wants to know about network 200.20.20.0, it must learn it

from its P3 (or possibly P5) interface.
Page 5
Module 10, Page 6
Interior/Exterior Gateway
Protocols
• Two applications of routing protocols

• Interior Gateway Protocols (IGP)
• Used to exchange routing information within the same Autonomous Systems
(AS).
• AS is a collection of networks under a common administrative domain
• RIP and Open Shortest Path First (OSPF) are examples of IGPs
• Exterior Gateway Protocols (EGP)
• Used to exchange routing information between Autonomous Systems
• This includes Border Gateway Protocol (BGP)
Page 6
Module 10, Page 7
Interior/Exterior Gateway
Protocols (cont.)
IGPs: RIP, OSPF EGPs: BGP
Autonomous System 100 Autonomous System 200
¾ An autonomous system is a collection of networks under a

common administrative domain
¾ IGPs operate within an autonomous system
¾ EGPs connect different autonomous systems
•Interior gateway routing protocols are used within a single autonomous system
The design criteria for an interior gateway routing protocol require it to find
the best path through the network. In other words, the metric and how that
metric is used is the most important element in an interior routing
protocol.
• Exterior routing protocols are used to communicate between autonomous
systems.
Exterior gateway protocols are used to exchange routing information
between networks that do not share a common administration. IP exterior
gateway protocols require the following three sets of information before
routing can begin:
A list of neighbor (or peer) routers or access servers with which to
exchange routing information
A list of networks to advertise as directly reachable
The autonomous system number of the local router
Page 7
Module 10, Page 8
Interior Routing Protocol Types
• Two types of IGP routing protocols

• Distance Vector (Based on the Bellman-Ford Algorithm)
• RIP is a Distance Vector Protocol
• Use distances to remote networks to judge the best path (Max 15 hops)
• By default, each router passed through is called a hop
• The router paths with the least amount of hops are considered as the best
• A hop count of 16 means unreachable
• Link State
• OSPF is a Link State Protocol
• Called shortest path first
• Tracks directly attached networks, and knows topology of the internetwork or
Area. Path selection is based on costs, rather than hops
• This means that link state protocols know more about the internetwork
than distance-vector protocols
Page 8
Module 10, Page 9
Distance Vector Routing Protocols
Distance—How
Distance—How farfar
Vector—In
Vector—In which
which direction
direction
Routing
Routing Routing
Routing Routing
Routing Routing
Routing
Table
Table Table
Table Table
Table Table
Table
ÔBased on the Bellman-Ford Algorithm

ÔPasses periodic copies of routing table to neighbor
routers and accumulates distance vectors (every 30 seconds)
Distance vector algorithms do not allow a router to know the exact topology of an
internetwork.
Routers running a Distance Vector Protocol know where the update came from
(what interface), and the number of hops associated with the advertised networks.
This information is somewhat like the information found on signs at a highway
intersection. A sign points toward a road leading away from the intersection and
indicates the distance to the destination. Further down the highway, another sign
also points toward the destination, but now the distance to the destination is
shorter. So long as each successive point on the path shows that the distance to
the destination is successively shorter, the traffic is following the best path.
Page 9
Module 10, Page 10

¾ Start RIP Process
¾ Specify Networks to Advertise
ÔOptional configuration tasks:

¾ Specify transmit/receive version
¾ Redistribute routes
¾ Remove split horizon
¾ Specify passive interface(s)
Page 10
Module 10, Page 11
Starting the RIP process
ÔIn global configuration EXEC

[no] router rip
¾ RIP task started
¾ By default RIP runs version 2
» Multicast updates to 224.0.0.9
RDN> en
RDN# config
RDN(config)# router rip
RDN(config-rip)#
Page 11
Module 10, Page 12
Specify the Networks to Advertise
ÔIn Router RIP configuration EXEC

[no] network <network mask>
¾ If a mask is not specified, then it will use classfull assumptions
RDN> en
RDN# config
RDN(config)# router rip
RDN(config-rip)#network 192.168.25.0 255.255.255.0
RDN(config-rip)#exit
RDN(config)#exit
RDN#
Page 12
Module 10, Page 13
Discovering Routes with RIP
Router A Router B Router C

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2 E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 10.3.0.0 P2 0 10.4.0.0 E0 0
ÔRouters discover the best path to destinations from each neighbor

ÔRIP information is broadcast every 30 seconds. These updates are not
synchronized
This figure continues the concept of how a router using a distance vector protocol
generally discovers the best path to destinations from each router neighbor.
Routing tables inside each router begin with entries for the 0 distance to directly
connected networks.
Page 13
Module 10, Page 14

(cont.)

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2 E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 0
10.3.0.0 10.4.0.0
10.1.0.0 P4 1
10.4.0.0 P2 1

synchronized
Here, router B learns about networks 10.1.0.0 from router A, and 10.4.0.0 from
router C, via RIP updates.
Page 14
Module 10, Page 15

(cont.)

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2 E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 0
10.3.0.0 10.4.0.0
10.3.0.0 P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
P3 2 10.4.0.0 P2 1 10.1.0.0 P2 2
10.4.0.0

synchronized
Here, router B advertises the routes it learned from routers C and A. Actually, all
routes are advertised, but the new entries are added to the routing table when
received.
Page 15
Module 10, Page 16
Maintaining Routing Information
Process
Process toto
update
update this
this
routing
routing
table
table
Topology
change
causes
routing
table
update
ÔUpdates proceed step-by-step

from router to router
The next couple of slides show how a router using a distance vector protocol
generally performs its routing information update process when the network
topology changes.
Page 16
Module 10, Page 17

(cont.)
Process
Process toto
update
update this
this
routing
routing
table
table
Router A sends Topology
out this updated change
routing table causes
after the routing
next period table
expires Router A update

Page 17
Module 10, Page 18

(cont.)
Process
Process toto Process
Process toto
update
update this
this update
update this
this
routing
routing routing
routing
table
table table
table
Router A sends Topology

out this updated change
routing table causes
after the routing
next period table
Router B expires Router A
update

Distance vector updates occur step by step. Typically, a router sends updates by
multicasting its table on each configured port, but other methods, such as sending
the table only to reconfigured neighbors, are employed by some routing
algorithms. Multicast is used by the RIP2, and OSPF. Broadcasts are used by
RIP version 1. The routing table can be sent routinely and periodically, or
whenever a change in the topology is discovered. Updates sent when changes
occur are called triggered updates.
Page 18
Module 10, Page 19
Administrative Distance:
Ranking Routes
OSPF Administrative
Distance=110
Router B
I need to send a packet to
Network F. Both router B
and D will get it there. Router C
Which route is best? Router A
RIP
Administrative
Distance=120
Network F
Router D Router E
An administrative distance is a rating of the trustworthiness of a routing

information source, such as an individual router or a group of routers. In a large
network, some routing protocols and some routers can be more reliable than
others as sources of routing information.
For instance, the default administrative distance for static routes 1. If a static route
existed to network E, then that route would be used. The lower the distance, the
more trustworthy the route is. For example, in the figure, the router would use the
route it learned via OSPF, instead of the RIP learned route.
•Connected interfaces have a default administrative distance of 0
•Static routes have a default administrative distance of 1
•OSPF routes have a default administrative distance of 110
•RIP routes have a default administrative distance of 120
Page 19
Module 10, Page 20
Examining Routing Information
ÔThere are two ways to look at RIP information:
¾ Examine the RIP database

» Directly connected interfaces participating in RIP
» All learned RIP routes
¾ Examine the routing table

» Directly connected networks
» Static routes
» Best dynamic routes
The routing table will show all directly connected networks, static routes and only
the best dynamic routes. The RIP database, on the other hand, will have
information about all directly connected interfaces participating in RIP as well as
all RIP learned routes.
Page 20
Module 10, Page 21
Examine the RIP Database
ÔIn Privilege EXEC

show ip rip database
RDN> en
RDN# sh ip rip database
192.168.24.0 255.255.255.0 directly connected 192.168.24.20 m:1 t:0
192.168.45.0 255.255.255.0 via 192.168.24.40 m:2 t:18
192.168.45.0 255.255.255.0 via 192.168.10.40 m:2 t:18
192.168.34.0 255.255.255.0 via 192.168.24.40 m:2 t:18
192.168.35.0 255.255.255.0 via 192.168.10.30 m:2 t:14
200.40.40.0 255.255.255.0 via 192.168.10.40 m:2 t:18
Showing the RIP database shows information on all directly connected and RIP
learned routes only. It shows metric (m:2) which is hop count, and time in
seconds (t) that the dynamic route was updated. This database is also known as
the routing information base (RIB). It is this database that is consulted by the
routing table manager (RTM) to decide which routes are actually put in the
Routing table, also known as the forwarding information base (FIB).
Page 21
Module 10, Page 22
Examine the Routing Table
ÔIn Privilege EXEC

show ip route [connected, or static,or RIP,or OSPF]
RDN> en
RDN# sh ip route
i - IS-IS derived L1-IS-IS level-1 route, L2- IS-IS level 2 route
C 192.168.10.0/24 is directly connected, ethernet 7/0
R 192.168.12.0/24 [120/2] via 192.168.10.10, ethernet 7/0
C 200.20.20.0/24 is directly connected, loopback 1
R 200.30.30.0/24 [120/2] via 192.168.10.30, ethernet 7/0
R 200.40.40.0/24 [120/2] via 192.168.10.40, ethernet 7/0
Showing the routing table gives you information about how forwarding decisions
are being made. This is the table that is used to route packets. In contrast, the
RIB is used to store information about all RIP learned routes, whether best route
or not. The reason for this, is a network or interface failure. By having other
routes in the database, a substitute route can be moved to the routing table, until a
better route appears in the RIB.
Page 22
Module 10, Page 23
Routing Loops

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2 E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 0
10.3.0.0 10.4.0.0
10.3.0.0 P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
P3 2 10.4.0.0 P2 1 10.1.0.0 P2 2
10.4.0.0
ÔEach node maintains the distance from itself to each possible destination
network
The next few slides describe one of the general problems that distance vector
protocols could face without the corrective influence of some countermeasure,
such as split-horizon.
This slide shows the original state of the network and routing tables. All routers
have consistent knowledge and correct routing tables. In this example, the routing
metric is hop count, so the cost of each link is 1. Router C is directly connected to
network 10.4.0.0 with a distance of 0. Router A’s path to network 10.4.0.0 is
through router B, with a hop count of 2.
Page 23
Module 10, Page 24
Symptom: Counting to Infinity

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 Down
10.3.0.0 10.4.0.0
10.3.0.0 P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
P3 2 10.4.0.0 P2 1 10.1.0.0 P2 2
10.4.0.0
Ô Slow convergence produces inconsistent routing
Router C has detected the failure of network 10.4.0.0 and stops routing packets
out its E0 interface. However, router A has not yet received notification of the
failure and still believes it can access network 10.4.0.0 through router B. Router
A’s routing table still reflects a path to network 10.4.0.0 with a distance of 2.
Page 24
Module 10, Page 25

(cont.)

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 2
10.3.0.0 10.4.0.0
10.3.0.0 P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
P3 2 10.4.0.0 P2 1 10.1.0.0 P2 2
10.4.0.0
Router C concludes that the best path to network

10.4.0.0 is through Router B
Because router B’s routing table indicates a path to network 10.4.0.0, router C
believes it now has a viable path to 10.4.0.0 through router B. Router C updates
its routing table to reflect a path to network 10.4.0.0 with a hop count of 2.
Page 25
Module 10, Page 26

(cont.)

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 2
10.3.0.0 10.4.0.0
P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
10.3.0.0
P3 4 10.4.0.0 P2 3 10.1.0.0 P2 2
10.4.0.0
Router A updates its table to reflect the new but

erroneous hop count
Router A receives the new routing table from router B, detects the modified
distance vector to network 10.4.0.0, and recalculates its own distance vector to
network 10.4.0.0 as 3.
If all routers in an internetwork do not have up-to-date, accurate information about
the state of the internetwork, they might use incorrect routing information to make
a routing decision.
The use of incorrect information might cause packets to take less-than-optimum

paths or paths that return packets to routers that they have already visited.
Page 26
Module 10, Page 27

(cont.)

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 4
10.3.0.0 10.4.0.0
P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
10.3.0.0
P3 6 10.4.0.0 P2 5 10.1.0.0 P2 2
10.4.0.0
Packets for network 10.4.0.0 bounce between routers A, B, and C

Hop count for network 10.4.0.0 counts to infinity
Both routers conclude that the best path to network 10.4.0.0 is through each other
and continue to bounce packets destined for network 10.4.0.0 between each
other, incrementing the distance vector by 1 each time. This condition, called
count-to-infinity, continuously loops packets around the network, despite the
fundamental fact that the destination network 10.4.0.0 is down. While the routers
are counting to infinity, the invalid information allows a routing loop to exist.
A related concept is the Time-to-Live (TTL) parameter. The TTL is a packet

parameter that decreases each time a router processes the packet. When the TTL
reaches zero, a router discards or drops the packet without forwarding it. A packet
caught in a routing loop is removed from the internetwork when its TTL expires.
Page 27
Module 10, Page 28
Solution: Defining a Maximum

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 16
10.3.0.0 10.4.0.0
10.3.0.0 P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
P3 16 10.4.0.0 P2 16 10.1.0.0 P2 2
10.4.0.0
ÔDefining a limit on the number of hops prevents infinite loops
Routing loops occur only when routing knowledge being propagated has not yet
reached the entire internetwork—when the internetwork has not converged after a
change. Fast convergence minimizes the chance for a routing loop to occur, but
even the smallest interval leaves the possibility open.
To avoid prolonging the count-to-infinity time span, distance vector protocols

define infinity as some maximum number. This number refers to a routing metric,
such as a hop count.
With this approach, the routing protocol permits the routing loop until the metric
exceeds its maximum allowed value. This example shows this defined maximum
as 16 hops. Once the metric value exceeds the maximum, network 10.4.0.0 is
considered unreachable.
Page 28
Module 10, Page 29
Solution: Split Horizon

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
X X
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 16
10.3.0.0 10.4.0.0
10.3.0.0 P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
P3 3 10.4.0.0 P2 2 10.1.0.0 P2 2
10.4.0.0
ÔIt is never useful to send information about a route back in

the direction from which the original route was learned.
The split horizon technique attempts to eliminate routing loops and speed up
convergence. The rule of split horizon is that it is never useful to send information
about a route back in the direction from which the original route was learned. In
the example:
Router C originally announced a route to network 10.4.0.0 to router B. It
makes no sense for router B to announce to router C that router B has
access to network 10.4.0.0 through router C.
Given that router B passed the announcement of its route to network

10.4.0.0 to router A, it makes no sense for router A to announce its
distance from network 10.4.0.0 to router B.
Because router B has no alternative path to network 10.4.0.0, router B

concludes that network 10.4.0.0 is inaccessible.
In its basic form, the split horizon technique simply omits from the message any
information about destinations routed on the link. This strategy relies either on
routes never being announced or on old announcements fading away through a
timeout mechanism.
Split horizon also improves performance by eliminating unnecessary routing

updates. Under normal circumstances, sending routing information back to the
source of the information is unnecessary.
Page 29
Module 10, Page 30
Configuring Split-Horizon
ÔConfigure split-horizon in interface mode

[no] ip split-horizon
» Default is on
» Must be disabled on CMTS interfaces in order to allow routing
across them
• CMTS interfaces are treated like local area (shared media) interfaces
RDN> en
RDN# conf
RDN (config)# int c 3/0
RDN (config-if)# no ip split-horizon
Page 30
Module 10, Page 31
Route Poisoning

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
16
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 Infinity
10.3.0.0 10.4.0.0
P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
10.3.0.0
P3 3 10.4.0.0 P2 1 10.1.0.0 P2 2
10.4.0.0
To speed convergence, another form of split-horizon can be used. Route

poisoning attempts to eliminate routing loops caused by inconsistent updates. It
involves setting the value of the downed directly connected network to infinity ,
and using a holddown timer so that updates from other sources (router B in this
case) will not be believed until the holddown timer expires.
Page 31
Module 10, Page 32
Poison Reverse

10.1.0.0 10.2.0.0 10.3.0.0 10.4.0.0
E0 P3 P4 P2 P2
X
E0
16
Routing
Routing Table
Table Routing
Routing Table
Table Routing
Routing Table
Table
10.1.0.0 E0 0 10.2.0.0 P4 0 10.3.0.0 P2 0
10.2.0.0 P3 0 P2 0 E0 Infinity
10.3.0.0 10.4.0.0
P3 1 10.1.0.0 P4 1 10.2.0.0 P2 1
10.3.0.0
P3 3 10.4.0.0 P2 Down? 10.1.0.0 P2 2
10.4.0.0
When router B sees the metric for network 10.4.0.0 jump to infinity, it sends an
update called a poison reverse back to router C stating that network 10.4.0.0 is
unreachable. This update overrides split-horizon to make sure all routers on that
segment (in this case only router C) have received information about the poisoned
route.
Page 32
Module 10, Page 33
Version 1 and 2 Differences
Ô Version 1 Version 2
Ô Packets sent broadcast Packets sent multicast
Ô No authentication mechanisms Next hop information is stored
Ô Class based routes, no subnet
mask field Simple text Authentication
Supports CIDR (classless inter-
domain routing)
There are some distinct differences between RIP versions one and two. As such,
it is important to check what version any participating router is running in order
ensure that routing updates will be exchanged successfully.
Page 33
Module 10, Page 34
Specifying RIP Version

version [0-3]
RDN (config-rip)#version 1
Ô In Interface configuration EXEC

[no] ip rip send |receive version [0-3]
RDN (config-if)#ip send version 1

RDN (config-if)#ip receive version 1
0=version 1&2, 1=version 1 only, 2=version 2 only, 3=don’t send
Interface mode RIP version commands takes precedence over global version
commands:
config-if# [no] ip rip send version [0-3] :
0 version 2 compatible 1
1 version 1
2 version 2
3 don’t send
config-if# [no] ip rip receive version [0-2] :
0 versions 2 and 1
1 version 1
2 version 2
Page 34
Module 10, Page 35
Configuring a Passive Interface
ÔPassive interfaces receive updates, but don’t send

them
¾Commonly used in conjunction with route redistribution
¾ In RIP configuration EXEC
passive interface <interface identifier>
RDN (config-rip)#passive interface e 3/0
ÔIn Interface configuration EXEC

[no] ip rip send version 3
RDN (config-if)#ip receive version 3
Page 35
Module 10, Page 36
Route Redistribution
ÔAllows us to advertise externally originating routes in RIP

¾ From OSPF, connected, static or OSPF

redistribute [connected | static | ospf ]
RDN (config-rip)#redistribute static
Page 36
Module 10, Page 37
Debugging RIP
Ô[no] debug ip rip

¾ Enables debug all of RIP’s debug output
Ô[no] debug ip rip database
¾ Enables debug statements that affect RIP’s routing tables, route
additions, deletions, changes in attributes
Ô[no] debug ip rip events
¾ Enables debug statements that affects RIP routing process such
as detail descriptions of route updates for incoming and outgoing
packets
Ô[no] debug ip rip trigger
¾ Enables debug statements that affect RIP triggered and periodic
updates
Page 37
Module 10, Page 38
Stopping the RIP process
ÔIn global configuration EXEC

no router rip
¾ RIP task stopped
¾ Removes all RIP-related configurations
» Including Network Statements
RDN> en
RDN# config
RDN(config)# no router rip
RDN(config-rip)#
Page 38
Module 10, Page 39
RIP Configuration Example
E0 P3 P3 P5 P5 E0
129.168.1.0 20.10.10.0
129.168.1.1 10.1.1.1 10.1.1.2 10.2.2.2 10.2.2.3 20.10.10.1
router rip 2.3.0.0 router rip 2.3.0.0

network 129.168.0.0 network 20.10.10.0
network 10.0.0.0 network 10.0.0.0
router rip
network 10.0.0.0
An administrator only specifies directly connected networks that he wishes to

publish to other routers. Without the network command, nothing is advertised.
With a network command, the router will advertise every subnet within the Class
A, B, or C network specified in the configuration.
Page 39
Module 10, Page 40
RIP Limitations
ÔRIP is primarily intended for use in networks of moderate size

¾ Less than 16 hops end to end
ÔBecause of this, RIP has the following limitations:
¾ AS diameter limitation of 15 hops (in most implementations)
¾ Slow convergence
ÔBest application: Networks of moderate size
Although there are mechanisms to minimize routing loops ( Split-Horizon for

example), loops can still occur causing slow convergence. When loops occur, a
behavior called “ Count to infinity” slows convergence.
Page 40
Module 11, Page 0
Module 11
Open Shortest Path First (OSPF)

Configuration
Rev. 1.0 11/01
Page 0
Module 11, Page 1
Introduction
Introduction In the last Module, you learned about the

RIP protocol. The BSR supports another
IGP (Interior Gateway Protocol) called
OSPF. It’s function, like RIPs, is to aid in
path determination for routed protocols.
Importance OSPF is widely used in large networks as
the routing protocol of choice. Although it is
more complex to architect and implement
than RIP, it pays dividends in lower
bandwidth overhead and greater control.
Lesson Overview We will examine how to implement single-
area and multi-area OSPF networks using
the BSR. As such, we will take a look at
how the underlying protocol works to
determine optimal routes and to share that
information with other hosts within the same
administrative domain.
Page 1
Module 11, Page 2
Objectives
Objectives Upon successful completion of this module you will

ÔExplain why OSPF is better than RIP in a large
environment
ÔExplain how OSPF discovers, chooses, and
maintains routes
ÔConfigure OSPF for proper operation according to
a specification
ÔMonitor proper OSPF operation
Page 2
Module 11, Page 3
Topic Order
Topic Order What is OSPF? Page 5
OSPF Routing Hierarchy Page 6
Starting the OSPF process Page 8
What does the OSPF Process do? Page 9
OSPF Initialization Page 10
OSPF Network Types Page 11
Examining Adjacencies Page 14
Neighbor States Page 15
Defining Interfaces in OSPF Page 17
Using Wildcard Masks Page 18
OSPF Areas Page 19
Multi-Area OSPF Config Page 20
OSPF Router Types Page 21
Types of Link State Advertisements Page 22
Examining Routing Info Page 23
Page 3
Module 11, Page 4
Topic Order (cont.)

Topic Order Examine the OSPF Database Page 24
Examine the Routing Table Page 25
Route Redistribution Page 26
Force Router ID Choice Page 27
Stopping the OSPF Process Page 28
Page 4
Module 11, Page 5
What is OSPF ?
ÔOSPF is an Interior Gateway Routing Protocol

Ô Link-State Routing Protocol
¾ Designed for IP networks
» Protocol 89
Ô Originally developed by IETF IGP Workgroup
¾ Updated by the OSPF Workgroup
Ô Has advantages over RIP
¾ Fast convergence
¾ Supports VLSM (like RIP II) as well as CIDR
¾ Has no hop count limitation
» Selects paths based on costs
» Information shared about the link states and network state within
the internetwork.
» It can utilize a routing hierarchy
Open Shortest Path First (OSPF) is a routing protocol developed for use with IP
networks by the interior gateway protocol (IGP) working group of the Internet
Engineering Task Force (IETF).
The working group was formed in 1988 to design an IGP based on the Dijkstra
(sometimes called shortest path first [SPF]) algorithm for use in the Internet. Just
like the Interior Gateway Routing Protocol (IGRP), OSPF was created because, in
the mid-1980s, as networks grew, the Routing Information Protocol (RIP) was just
unable to serve their needs (especially in heterogeneous environments).
Version one published in RFC 1131, updated in RFC 1247 (version two), which
was deprecated with RFC 1583 (still OSPF version 2) and RFC 2328, April 1998.
The BSR64000 supports OSPF Version 2, RFC 2328 including the use of area
border routers (ABR), stub areas, Virtual Links and Transit areas, Autonomous
System border router (ASBR), Type 1 and 2 external routes, simple and MD5
authentication.
Also supported is the OSPF MIB (RFC 1850 - Nov 1995) and not-so-stubby areas
(NSSAs); RFC 1587, March 1994.
Page 5
Module 11, Page 6
OSPF Routing Hierarchy
Autonomous System (AS)
Area 0
Area 1 Area 2
OSPFs’ ability to separate a large inter-network into multiple areas is referred to

as hierarchical routing. Hierarchical routing allows separation of large
internetworks (Autonomous systems) into smaller inter-networks called areas.
With this technique, routing still occurs between areas (intra-area routing), but
minute internal routing operations such as re-calculating the database are kept
within the area.
Page 6
Module 11, Page 7

¾ Start OSPF Process
¾ Define interfaces on which OSPF runs
» Assign an interface to an area
ÔOptional basic configuration tasks:

¾ Force Router I.D. Choice Using Loopback
Address
¾ Segregate Routers Into Separate Areas
¾ Redistribute External Routes
Basic configuration tasks include starting the OSPF routing process on your
BSR64000, specifying which interfaces will be included and grouping those
interfaces into areas.
It is possible to include externally discovered (i.e., RIP, BGP or static) routes in
OSPF, as well as to segregate routers into administrative areas. These two,
along with the forcing of the Router I.D. by using a Loopback Address are optional
implementation tasks that you may or may not do, based on your network
environment.
Page 7
Module 11, Page 8
Starting the OSPF Process
ÔFrom Global Configuration EXEC

[no] router ospf
» Starts the OSPF process
RDN> en
RDN# config
RDN(config)# router ospf
RDN(config-ospf)#
…
The router ospf command starts the OSPF routing process on the BSR64000.
This means that the routing daemon looks for indicators as to which networks to
advertise. If no networks (interfaces) are specified, then the router will not
advertise or listen for those advertisements.
Page 8
Module 11, Page 9
What does the OSPF Process do?
ÔEach router maintains an identical topological database

¾ Link State Database (LSDB)
» Built based on Link State Advertisements (LSA)
ÔEach router constructs a tree of shortest paths
¾ By running shortest path first (SPF) algorithm on the database
» Tree provides route to each known destination
¾ OSPF uses cost as a metric, not hops
A topological database is an overall picture of networks in relationship to routers.

The topological database contains a collection of advertised networks received
from all routers in the same area. Because routers within the same area share
the same information, they have identical topological databases.
The shortest path first (SPF) routing algorithm is the basis for OSPF operations.
When an SPF router is powered up, it initializes its routing-protocol data
structures and then waits for indications from lower-layer protocols that its
interfaces are functional.
When no OSPF areas are configured, each router in the Autonomous System has
an identical topological database, leading to an identical graphical representation.
A router generates its routing table from this graph by calculating a tree of shortest
paths with the router itself as root. The shortest-path tree depends on the router
doing the calculation.
The tree gives the entire route to any destination network or host. However, only
the next hop to the destination is used in the forwarding process. Note also that
the best route to any router has also been calculated. For the processing of
external data, we note the next hop and distance to any router advertising external
routes.
A cost is associated with the output side of each router interface. This cost is
configurable. The lower the cost, the more likely the interface is to be used to
forward data traffic.
Page 9
Module 11, Page 10
OSPF Initialization
ÔOSPF routers exchange Router-IDs

¾ Done with Hello Protocol
» Packet type 1
» Send to multicast address of 224.0.0.5
• All SPF routers
» Send periodically on all OSPF configured
ÔWhat happens next, network-type dependant

¾ Broadcast
¾ Point-to-point
¾ Non-broadcast multiple-access (NBMA)
After a router is assured that its interfaces are functioning, it uses the OSPF Hello
protocol to acquire neighbors, which are routers with interfaces to a common
network. The Hello Protocol is responsible for establishing and maintaining
neighbor relationships. It also ensures that communication between neighbors is
bi-directional. The router sends hello packets to its neighbors and receives their
hello packets. Bi-directional communication is indicated when the router sees
itself listed in the neighbor's Hello Packet. In addition to helping acquire
neighbors, hello packets also act as keep-alives to let routers know that other
routers are still functional.
The next step is different on broadcast networks, as compared to non-broadcast
networks, which we will see in a moment.
Page 10
Module 11, Page 11
OSPF Network Types
ÔBroadcast Networks - make use of IP multicast.

¾ Designated Router Election
» Designated Router (DR)
• Router generates Network LSA changes
¾ Backup Designated Router (BDR)
¾ DR-Other
Priority 1 Priority 4
200.10.10.1 200.10.10.2
Hello? Hello? Hello?

Hello?
Hello?
200.10.10.3 200.10.10.4
Priority 3 Priority 2
In an OSPF broadcast network, the mechanism used to control the size of the
Link state database and number of adjacencies is to elect a designated and
backup designated router. Two values are used , priority and router ID. The
highest Priority wins, and if there is a tie, then the highest router ID wins the
election. All routers on a broadcast network will form an adjacency with the
designated router. Routers discover neighbors through a protocol called the
“Hello” protocol. On a point to point link, you neighbor is at the other end, so there
is no need to have an election. After the election, all routers will have a full
adjacency to the designated router, but no to the other routers in the broadcast
network. The backup designated knows everything the designated router knows,
so if the designated router fails, the backup designated router can take over
immediately. The multicast addresses used are 224.0.0.5 (all SPF Routers) and
224.0.0.6 (All DRs and BDrs).
Page 11
Module 11, Page 12
OSPF Network Types (cont.)
ÔPoint-to-point links
¾ No Designated Router
¾ Router with highest Router-ID controls initial exchange-process
192.168.1.0
Hello?
.1
POS
.2
POS
Hello?
Point to point networks are seen by OSPF as “transit” networks, because there
are no nodes other than the two connected routers on the link. When you have
point-to-point networks in your OSPF domain, the will show up twice in your IP
routing table. Once as a normal network with the corresponding mask, and once
as a “host route”. The host route will show up in the table with a mask of /32. An
example would be:
• 192.168.1.0/24 Network
• 192.168.1.2/32 Host Route
Page 12
Module 11, Page 13
OSPF Network Types (cont.)
ÔNonBroadcast Multi-Access (NBMA), point-to-

Multipoint, Networks
¾ Such as ATM VC, FrameRelay, …
¾ Not Supported by the BSR today
Page 13
Module 11, Page 14
Examining Adjacencies
From Priveleged EXEC

show ip ospf neighbor [<neighbor-id>] [detail]
RDN> en
RDN# show ip ospf neighbor
…
Page 14
Module 11, Page 15
Neighbor States
¾ Down
» No information has been received from neighbor(s)
¾ Attempt
» Effort is made to contact the neighbor
• No information has yet been received from the neighbor
• Valid for neighbors on NBMA network
¾ Init
» Hello packet has been received form the neighbor
• Receiving router is not listed in that hello packet
¾ 2Way
» Bi-directional communication with the neighbor is established
» Beginning of adjacency
• DR/BDR are elected in this state
There are multiple steps towards building adjacencies. Fully adjacent routers will
have the same link state database.
Routers within non-broadcast multi-access networks, such as Frame Relay and
X.25 will reflect the Attempt State when no recent information has been received
from a neighbor. An effort will be made to contact the neighbor by sending Hello
packets at a reduced rate.
Page 15
Module 11, Page 16
Neighbor States (cont.)
¾ Exstart
» Used in deciding the DBD sync, which router will be
master/slave and what will be the first sequence number for
DBD packet
¾ Exchange
» In this state the router describes the entire link state database
through the DBD packet, each DBD sequence is explicitly
acked, only one DBD packet is allowed outstanding at one
time, link state request packets are also sent to request the
newer LSA
¾ Loading
» In this state, link-state request packets are requested for the
more recent LSA that have not been received during Exchange
¾ Full
» In this state complete information has been exchanged
Page 16
Module 11, Page 17
Defining Interfaces in OSPF
ÔFrom Router OSPF Configuration EXEC

network <ip address> <wildcard mask> area <area-id>
» Specifies networks to advertise
RDN> en
RDN# config
RDN(conifg-ospf)# network 192.168.50.0 0.0.0.255 area 0
RDN(config-ospf)# network 200.20.20.1 0.0.0.0 area 0
…
Page 17
Module 11, Page 18
Using Wildcard Masks
ÔCan be used to reduce number of network statements

¾ 0.0.0.0 means match all 32 bits exactly
¾ 255.255.255.255 means anything matches
ÔExamples
¾ 192.168.32.3 0.0.0.0 area 0
» OSPF will advertise on this specific network interface and
place it in Area 0
¾ 192.168.32.0 0.0.0.255 Area 0
» Any interfaces that start with 192.168.32.x will belong to Area 0
Wildcard masks are used to cut down the number of network statements that are
entered after enabling OSPF routing. The Wildcard masks are also called inverse
masks, as they appear to work inverse to subnet masks. They aren’t hard. Just
remember that 1’s mean “I don’t care”, and 0’s mean “oh, this is important”.
Page 18
Module 11, Page 19
Force Router ID Choice
ÔFrom Interface Configuration EXEC

ip address <ip address> <network mask>
ÔFrom OSPF Configuration EXEC
network <loopback address> <wildcard mask> area <area-id>
RDN> en
RDN# conf
RDN(config)# int loopback 1
RDN(config-if)# ip address 200.20.20.1 255.255.255.0
…
RDN(router-ospf)# network 200.20.20.1 0.0.0.0 area 2

…
Page 19
Module 11, Page 20
OSPF Areas
Network 5
Network 3
Network 1 Router1
Router 3
Network 2 Router 2 Network 4
Area 1 Area 0 Area 2

Backbone Area
OSPF allows you to break up your inter-network into Areas. Remember, an area
is collection of routers and networks that have the same area ID. The purpose of
an area is to cut down the amount of routing information that a router has to keep
in its’ routing table. Areas allow for summarization, so that many known networks
might be advertised as a single route. Additionally, flooding of LSAs is contained
within the area and does not affect other areas which saves on bandwidth. The
end result is smaller routing tables in other areas. To take advantage of this
feature, a network administrator has to give some thought to the network
addressing in each area, so that the addresses lend themselves to the
summarization process.
Page 20
Module 11, Page 21
Multi-Area OSPF Configuration
From OSPF Configuration EXEC

[no] network <ip address> <wildcard mask> area <area-id>
• No different than single-area network statement
RDN> en
RDN# conf
RDN(config-ospf)# network 192.168.50.4 0.0.0.0 area 3
…
Page 21
Module 11, Page 22
OSPF Router Types
Area 0
Backbone
internal
ABR and Backbone ABR and Backbone
ASBR
Internal
Area 1 Area 2
External
Network
Internal routers have all their interfaces in the same area. Backbone routers that
sit on the perimeter of the backbone area. They have at least one interface
connected to area 0. Area border routers (ABRs) have interfaces connected to
multiple areas. ABRs are exit points for the area, which means routing
information destined for another area can only get there via the local areas ABR.
Autonomous System Border Routers (ASBRs) have at least one interface into an
external inter-network (autonomous system), such as a RIP based network.
These routers can import or export (called redistribution) non-OSPF network
information into the OSPF network, or advertise OSPF networks into the RIP
inter-network. As you can see from the diagram, a router can act as more than
one specific router type.
Page 22
Module 11, Page 23
Types of Link State Advertisements
Type 1 All Internal Routers
Type 2 Designated Routers
Type 3 Area Border routers
Type 4 ASBRs and ABRs
Type 5 ASBRs
Type 7 ABRs
A router has a separate LSDB for each area it belongs to. All routers belonging to
the same area have identical LSDBs. The LSDB is used by each router to
calculate which routes are entered into the routing table. Type 1 LSAs are
generated by each router for each area it belongs. They describe the states of the
routers links to the area, and are only flooded within an area.
Type 2 LSAs are generated by DRs (In a broadcast network), and describe the set
of routers attached to a particular network. They are flooded in the area that
contains the network.
Type 3 LSAs (called summary LSAs) are generated by Area Border Routers, and
describe inter-area routes to various networks (IA).
Type 4 LSAs (also considered summary LSAs) describe links to Autonomous
System Border (Boundary) routers and are also generated by Area Border
Routers. (IA)
Type 5 LSAs are generated by ASBRs. They describe routes to destinations to
the Autonomous System (Internetwork). They are flooded everywhere, with the
exception of Stub areas.
Type 7 LSAs are generated by ABRs. They describe routes within the Not So
Stubby Area. They can be summarized and converted to type 5 LSAs by the
ABRs. After conversion, they will be flooded to areas that can support type 5
LSAs.
Page 23
Module 11, Page 24
Examining Routing Information
ÔThere are Two ways to look at OSPF routing information:
¾ Examine the OSPF database

» Directly connected interfaces participating in OSPF
» All learned OSPF routes
¾ Examine the routing table

» Directly connected networks
» Static routes
» Best dynamic routes
Page 24
Module 11, Page 25
Examine the OSPF Database
ÔFrom Privilege EXEC

show ip ospf database
¾ Routers participating in multiple areas will have one database for
each area they belong to
RDN> en
RDN# sh ip ospf database
…
Showing the OSPF database shows information on all directly connected and
OSPF learned routes only. This database is also known as the routing
information base (RIB). It is this database that is consulted by the routing table
manager (RTM) to decide which routes are actually put in the Routing table, also
known as the forwarding information base (FIB).
Page 25
Module 11, Page 26
Examine the Routing Table
ÔIn Privilege EXEC

show ip route [connected, or static,or RIP or OSPF]
RDN> en
RDN# sh ip route
i - IS-IS derived L1-IS-IS level-1 route, L2- IS-IS level 2 route
C 192.168.10.0/24 is directly connected, ethernet 7/0
R 192.168.12.0/24 [120/2] via 192.168.10.10, ethernet 7/0
C 200.20.20.0/24 is directly connected, loopback 1
R 200.30.30.0/24 [120/2] via 192.168.10.30, ethernet 7/0
O 200.40.40.0/24 [110/2] via 192.168.10.40, pos 4/0
Showing the routing table gives you information about how forwarding decisions
are being made. This is the table that is used to route packets. In contrast, the
RIB is used to store information about all RIP learned routes, whether best route
or not. The reason for this, is a network or interface failure. By having other
routes in the database, a substitute route can be moved to the routing table, until a
better route appears in the RIB.
Page 26
Module 11, Page 27
Route Redistribution
ÔFrom OSPF Configuration EXEC

[no] redistribute [rip | static | bgp | connected]
ASBR Route Redistribution
» Supports tagging of external derived routing information.
» AS Boundary Router (ASBR) imports external routing
information into the OSPF routing domain.
» Type 1 and Type 2 metrics (E1,&E2)
RDN> en
RDN# conf
RDN(config-ospf)# redistribute rip
…
Type 1 metrics are comparable to the link state metrics. Link costs are added as
if it were an area route being advertised by the ABR.
Type 2 metrics are assumed to be larger than the cost of any intra-AS path.
When redistributing a rip route, you assign a metric value which is higher than any
Intra-AS path.
Page 27
Module 11, Page 28
Stopping the OSPF Process
ÔFrom Global Configuration EXEC

no router ospf
» Stops the OSPF process
» Removes all OSPF-related configurations
• Including Network Statements
RDN> en
RDN# config
RDN(config)# no router ospf
RDN(router)#
…
The router ospf command starts the OSPF routing process on the BSR64000.
This means that the routing daemon looks for indicators as to which networks to
advertise. If no networks (interfaces) are specified, then the router will not
advertise or listen for those advertisements.
Page 28

CMTS Motorola BSR PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CMTS Motorola BSR PDF

Uploaded by

Copyright:

Available Formats

Introduction, Page 0

Implementing the BSR for Basic Data Services

Implementing the BSR for Basic

Rev. 1.1 3/02

Introduction Implementing the BSR for Basic Data Services (IRI-

Course Overview Attendees will learn how to install and configure

Course Objectives Upon successful completion of this course, you will

Prerequisites To fully benefit from this course, you should already

Topic Order The Big Picture Module 1

Please enjoy the class; we certainly enjoy providing it.

How We Develop Our Classes

Implementing the BSR for Basic Data Services

The Big Picture

Rev. 1.1 11/01

Introduction Cable systems have developed devices and

Importance It is important to know what components make up the

Objectives Upon successful completion of this module, you will be

Topic Order The Big Picture Page 5

Topic Order (cont.)

Topic Order Amplifiers Page 19

The Big Picture

Traditional Headend Functions

Headends are the nerve center of CATV (Community Antenna Television)

Digital Broadcast 8-VSB

Headend system components provide for signal collection, signal

System Components (cont.)

Distribution System Access

Tree and Branch Architecture

Originally intended as a means to provide TV signals to remote areas, CATV

Fiber Backbone Architecture

Hybrid Fiber Coax Architecture

Originally intended as a means to provide TV signals to remote areas, CATV

DFB (Distributed Feedback) laser with 1310 nm operational wavelength using

Amplifiers are used to maintain unity gain in the distribution system.

Forward Signal Path

Return Signal Path

Customer Premises Equipment

ÔCommonly Referred to as CPE

Splitter HFC Plant

Set Top Box

From a cable data perspective, the CPE - Customer Premises Equipment is

Implementing the BSR for Basic Data Services

Rev. 1.0 11/01

Introduction High speed cable data communications share

Importance Frameworks and models provide a means of

Lesson Overview We will look at models as they relate to

Objectives Upon successful completion of this module, you will

Topic Order Components of a Conversation Page 5

Topic Order (cont.)

Topic Order The Data Link Layer – Layer 2 Page 19

Consider a conversation between two people. There are a number of common

Who Makes the Rules?

Ô International Telecommunication Union (ITU)

ÔRule, condition, or requirement

Request for Comments

ÔInternet Society (ISOC)

A communications protocol is an agreement among parties on a format and

The ISO’s OSI Model

The International Organization for Standardization (ISO) developed the Open

Remembering the OSI Model

 Again  Again  All  All

The Application Layer — Layer 7

Ô Service use and advertisement

Again Again All All