LANs, WANs, MANs

Febriliyan Samopa
LANs, WANs, and MANs
• Ownership
• WANs can be either public or private
• LANs are usually privately owned
• Capacity
• LANs are usually higher capacity, to carry greater
internal communications load
• Coverage
• LANs are typically limited to a single location
• WANs interconnect locations
• MANs occupy a middle ground
Network Services Available for MAN
and WAN

• Dialed Circuit Services

• Direct Dialing (DD) & Wide Area Telephone Services (WATS)
• Dedicated Circuit Services
• Voice-grade circuits
• Wideband Analog Services
• T-Carrier Circuits
• Synchronous Optical Network (SONET)
• Digital Subscriber Line (DSL)
• Cable modem
• Circuit-Switched Services
• Integrated Digital Services Network (Narrowband & Broadband)
• Packet-Switched Services
• X.25, Frame Relay, ATM, and SMDS
Tiered LANs
• Cost of attachment to a LAN tends to
increase with data rate
• Alternative to connecting all devices is to
have multiple tiers
• Multiple advantages
• Higher reliability
• Greater capacity (less saturation)
• Better distribution of costs based on need

4
Tiered LAN Diagram

5
Network Interface Card (NIC)

NIC

NIC (Network Interface Card or Network Adapter)

• Interface between a computer and a LAN
• CPU can’t process binary data at network speeds (10Mbps, 100Mbps,
1GMbps)
• NIC contains sufficient hardware to process data independent of system
CPU
• System CPU forms message request
• Sends instructions to NIC to transmit data R
• eceives interrupt on arrival of incoming data
• NIC is built for one kind of physical network
• Ethernet interface can't be used with token ring
• ATM interface can't be used with FDDI
NIC Example in Desktop Computer
Some NIC can be used with different
but similar networks: thick, thin and
10Base-T Ethernet, Fast Ethernet
LAN Topology
• Arrangement of workstations in a shared
medium environment
• Logical arrangement (data flow)
• Physical arrangement (cabling scheme)

8
LAN Topology, cont...
LAN Topologies: Bus
• Multipoint medium
• Stations attach to linear medium (bus) using
tap
• Full-duplex between station and tap
• Transmission from any stations travels
entire medium (both directions)
• Termination required at ends of bus

10
Bus LAN Diagram

11
LAN Topologies: Tree
• Generalization of bus topology
• Branching cable with no closed loops
• Cable(s) begin at headend, travel to
branches which may have branches of their
own
• Each transmission propagates through
network, can be received by any station

12
Tree LAN Diagram

13
Bus/Tree Topology Problems
• How do you identify who the transmission is
intended for?
• Data transmitted in frames
• Each frame has header with addressing info
• How do you regulate access?
• Stations take turns sending, by monitoring
control information in frames

14
LAN Topologies: Ring
• Repeaters are joined by unidirectional
point-to-point links in a ring
• As a frame circulates past a receiver, the
receiver checks its address, and copies
those intended for it into a local buffer
• Frame circulates until it returns to source,
which removes it from network

15
Ring LAN Diagram

16
LAN Topologies: Star
• Each station connected directly to central
node, usually with two unidirectional links
• Central node can broadcast info, or can
switch frames among stations

17
Star LAN Diagram

18
Choosing a Topology
• Factors to consider include reliability,
flexibility/expandability, and performance
• Bus/tree is most flexible
• Tree topology easy to lay out
• Ring provides high throughput, but reliability
problems
• Star can be high speed for short distances, but
has limited expandability

19
Medium and Topology

20
LAN Standards (IEEE802.x)
• Advantages of standards
• Assure sufficient volume to keep costs down
• Enable equipment from various sources to
interconnect
• IEEE 802 committee developed, revises,
and extends standards
• Use a three-layer protocol hierarchy:
physical, medium access control (MAC),
and logical link control (LLC)

21
 IEEE LAN Standards

• IEEE 802.2: Logic link control (LLC) layer of data link layer
• IEEE 802.3: Ethernet
• IEEE 802.4: Token bus, an old protocol
• IEEE 802.5: Token ring
• IEEE 802.6: Distributed queue dual bus (DQDB) protocol, similar to
FDDI
• IEEE 802.9: Integrated voice and data networking, including ISDN,
Iso-ethernet
• IEEE 802.11: Wireless LAN
• IEEE 802.12: 100Base-VG
• IEEE 802.13: 100Base-X
• IEEE 802.14: Cable modem

22
*Logical Link Control
• Specifies method of addressing and controls exchange
of data
• Independent of topology, medium, and medium access
control
• Unacknowledged connectionless service (higher layers
handle error/flow control, or simple apps)
• Connection-mode service (devices without higher-level
software)
• Acknowledged connectionless service (no prior
connection necessary)

23
*Medium Access Control

• Provides a means of • LLC frames data,

controlling access to passes it to MAC
a shared medium which frames it
• Two techniques in again
wide use • MAC control
• CSMA/CD (e.g. priority level)
• Token passing • Destination physical
address
• Source physical
address

24
Ethernet (IEEE 802.3)

The Ethernet LAN standard was originally

developed by DEC, Xerox, and Intel, but
has since become a formalized standard by
the Institute of Electrical and Electronics
Engineers as IEEE 802.3

25
Inventors of Ethernet
Robert Metcalfe (1946, US)
David R. Boggs (1950, US)
- Developed at Xerox PARC, 1973-1974
• Metcalfe identifies the day Ethernet: May 22,
1973, circulated a memo titled "Alto
Ethernet" containing a rough scheme.
• Boggs identifies another date as birth of
Ethernet: November 11, 1973, the first day
the system actually functioned.
• The two would co-invent Ethernet, with
Metcalfe generating the ideas and Boggs
figuring out how to build the system.
• Patented with the 2 & Thacker, Lampson (75)
• Metcalfe, Boggs, "Ethernet: Distributed
Packet Switching for Local Computer
Networks". Communications of the ACM
(1976)
27
 Ethernet Topology
• Ethernet uses a bus topology (a high speed circuit and a
limited distance between the computers, such as within
one building).
• From the outside, an ethernet LAN appears to be a star,
because all cables connect to the central hub.
• Most ethernet LANs span sufficient distance to require
several hubs, but some ethernet LANs are build without
the use of hubs (coax bus).

28
 Ethernet Media Access Control
Carrier Sense Multiple Access with Collision
Detection (CSMA/CD)

• Wait until the bus is free and then transmit.

• If no collision, transmission is completed.
• If the collision is detected, send a jamming signal.
• Wait a random amount of time, then re-broadcast.

29
CSMA/CD Flowchart
 Ethernet Wiring
segment
Thick Ethernet wiring (Thicknet)
- the first generation
- use thick coax cable (1cmD)
- AUI cable connects NIC to transceiver
- digital signal in AUI
(Attachment Unit Interface)
- analog signal in coax
- 10Base5
10Mbps
max segment: 500m
computer spacing: 2.5m

Thin Ethernet wiring (Thinnet)

- the second generation
- use thin coax cable (0.5cmD)
- BNC connects to thin cable
- 10Base2
10Mbps
max segment: 185m
computer spacing: 0.5m
 Ethernet Wiring, cont...
Hub Ethernet wiring
- the third generation, most popular
- use hub
a device with connections to computers
physically star topology, logically bus
same frame format, follow CSMA/CD
 “Ethernet-in-a-box” Hub
- 10Base-T
10Mbps, twisted pair, RJ-45 connector
max wire length: 100m
number of computers: port number
- 100Base-Tx/T4
100Mbps, twisted pair
max wire length: 100m
- 1000Base-T/CX
1000Mbps, twisted pair
max wire length: 25m
- 1000Base-SX/LX
1000Mbps, optical fiber
max fiber length: 550m
Comparison of
Wiring Schemes
Thicknet

Thinnet

Hub Ethernet

-Hybrid Wiring in practical-

Ethernet Frame Format

Preamble: 64bits (8 bytes) for receiver synchronization with incoming

signal
Destination address and source address
Frame type (2 bytes):
- 0000~05DC Reserved for use with IEEE LLC/SNAP
- 0800 Internet IP Version 4
- 8008 AT&T Corporation
- 8014 Silicon Graphics Corporation network games
- 8137-8138 Novell Corporation IPX
-…
Data or Payload: Minimum 46 bytes and maximum 1500 bytes
CRC (4 bytes): Cyclic redundancy check
Token Ring (IEEE 802.5)
Token Ring was originally developed
by IBM, and have since been
standardized by IEEE as IEEE 802.5.

35
Topology

36
 Token Ring Media Access Control
• Token ring uses a controlled-access technique called token passing.
• The “token” is a series of bits, travels between the computers in a
predetermined sequence.
• A computer with a message waits to transmit until it receives a free
token. The computer changes the free token to a busy token and
attaches its message to it. Then it retransmits it on the circuit to the
next computer in the sequence.
• The computer receiving the message, changes the acknowledgement
to ACK (or NAK) and sends the message back to the sender, who
creates a new free token.

37
Token Ring Media Access Control

38
Token Ring Media Access Control

39
 Token Ring Media Access Control

Token loss:
• The token crashes before being transmitted - lost a free token
• A computer in the ring crashes - lost a busy token
• A token is always busy.
A solution for the “lost” token problem:
• Designate one computer to be the token monitor and another
computer to be a backup token monitor.
• If no token circulated through the network for a certain length of time
or if a busy token circulates too often, the token monitor will create a
new free token (and destroy the busy token if necessary.)

40
FDDI - Ring Topology
SD FC ED Token frame (3 bytes)

Data frame

SD FC DAdd SAdd Payload CRC ED FS

Up to 4550 bytes

Features: multiple frames for MM data or RT trans.

• Fiber Distributed Data Interconnect (FDDI) using fiber optics cable and
Copper Distributed Data Interconnect (CDDI) using copper cable
• Topology: Two rings
• Transmit: Bits can flow in 2 directions for improving reliability
• Speed: 100 Mbps
• Status: ANSI standard, expensive than Ethernet and Token Ring
• Token: A permission to send data across the ring.
Only one token in a ring and the token rotates one by one.
If no data to be sent, computer passes received token to the next.
When a computer obtains the token, it can send a single packet.
A packet travels around entire ring and finally arrives source.
ATM LAN - Star Topology

• ATM: Asynchronous Transmission Mode

• Media: Fiber
• Topology: Star
• Transmit: More than one can transmit because of using switch
• Speed: 25, 45, 155, 622 Mbps
• Status: ATM Forum and ATM Consortium, high speed for multimedia com
• Switch: Each station connects to switch with full-duplex connections.
Computers send data independently and simultaneously.
Packet size is very small, 53 bytes, called cell.
Very expensive
Extending LAN

• Distance and computer number Limitation of LAN

• Thicknet: 500m, Thinnet: 185m, 10/100Base-T: 100m
• Thicknet: 100 computers, Thinnet: 30 comp, 10/100Base-T: <100
• Signal attenuation and noise across long distance
• CSMA/CD can’t work across arbitrary distance for Ethernet
• Long circulation time for token passing in Token Ring and FDDI
• Motivation to extend LAN
• Large organization has many computers in different places
• Requirement in extending LAN
• Must not violate designs assumptions in original LANs
• Methods and hardware in extending LAN
• Optical fiber and fiber modem
• Repeater or hub
• Bridge
• Switch
Fiber Optical Extension

• Optical fiber has high bandwidth and low delay

• connect one computer to a far LAN
• connect two LANs in certain distance (e.g., 2 LANs in two
buildings)
• Fiber modem
• convert electronic signal in LAN cable into light signal in fiber
• Note: can’t use telephone line modem to directly connect a
computer to a remote LAN because its low speed
(34/56Kbps)
Repeater

• Repeater:
• A bi-directional device connect two LAN segments
• Doubles Thicknet segment from 500m to 1000m
• Accepts attenuated and noise signal and output reconstruct signal
• Does not understand frame format and have no hardware address
• Propagates error bits and collisions
Hubs
• The active central element of the star
layout.
• When a single station transmits, the hub
repeats the signal on the outgoing line to
each station.
• Physically a star; logically a bus.
• Hubs can be cascaded in a hierarchical
configuration.

46
Hubs
Operating at the physical layer, hubs are very
simple devices that pass all traffic in both
directions between the LAN sections they link.
They may connect different types of cable, but use
the same data link and network protocol.
Strictly speaking, hubs are not considered part of a
backbone network, but are usually repeaters or
amplifiers.

47
Application of Repeater and Its Limit

Extended LAN in a building

• Can’t extend Ethernet with repeaters infinitely

• CSMA/CD won’t work if medium is too long and has large delay
• Maximum 4 repeaters between any two Ethernet computers
• A hub can connect other hub(s) and it functions as a repeater
• Be careful in such connection !!
Bridges
• Allow connections between LANs and to WANs
• Operates at Layer 2 (Data Link Layer) of OSI
• Used between networks using identical
physical and link layer protocols
• Handles both bit reconstruction like repeater and complete frame
• Knows which computer is attached to which segment using a table
• Checks packet destination address and forward it if necessary
• Forwards all broadcast and multicast packet
• Does not forward error packet and collision
• Allows concurrent use of different segments if traffic is local
• Provide a number of advantages
• Reliability: Creates self-contained units
• Performance: Less contention
• Security: Not all data broadcast to all users
• Geography: Allows long-distance links
49
Bridges

50
Bridge Functions
• Read all frames from each network
• Accept frames from sender on one network that are
addressed to a receiver on the other network
• Retransmit frames from sender using MAC protocol
for receiver
• Must have some routing information stored in order to
know which frames to pass

51
Set Up Bridge Table

• Initially, the forwarding table in a bridge is empty.

• Bridge uses source address to learn location of each computer
• Learning is completely automated and fast
• Examines source address in each frame, add entry to list for a
segment
Bridges
If a bridge receives a packet with a destination
address that is not in the address table, it
forwards the packet to all networks or network
segments except the one on which it was
received.
Bridges are a combination of both hardware and
software, typically a “black box” that sits between
the two networks, but can also be a computer with
two NICs and special software.

53
Bridge Operation

54
Optical Fiber Bridging between
Buildings

• Similar to extending AUI with fiber modems

• Can put a bridge in one building with a long connection to
other LAN segment in different building
• Avoids extended AUI connection for each computer in
remote building
Bridging Across Longer Distance

• Can use leased line, microwave, laser or satellite to

connect two LANs
• Using two bridges instead of one
• Filters at both ends, reducing traffic across slow link
• Provides buffering at both ends, matching dissimilar
transmission speeds
Bridge Tree and Bridge Cycle
Bridge tree

Bridge cycle

• Bridge Tree : Use multiple bridges to connect LANs to form a large network
e.g., campus network, concurrent transmissions in each LAN
• Bridge Cycle : Several bridges form a loop
• A computer receives two copies of a same packet
• A broadcast packet results in infinite packets in the loop
• A loop is hard to avoid in a large and dynamic network
• Spanning Tree Algorithm (IEEE 802.1 Group)
Switches

Like bridges, switches operate at the data link

layer. Switches connect two or more
computers or network segments that use
the same data link and network protocol.
They may connect the same or different
types of cable.

58
Switches
Switches operate at the same layers as bridges but
differ from them in two ways:

• First, most switches enable all ports to be in use

simultaneously, making them faster than bridges.

• Second, unlike bridges, switches don’t learn addresses,

and need to have addresses defined.

Example: Intel Express 510 T switch.

59
Switched Ethernet
• A simple concept behind switched Ethernet
- replace the LAN hub with a switch. Each
computer now has its own dedicated point-
to-point circuit.
• By increasing the number of connections
from the server to the switch, the
throughput of the server will be improved
because of more circuits.

60
Switched Ethernet

61
Ethernet Hubs and Switches
Hub

(not an actual implementation)

• Switch
• A device interconnects computers or LANs
• Physically similar to a hub and logically similar to a bridge array
• One LAN segment per host and bridges interconnect segments
• Operates on packets, understand addresses, only forward if necessary
• Permits concurrent/simultaneous transmissions
• Higher cost than hub per port
Ethernet Hubs and Switches

• Shared
medium
hubs
x
• Switched
LAN hubs

63
Types of Switches
• Store and forward switch
• Accepts a frame on input line
• Buffers it briefly
• Routes it to appropriate output line
• Cut-through switch
• Begins repeating the frame as soon as it recognizes the
destination MAC address
• Higher throughput, increased chance of error

64
Routers
Routers operate at the network layer. Routers connect two
or more LANs that use the same or different data link
protocols, but the same network protocol.
Routers may be “black boxes,” computers with several NICs,
or special network modules in computers.
In general they perform more processing on each message
than bridges and therefore operate more slowly.

65
Routers

66
Routers vs Bridges
• Routers can choose the best route.
• Routers also only process messages specifically
addressed to it.
• Routers can connect networks using different data
link layer protocols. Therefore, routers are able to
change data link layer packets.
• Routers may split a message into several smaller
messages for transmission.

67
Layer 3 Switches
• Problems With Layer 2 Switches
• Broadcast overload because of the single MAC broadcast address (e.g.
using ARP for Data Link Layer address resolution)
• Lack of multiple links - only one path
• Normally, the above problems can be solved with several subnets
connected by routers. However,
• A MAC broadcast frame is then limited to only the devices and switches
contained in a single subnet.
• A router does all IP-level processing, some of which could be not
necessary.
• It is implemented in software and slow.
• Layer 3 switches implement the packet-forwarding logic of the router
in hardware.

68
Gateways
Gateways operate at the network layer and use
network layer addresses in processing messages.
Gateways connect two or more LANs that use the
same or different (usually different) data link and
network protocols. They may connect the same
or different kinds of cable.
Gateways process only those messages explicitly
addressed to them.

69
Gateways
Gateways translate one network protocol into
another, translate data formats, and open
sessions between application programs, thus
overcoming both hardware and software
incompatibilities.
A gateway may be a stand-alone microcomputer
with several NICs and special software, a FEP
connected to a mainframe computer, or even a
special circuit card in the network server.

70
Gateways

One of the most common uses of gateways is to

enable LANs that use TCP/IP and ethernet to
communicate with IBM mainframes that use SNA.

The gateway provides both the basic system

interconnection and the necessary translation
between the protocols in both directions.

71
Gateways

72
A Caveat

The terminology used in the marketplace

may differ substantially. One vendor’s
bridge may provide the functions of a
router.
• Multiprotocol routers -can understand several
different network layer protocols.
• Brouters – Combine the functions of both bridges
and routers. They operate at both data link and
network layers.

73
Cable Modem Configuration

20:02:43
IP Address 68.1.216.37
Subnet Mask 255.255.248.0
Gateway 68.1.216.1
Domain Name Server 68.1.208.30, 68.1.18.30

74
Comparison of Networking Options
Types of WANs
Circuit-Switching

• Definition: Communication in which a

dedicated communications path is
established between two devices through
one or more intermediate switching nodes
• Dominant in both voice and data
communications today
• e.g. PSTN is a circuit-switched network
• Relatively inefficient (100% dedication
even without 100% utilization)
Circuit-Switching Stages
• Circuit establishment
• Transfer of information
• point-to-point from endpoints to node
• internal switching/multiplexing among nodes
• Circuit disconnect
Circuit Establishment
• Station requests connection from node
• Node determines best route, sends message to
next link
• Each subsequent node continues the
establishment of a path
• Once nodes have established connection, test
message is sent to determine if receiver is
ready/able to accept message
Information Transfer
• Point-to-point transfer from source to node
• Internal switching and multiplexed transfer
from node to node
• Point-to-point transfer from node to receiver
• Usually a full-duplex connection throughout
circuit
Circuit Disconnect
• When transfer is complete, one station
initiates termination
• Signals must be propagated to all nodes
used in transit in order to free up resources
 Long Distance Telephone Network
• A single LAN can’t handle an arbitrary number of computers.
• A LAN is usually owned by a company, organization, university, …, which can’t
build their own long distance links to connect computers in arbitrary sites.
• Telephone and other companies offer long distance connections which can be
leased to build large computer networks covered many distance sites.

Telephone Company
Circuit Switch Analog Network
Modem Modem
AD/DA AD/DA
Circuit Switch Digital Network
Convert Convert
Comp
LAN
Packet Switch Digital Network Comp
LAN

Other Company
Modem CATV/Power Network Modem

Modem Satellite Network Modem

...
 Digital Telephone Network and PCM
Analog telephone network: transfers analog voice signal in the frequency range [0, 3.5KHz]
Digital telephone network: transfers digital voice signal
PCM (Pulse Code Modulation): a standard of conversion from analog voice to digital voice
- Analog signal is first sampled in sampling interval T or with sampling frequency F=1/T
- Nyquist sampling theorem: Original analog signal can be reconstructed
from a sequence of sampled values if F>2B where B is the maximum signal frequency
- Sampling frequency of digital voice = 8 KHz
- Each sampled value is quantized to an integer in [-127, 127]
- Encoding the integer to an 8-bit binary value, PCM hardware is a analog-to-digital (AD) device
- Bit rate of one digital voice channel: 8KHz x 8 bits = 64 Kbps
Nyquist Rate
00000111 F<2B
00000110
00000101
00000100 F=2B
00000011
00000010
00000001 F>2B
00000000
T T, sampling interval
F=1/T, sampling frequency
DSU/CSU

- Computer industry and telephone industry use own standards in different encoding
- DSU/CSU perform the conversion between two standards
- DSU (Data Service Unit): Translations between two encoding
- CSU (Channel Service Unit): management of line termination and diagnose

DSU/CSU  Gateway
Digital Telephone Line Standard

- Specified by the telephone industry in each country, different from the world
- Fractional T1= 64, 9.6 or 4.8Kbps, use multiple fractional T1 with multiplexer
- Can be leased in two points
High Capacity Digital Lines

• Use both electrical signal and optical signal

• Electrical standard called Synchronous Transport Signal
(STS)
• Optical standard called Optical Carrier (OC)
• Engineers usually use OC- terminology for everything
• OC-3 is popular
• OC-3: three OC-1 lines, OC-3C: a single line operating at
155.520 Mbps
Local Loop

Analog line Local

ISDN line Central Telephone
Office network
DSL line ( CO )
Subscribers

Local Loop

• Analog line :
• Analog signal (300~3300Hz), modem needed, low speed (34/56Kbps)
• Also called PSTN
• ISDN (Integrated Services Digital Network):
• BRI (Basic Rate Interface) = 2B+D, B=64Kbps for data, D=16Kbps for control
• PRI (Primary Rate Interface)=23B+D=1.544Mbps=T1, D=64Kbps for control
• xDSL (Digital Subscriber Line)
• Use existing telephone line to provide high speed transmission, current technology
Public Switched Telephone Network
(PSTN)

• Subscribers • Trunks
• Local loop
• Connects subscriber to
local telco exchange
• Exchanges
• Telco switching centers
• Also known as end office
• >19,000 in US
• Connections between
exchanges
• Carry multiple voice
circuits using FDM or
synchronous TDM
• Managed by IXCs (inter-
exchange carriers)

Services:
1. Dial-up line
2. Dedicated line
ISDN

• 1st generation: narrowband ISDN

• Basic Rate Interface (BRI)
• two 64Kbps bearer channels + 16Kbps data channel
(2B+D) = 144 Kbps
• circuit-switched
• 2nd generation: broadband ISDN (B-ISDN)
• Primary Rate Interface (PRI)
• twenty-three 64Kbps bearer channels + 64 kbps data
channel (23B+D) = 1.536 Mbps
• packet-switched network
• development effort led to ATM/cell relay
Past Criticism of ISDN
• “Innovations Subscribers Don’t Need” , “It Still
Doesn’t Network” , “It Still Does Nothing”
• Why so much criticism?
• overhyping of services before delivery
• high price of equipment
• delay in implementing infrastructure
• incompatibility between providers' equipment.
• Didn’t live up to early promises
ISDN Principles
• Support of voice and nonvoice using limited set of
standard facilities
• Support for switched and nonswitched
applications
• Reliance on 64kbps connections
• Intelligence in the networks
• Layered protocol architecture (can be mapped
onto OSI model)
• Variety of configurations
ISDN Network Architecture
• Physical path from user to office
• subscriber loop, aka local loop
• full-duplex
• primarily twisted pair, but fiber use growing
• Central office connecting subscriber loops
• B channels: 64 kbps
• D channels: 16 or 64 kbps
• H channels: 384, 1536, or 1920 kbps
ISDN B Channel

• Basic user channel (aka “bearer channel”)

• Can carry digital voice, data, or mixture
• Mixed data must have same destination
• Four kinds of connections possible
• Circuit-switched
• Packet-switched
• Frame mode
• Semipermanent
ISDN D Channel
• Carries signaling information using
common-channel signaling
• call management
• billing data
• Allows B channels to be used more
efficiently
• Can be used for packet switching
ISDN H Channel
• Only available over primary interface
• High speed rates
• Used in ATM
ISDN Basic Access
• Basic Rate Interface (BRI)
• Two full-duplex 64 kbps B channels
• One full-duplex 16 kbps D channel
• Framing, synchronization, and overhead bring
total data rate to 192 kbps
• Can be supported by existing twisted pair local
loops
• 2B+D most common, but 1B+D available
ISDN Primary Access
• Primary Rate Interface (PRI)
• Used when greater capacity required
• No international agreement on rates
• US, Canada, Japan: 1.544 mbps (= to T1)
• Europe: 2.048 mbps
• Typically 23 64 kbps B + 1 64 kbps D
• Fractional use of nB+D possible
• Can be used to support H channels
 xDSL Technology
xDSL is a family of technologies to use existing analog telephone line (copper pair)
for delivery of high-speed data between a subscriber and telephone central office.
xDSL family
Telephone Network
ADSL: Asymmetric Digital Subscriber Line
HDSL: High-rate Digital Subscriber Line
SDSL: Symmetric Digital Subscriber Line
Local Central Office (CO) VDSL: Very high bite rate Digital Subscriber Line

Analogy phone DSL xDSL ADSL HDSL SDSL VDSL

Bit rate 32K~3M Up 1.5M 1.5M 1.5~2.3M Up
32K~40M Down 1.5M 1.5M 13~52M Down
f (KHz) Mode Asymmetric Symmetric Symm Asym
0 20 25 Fa Fb Fc wire 1 pair 2 pairs 1 pair 1 pair
Upstream Downstream Length 3.7~5.5km 3.7km 3.0km 1.4km

Phone Computer FDM – Frequency Division Multiplexing

Frequency range (<20KHz) for analog phone
Frequency range (>25KHz) for data transmission
Local line quality
Spectrum of copper pair
Low frequency is better than high frequency
Characteristic of each line is different from others

0 f (KHz)
 ADSL

DMT (Discrete MultiTone) by ANSI

- FDM and QAM modulation
- 31 channels (carriers) for upstream
- 255 channels (carriers) for downstream
Analogy
phone

… … … f (KHz)

0 20 25 200 250 1M
Upstream Downstream

Spectrum of copper pair

ADSL2/2+, VDSL2, etc.

http://en.wikipedia.org/wiki/Asymmetric_Digital_Subscriber_Line_2
0 f (KHz)

• Concurrent and independent uses of analog phone and data transmission

• Upstream ~few Mbps, Downstream ~tenths M, suited for accessing Internet
Cable Modem Technology
• Cable TV already brings high bandwidth coax into houses

• Conventional Cable TV is one direction …

• Signal broadcast at central location
• Amplifiers boost signal through network M M
• Amplifiers are unidirectional
CATV
Center
• Solutions:
Amplifier
• Alternate upstream path - e.g., dialup M - Cable Modem
• Replaced by bi-directional amplifiers

• Cable modems encode and decode data from cable TV coax

• One in cable TV center connects to network
• One in home connects to computer

TV TV …
f (MHz)
• Users share the cable
• Each subscriber is assigned an address
Upstream Downstream
• TDM is used like multiple computers in LAN (shared by multiple users)
Cable Modem Connections

Hybrid Fibre-Coaxial (HFC)

Cable Modem
Connections
FTTX - Fiber To The X
WAN and Switch
• WAN building blocks:
1. Long-distance connection (e.g., T1/T3 lines,
OC1/OC3 lines, etc.)
2. Packet switch :
• A hardware device connected to other switches and
computers
• Has CPU, memory, I/O interface, etc.
• Handles packet
Packet
Header Frame Data or Payload Trailer

Dest Addr Src Addr

Illustration of A WAN

• Place one or more switches at each site

• Interconnect switches:
• LAN technology for local connections
• Leased digital lines for long distance connections
• Interconnections depend on traffic amount and reliability requirement
• Packet is sent from source, travels switch-to-switch, and delivered to destination
• Switch stores packet in memory, examines address, and forwards it toward destination
Addressing in WAN
• Each computer has a unique address
• It is a two-part hierarchical address
including 2 integers: [Integer1, Integer2]

Integer1 for switch number Integer2 for computer number

Next-Hop Forwarding and Routing
• Packet switch must choose outgoing connection for forwarding
• If destination is local computer, packet switch delivers computer port
• If destination is attached another switch, this packet switch forwards to next hop through
connection to another switch
• Packet switch doesn't keep complete information about all possible destination
• A routing table just keeps next hop
• For each packet, packet switch looks up destination in its routing table and
forwards through connection to next hop
• Next hop to destination does not depend on source of packet
• Called source independence

Routing table of switch 2

 Routing Table and Graph

Graph: Node models switch

Edge models connection

Routing tables
without default routes

Routing tables
with default routes

- Each switch or node has a routing table: Left entry right entry
destination switch edge number pair
- Default route can remove duplicate routes, reduce memory and improve performance
 Routing Table Computation
- Static routing table: fixed in switch booting and does not change
- Dynamic routing table: initialized in booting and alters as conditions in the network change

Shortest path computation:

- Label on edge represents “distance”
- Possible distance metric: geographic distance, economic cost, capacity, etc
- “Shortest” means the minimum sum of distances in all paths between two nodes
- Darkened path is minimum from node 4 to node 5

Algorithms for computing shortest paths

- Dijkstra’s algorithm
- Distance Vector (DV) algorithm
History of WAN Technologies
• ARPANET
• The first WAN, began in 1960s, low speed: 56Kbps
• Funded by Advanced Research Projects Agency, an organization of US Defense Department
• Incubator for many of current ideas, algorithms and internet technologies
• X.25
• Early standard for connection-oriented networking from ITU, which was originally CCITT
• Early commercial service, more popular in Europe
• Predates computer connections, used for terminal/timesharing connection
• Frame Relay
• Telco service for delivering blocks of data
• Connection-based service;
• Typically 56Kbps or 1.5Mbps; can run to 100Mbps
• SMDS - Switched Multi-megabit Data Service
• Connectionless service; any SMDS station can send a frame to any other station
• Typically 1.5-100Mbps
• ATM - Asynchronous Transfer Mode
• Designed as single technology for voice, video, data, ...
• Low jitter (variance in delivery time) and high capacity
• Uses fixed size, small cells - 48 bytes data, 5 bytes header
• Can connect multiple ATM switches into a network
• Potential in future B-ISDN
• Can be worked as a LAN