Payload 4 02 | Digital Subscriber Line | Computer Network

Chapter 5: SONET/SDH

Reference textbook: ³Understanding SONET/SDH and ATM: Communications Networks for The Next Millennium´ By S1. V. Kartalopoulos

Introduction 
³Humans are social animals!´  Socializing p exchanging ideas and information p build networks to facilitate information exchange and communication  Different ways of communication technologies  Telegraph  Telephony  Digital electronic technology of various data types
‡ ‡ ‡ ‡ Telemetry data User data Video data Voice data

Fig. 5.1 Characteristics per service

5.1 Legacy Communications Systems: Concepts
Basic Technology and Services

Pulse-Coded Modulation ( PCM ) 
Voice : the primary services in the communication industry  Voice : analog signal.  Words : acoustic waves generated by vibrating vocal cords and the mouth cavity modulates them into recognizable and distinguishable compounded sounds  Microphone: a transducer converts acoustic waves into (analog) electrical signals  Speaker: a transducer converts (analog) electrical signals into acoustic waves and reproduces voice sounds  POTS ( Plain Old Telephone Service )  Ringing  call initiation  number dialing  Transmits analog signals over the telephone network

Pulse-Coded Modulation ( PCM ) 
          PCM form a binary (or digital) bit stream at 64000 bits/second Also know as Digital Signal level 0 ( DS0 ) (en)Coder : converts analog signal into digital bit steams Decoder : converts digital bit streams into analog signal (voices) CODEC ( coder/decoder ) : periodically samples the analog signal, and based on a conversion scheme, it translates each sampled value into a binary representation Two major schemes for PCM : Q-law in United States, E-low in Europe Acoustical signal of speaking voice : usually under 3.4 kilocycles per second, or kilohertz ( 3.4 kHz ) A low-pass filter removes all components of frequency higher than 3.4 kHz By sampling theorem, to truly represent a signal, we must sample at least twice as its maximum frequency content So PCM CODEC samples at analog voices 8000 times per second, or every 125 ms, and it converts each sample into 8 bits PCM The total bit rate is 8000 × 8 = 64000 bits / second, or 64 kilobits per second ( 64 Kbps ), this is the rate termed DS0.

Fig. 5.2 Analog to PCM

Fig. 5.3 DS0 rate

Pulse-Coded Modulation ( PCM ) 
Differential PCM ( DPCM ) , adaptive DPCM ( ADPCM ), and sigma-delta PCM ( PCM ) : by more sophisticated digital signal processing algorithms to compress the 64 Kbps to 32 Kbps, or 16 Kbps, or even lower than that.  Local loop : a pair of twisted copper wires between user¶s equipment ( a POTS telephone ) and the service provider¶s equipment, where the CODEC is located. The signal transmitted on a local loop cable is an analog electrical signal.

Time-Division Multiplexing ( TDM ) 
POTS telephone converts voice signals into an electrical signal  The electrical signal is sent over a pair of copper wires to a communications systems where the CODEC function is performed  The backbone network only transmits digital signals and no analog signals : all-digital communications network  On an all-digital communications network, in addition to digitized voice signals, other digital data can also be transmitted ( raw digital data, encoded video, encoded sound, etc. )

Time-Division Multiplexing ( TDM ) 
Some digital services on DS0  Digital Data Service ( DDS )
‡ 56 Kbps for data, 8 Kbps for ³in-band´ signaling ‡ Only for the backbone network, not for the end users 

Basic Rate Integrated Services Digital Network ( BRISDN or BRI )
‡ Uses two 64 Kbps channels ( channels B ) and a 16 Kbps subrate channel ( channel D ) ‡ Support a combination of voice and/or data services over a single pair of wire ‡ Modes available for end users :
± ± ± ± ± 2 B channels for voice and a D channel for data 1 B channel for voice and a B+D channels for data 2 B channels for data and a D channel for signaling All channels for data Can only be used if both end users has ISDN and the network supports ISDN

Table 5.1 DS0 Rates and Line Codes
Source Voice : 64 Kbps DDS 56 Kbps + 8 Kbps signalling B-ISDN 64 Kbps B channel ( 2B + D ) Abbreviations : CCC, clear channel capability; B = 64 Kbps voice or data channel; DDS, digital data services; D = 16 Kbps signaling or data channel Format, overhead line code Frame, bit robbed 64 CCC B8ZS 64 CCC B8ZS

Time-Division Multiplexing ( TDM ) 
User-to-network interface ( UNI ) : the interface where the user signal, a 64 Kbps data stream ( organized in 8-bit bytes ), first meets the network  At UNI, users¶ signals are synchronized with a 8 kHz system clock, or a multiple of it.  Then based on a round-robin principle, the signals are sequentially polled one byte at a time and place one after the other in a fixed order, this is called byte-interleaving.  Time slot : the location of each byte source in this ordered digital signal.  The whole process is called Time-Division Multiplexing ( TDM ).

DS1 Rate 
Multiplex 24 DS0 users  Each user contributes one byte in a DS1 frame  An additional bit, the F-bit, is used as the beginning of a DS1 frame  Total 24 × 8 + 1 = 193 bits in 125 Qs.  The bit rate is 193 bits / 125 ms = 1.544 Mbps, knows as the digital signal level 1 ( DS1 ) rate.  M1 multiplexer ( M1 mux ) : function that TDM 24 DS0 into a DS1.  The F-bit of each frame forms an F subrate channel in DS1 superframe  12 or 24 frames in each super frame  Provide a data link over which the network data are sent  Provide error control

Fig. 5.4 DS1 rate ± M1 Mux

DS1 Formats 
DS1 frame = 24 time slots and the F-bit  For primary rate ISDN ( PRI )  23 time slots : 23 B channels ( at 64 Kbps each )  1 time slot : for 4 D channels  DS 1 signal  Alternate mark inversion ( AMI ), or called bipolar with 8zero substitution, a.k.a. bits 8-zero suppression ( B8ZS )  B8ZS or more generally, BnZS is used to prevent transmit long strings of 0¶s by the minimum number of 1¶s

Table 5.2 DS1 Rate and Line Codes
Source Voice : 24 × DS0 DDS 23 × DS0 + sync + frame P-ISDN 23B + D B channel ( 2B + D ) Format, Overhead Line Code SF, ESF, AMI SF, ESF, B8ZS SF, ESF AMI

Abbreviations : SF, superframe; ESF, extended superframe; AMI, alternate mark inversion.

DS1 Over Long Distnace 
DS1 signal is transmitted over a T1 line  One pair of wires is used for each direction  Communication systems at the DS1 level are designed with an error rate of 10-6 or better.  A regenerator is placed after 3000 ft from user and every 6000 ft between regenerators.  A regenerator every 6000 ft places a maintenance and troubleshooting overhead.

Fig. 5.5 T1 characteristics

xDSL 
Digital Subscriber Line ( DSL ) : the local loop ( i.e. the copper twister-pair wires between most homes and the telephone service provider equipment ) in digital form, such as the BRI.  DSL provides 1.544 Mbps and in some cases up to 7 Mbps over existing twisted-pair copper cable.  DSL cannot be used on ³loaded´ loops, i.e., no inductors or coils on the loop cable.  Effective distance of a DSL depends on the data rate supported  1.5 Mbps : several miles  25 Mbps : only half a mile  xDSL : ³x´ refers to one of many DSL formats and rates.

xDSL
VDSL : very high-bit-rate DSL HDSL : high-bit-rate DSL ADSL : asymmetric DSL SDSL : symmetric DSL RADSL : rate adaptive DSL MSDSL : multi-rate symmetric DSL  Build on a single pair of DSL technology  On unloaded copper pair, can provide 64/128 Kbps up to 8.9 Km, or 2 Mbps up to 4.5 Km  DSL requires terminating devices at both ends of the loop, at the user and at the service provider, to terminate the upstream and downstream digital signals. 
    

Modulations on DSL 
Two-bits-to-one quaternary ( 2B1Q ) : Digital PAM with -3, -1, +1 and +3, four possible levels, each represents 2 bits.  Its transmitted power is superior to that of AMI  Bit rate is limited to 392 Kbps  Suitable for upstream transmission on local loos  This is used for BRI signals.  Discrete multi-tone ( DMT ) : divides the bandwidth into frequency channels.  When a certain channel is detected to have inferior transmission characteristics, the traffic is assigned another frequency channel, a.k.a. frequency hopping.  The official standard of the ANSI T1E1.r working group  Supports upto to 6 Mbps services
‡ Up to four MPEG-1 compressed video data ‡ A single MPEG-2 compressed video data

Modulations on DSL 
Carrierless amplitude phase ( CAP ): a derivative of the quadrature amplitude modulation ( QAM )  Translates a 4-bit code in one of the 16 voltage phase points  May be viewed as a 2B1Q 2-dimensional approach.  Transmitting power is superior to that of AMI and 2B1Q  Effective bit rate is in the range of 10 ± 175 Kbps  The de facto standard that by 1996 was deployed in almost 97% of all ADSL applications.

Table 5.3 Various Modulations on DSLs
Modulations # of wires in pairs 2 2 1 1 1 1 Downstream bit Upstream bit rate in Kbps rate in Kbps 144 2048 768 1500 6000 1500 144 2048 768 176 640 64 Max. length in one loop ( in ft ) 18000 13000 12000 12000 12000 18000

ISDN HDSL Two pairs HDSL Single pair ADSL DMT ADSL CAP ADSL CAP

LVDS 
Low-Voltage Differential Signaling ( LVDS )  Once defined for high-speed data transmission over relatively long cables  Now used as a high speed, 155.5 Mbps, low-power general purpose data transmission technology at the board / bus levels  LVDS differential voltage swing is between the voltage levels VOH = 1.4 V, and VOL = 1.0 V  Scalable Coherent Interface LVDS ( SCI-LVDS )  Specified by the IEEE 1596.3 standard  For unidirectional point-to-point links, from a transmitter to a receiver

1000BaseT 
An evolutionary standard derived from the 100BaseT used in LANs.  Working on by the IEEE 802.3ab task force.  Allows for transmission of a balanced digital signal at 1 Gbps over UTP-5 cable for the distance of 100 m.  Most of the applications are in the LAN

Coding Schemes 
Unipolar : a two-voltage-level signal that typically swings between zero voltage and + V.  Bipolar : a three-voltage-level signal that typically swings between a positive and a negative voltage  May return to zero ( RZ ) or nonreturn to zero ( NRZ )  In a digital bipolar signal, the 1¶s alternate between the two voltages, ³+´ and ³-´ and this will result in a zero DC component on the transmission line.  A unipolar and a NRZ bipolar are considered as on-off signals  May applied to either electrical or optical signals  Multi-level signal : several voltage levels are use

Fig. 5.6 Unipolar and bipolar coding

Coding Schemes 
RN and NRZ  For both methods, the signal alternates between a positive ( +V ) and a negative ( -V ) voltage  Logic ³1´ : positive voltage  Logic ³0´ : negative voltage  For NRZ, transitions between logic ³0´ and logic ³1´, and vice versa, are directly crossing the zero voltage level  For RZ : transitions between logic ³0´ and logic ³1´, and vice versa, stay temporarily on the zero voltage level

Fig. 5.7 RZ and NRZ coding

Coding Schemes 
4B/5B coding : translates 4 bits into one of the 16 predetermined 5-bit codes.  The original 4-bit code 0000 is translated to a 5-bit not-all-zero code.  Bipolar Violations ( BV )  Bipolar signal is a 3-level signal where consecutive 1¶s in the bit stream are alternating polarity  BV is when two consecutive µ1s do not change polarity  Can be used to detect errors in the bit stream  Or used to mark a specific bit manipulation ( coding ) in the bit stream  B8ZS ( bit 8-zero suppression ) : find 8 consecutive 0¶s in the bit stream and substitutes them with a bipolar violation  At the receiver, BV is detected and the bit stream is restored to its original form  HDB3 : substitueds four 0¶s by a code that contains a violation.  At the receiver, BV is detected and the bit stream is restored to its original form

Fig. 5.8 Bipolar and bipolar violation

5.1 Legacy Communications Systems: Concepts
Hierarchical Multiplexing

Multiplexing in North America 
24 DS0 signals + a framing F-bit p level 1 multiplexer ( M1 ) p a DS1 signal  4 DS1 signals + a control C-bit ( for every 48 payload bits ) p bit interleaved by a level 1-to-2 multiplexer ( M12 ) p a DS2 signal  7 DS2 signals + a control C-bit ( for every 84 payload bits ) p bit interleaved by a level 2-to-3 multiplexer ( M23 ) p a DS3 signal  Two-stage multiplexing : multiplexing from DS1 to DS3  DS2, although defined, are seldom used. Instead, an M13 multiplexer is used that multiplexes in one-stage 28 DS1 signals and all the F-bits, C-bits into a DS3 signal with bit rate of 44.736 Mbps.

Fig. 5.9 T-carrier hierarchy

Stuffing Bits 
A DS2 signal is formed by bit interleaving 4 DS1 signals  In a recurrent manner, a bit from each of the 4 DS1¶s is placed ina sequential order  But the four DS1 signal sources may be a slightly different bit rage then the expected 1.544 Mbps  Stuffing additional bits in specfic time slots of slower DS1 signal so that the composite DS2 signal is exactly 6.183 Mbps  Stuffing bits are also used when 7 DS2 signals are multiplexing into a DS3 signals

Fig. 5.10 DS2 frame ( 6.312 Mbps ) with Bit Stuffing

Fig. 5.11 DS3 frame ( 44.736 Mbps ) with Bit Stuffing

T1c Rate 
Another legacy system format.  2 DS1 signals and a control bit ( C-bit ) p a M11c bitinterleaved multiplexer p a DS1c signal  DS1c signal uses a duo-binary modulation and in effect doubles the bandwidth capacity of a T1 line

Fig. 5.12 T-carrier hierarchy

Multiplexing in Europe 
European systems time-division multiplex 30 user time slots and one ( framing and signaling ) time slot for every 15 user time slots to yield an E1 ( or E11 ) frame with 32 time slots.  The functions of the multiplexer are round-robin polling, byte interleaving, and time compression.  In the receiving end, an E1 signal is demultiplexed into 30 user channels and 2 added channels for framing and signaling and each channel is decompressed to 64 Kbps.  Each frame is transmitted within 125 Qs, so the total bit rate is 2.048 Mbps.  E1 is different from DS1 in bit rate and also the absence of the F-bit.  E1 signal uses an alternate mark inversion ( AMI ) with a high-density bipolar three-zero ( HDB3 ) techniques.

Fig. 5.13 E1 rate ( 2.048 Mbps )

Multiplexing in Europe 
30 DS0 channels + 2 64-Kbps channels ( for framing and signaling ) p time compressed and byte multiplexed by a level 1 multiplexer p an E1 signal  4 E1 signals + 4 64-Kbps channels ( for signaling ) p time compressed and bit interleaved by a level 1-to-2 multiplexer p an E2 signal ( 8.448 Mbps ).  4 E2 signals + 9 64-Kbps channels ( for signaling ) p time compressed and bit interleaved by a level 2-to-3 multiplexer p an E3 signal ( 34.368 Mbps ).

Fig. 5.14 European hierarchy

Table 5.4 Legacy Rates
Facility DS0 DS1 E1 DS1c DS2 E2 DS3 DS3c E3 DS4 274.176 Mbps 397.2 Mbps 44.736 Mbps 91.053 Mbps 97.728 Mbps 139.264 Mbps United States 64 Kbps 1.544 Mbps 3.152 Mbps 6.312 Mbps 8.448 Mbps 34.368 Mbps Europe 64 Kbps 2.048 Mbps 3.152 Mbps 6.312 Mbps 32.064 Mbps Japan 64 Kbps 1.544 Mbps

PROS and CONS of Legacy Networks 
Legacy networks were initially designed to carry voice signal from many sources to many destinations.  Switching functions play very important role in the legacy networks.  Due to the circuit-switching nature, time slots are assigned to a particular user for as long as the connection is alive, therefore the system is deterministic and its design simpler.  Pre-assigned time slots uses bandwidth inefficiently  So, legacy networks do not cost-effectively accommodate highbit-rate services

5.1 Legacy Communications Systems: Concepts
Legacy Data Networks

Legacy Data Networks 
Legacy synchronous communications networks do not address the requirements for voice and data applications in an equitable manner.  For voice communications, short delays through the network were the primary concern.  For data applications, the major concerns are high bit rates and low cost per unit of time.  For example, banks, stock-brokerage firms, publishing companies, image transmission.  Data will be transferred in chunks, up to about 9000 bytes, known as packets.  Such systems are local area network ( LAN ), metropolitan area network ( MAN ), wide area network ( WAN ), and the switched multimegabit data services ( SMDS ).

LAN 
Networks for a limited area ( building, campus ).  Many terminals were connected on a LAN backbone, and based on a LAN communications protocol, terminals were able to communicate among them.  Two major types : Ethernet and ring.  Ethernet LANs ( IEEE 802.1 ) : hierarchical and transmission rates is at 10 or 100 Mbps.  Ethernet 10BaseT : 10 Mbps over a twisted pair of wires  FDDI ( Fiber Distributed Data Interface ) ( IEEE 802.5 ) : a counter-rotating fiber ring LAN at 100 Mbps effective data rate.  FDDI employs 4B/5B coding and thus the actual bit rate is 125 Mbps.

MAN 
LANs were expanded to cover a city or a metropolis and the MAN is evolved  Traffic from one LAN was brought to another via a function known as bridge.  LAN/MAN traffic was routed on the communications networks so that connectivity between 2 or more remote LANs/MANs could be established.  Dedicated links can be deployed if high traffics between two LANs/MANs is sustained 24 hours a day.

SMDS 
A public MAN service developed by Bellcore  Primarily for LAN interconnections.  Based on the distributed queue dual bus ( DQDB )transport and multiplexing mechanism and is defined in IEEE 802.6.  The use of the DQDB format is based on a 53-byte cell structure that is similar to the ATM cell.  SMDS is a connectionless technology specified over synchronous carriers like DS1, DS3, E1 and E3 lines.

Frame Relay ( FR ) 
FR is a packet service.  Design philosophy is similar to that of a highway with ramps on and off.  At the ramps, any car, small or large, may approach the highway at a slower speed.  Once on the highway, one can travel any length of the highway and as fast as the limit.  So highway is equally accessible and shareable by all.  In FR, the ramps are the access points and the number of lances determines the bandwidth of the FR system.  FR is a shared-bandwidth solution that takes advantage of traffic pattern variability and oversubscription to provide a cost-effective data service.

Frame Relay ( FR ) 
At the access points, or the user-to-network interface, circuitry concentrates the packet traffic from a number of users, typically over leased lines ( T1 / E1 ).  The concentrated traffic is switched by means of a FR switch and it is put on a common backbone ( the highway ).  A number of FR switches are interconnected to form a FR network.

5.2 SONET and SDH

Introduction

Introduction 
The evolution from copper wires to optical fibers led to  Synchronous Optical Network ( SONET ) in US  Synchronous Digital Hierarchy ( SDH ) in Europe  Since first deployment in the 1980s, SONET / SDH have almost replaced all long-haul copper cables.  The optical fiber has responded to an unexpected increase in traffic demand and make ³ information superhighway ³ a reality

SONET / SDH 
SONET is a set of standard interfaces in an optical synchronous network of elements ( NE ) that conform to these interfaces.  SONET interfaces define all layers, from the physical to the application layer.  SONET is a synchronous network  SDH is also a synchronous network with optical interfaces.  SDH is a set of standard interfaces in a network of elements that conform to these interfaces.  Like SONET, SDH interfaces define all layers, from the physical to the application layer.

Fig. 5. 15 SONET / SDH services

SONET versus SDH 
Technical similarities between SONET and SDH  Bit rates and frame format organization  Frame synchronization schemes  Multiplexing and de-multiplexing rules  Error control  Major differences :  The variations about overhead bytes to accommodate differences between US and European communications nodes and networks  The SDH photonic interface specifies more parameters than SONET  SONET and SDH have enough minor technical and linguistic differences to add complexity ( and cost ) in their design

SONET versus SDH 
Some nomenclature examples  Synchronous transport signal ( STS ) versus synchronous transport module ( STM ), e.g., STS-1, STS-3, STS-12, STS48 versus STM-1, STM-4, STM-16.  Synchronous payload envelope ( SPE ) versus virtual container ( VC )  Virtual tributary ( VT ) versus tributary unit ( TU )

SONET / SDH Benefits 
Reduced cost:  It lowers operations cost  It has the same interface for all vendors  Integrated network elements  It allows for multi-vendor internetworking  It has enhanced network element management  Remote operations capabilities : provision, test, inventory, customize, and reconfigured  It offers network survivability features.  It is compatible with legacy and future networks

SONET and SDH Rates 
SONET and SDH rates are in the range of 51.85 ± 9953.28 Mbps  SONET signal in electrical nature : synchronous transport signal level N ( STM-N )  SDH signal in electrical nature : synchronous transport module level N ( STM-N )  In optical signal nature : optical carrier level N ( OC-N )

Table 5.5 SONET / SDH Rates

Why Use SONET / SDH ? 
Why is glass fiber better than copper wire ?  Higher transmission reliability.
‡ Glass fiber is not as susceptible to radio frequency or electromagnetic interference ( RFI, EMI ) as copper wire unless it is shielded and well grounded 

Lower bit error rate ( BER ) : lower inter-symbol interference among fibers than copper wires, so lower bit error rate.  Higher bandwidth per fiber  Fiber can transmit without repeaters at longer distance compared to copper wires.  Fiber yields thinner cable than copper.  SONET/SDH is based on standards, which enables multivendor compatibility and interoperability.

5.2 SONET and SDH

SONET/SDH Networks

SONET / SDH Network 
SONET / SDH network consists of nodes or network elements ( NEs )  Are interconnected with fiber cable  User and network information is transmitted over the fiber cable  SONET NEs receive signals from DS1, DS3, ATM, Internet, LAN/MAN/WAN, networks of different topologies such as rings, trees. For example, LAN at 10 Mbps, 100 Mbps or higher bit rates.  SDH signals may also be connected with a SONET and vice versa, to exchange information between two kinds of networks.

Fig. 5. 16 SONET network

Network Topologies 
Three major network topologies : tree, ring, and mesh  Tree  A hierarchical distribution of NEs  Most used in LANs, such as Ethernet  A source is connected to a distribution function ( known as hub ) that routes a source packet to its destination node.  A connection between source and destination is established for the duration of the packet through the hub.  Very efficient for asynchronous data transmission but not for real-time data and voice.  If a hub fails, all connections through it die.

Network Topologies 
Ring : consists of NEs interconnected with a dual fiber, the primary and secondary, to form a ring.  Some of the NEs are also assigned to communication with other rings or topologies  When one fiber fails, the other can still keep the integrity of the ring  If both fiber breaks, then the network is reconfigured, forming a ring using both the primary and secondary and the packets will flow through all fibers but the broken ones.  Ring offers fast path protection and is widely used in LANs or within a relatively limited radius ( campus, town, high risers).

Network Topologies 
Mesh : consists of NEs fully interconnected  When one link breaks, the adjacent NE detects the breakage and reroutes the traffic to another NE.  Mesh provides transmission protection and network restoration capabilities.  Mesh also provides disaster avoidance capabilities when a cluster of NEs may fail.  Better applicable in densely populated areas.  SONET / SDH networks are based on ring topologies  Other topologies may be mixed into networks as well.

Fig. 5. 17 Ring, tree and mesh topologies

A Hierarchical Process 

1. 2. 3. 4. 

A hierarchical process can transformed any type of nonSONET signal into SONET networks. Segmenting the signal and mapping the segments in small containers known as virtual tributaries ( VTs ). Once the VTs have been filled with segmented payload, they are grouped in larger containers that are know as groups. Groups are then mapped into a SONET frame. Many contiguous frames entails the SONET signals to be transmitted over the OC-N fiber. In SDH, VT are called tributary units ( TUs ) and the groups are called tributary unit groups ( TUGs ).

Fig. 5. 18 SONET hierarchy

Table 5.6 Broadband Services and Rates
SONET VT1.5 VT2 VT3 VT6 STS-1 STS-3 STS-12 STS-48 STS-192 STM-0 STM-1 STM-4 STM-16 STM-64 SDH Bit Rate ( Mbps Sample Services ) 1.728 Voice / high-speed digital services 2.304 High-capacity digital services 3.456 6.912 51.840 155.520 SMDS, broadband-ATM, high-definition TV ( HDTV ) 622.080 Uncompressed Extended Quality TV 2488.320 Uncompressed HDTV 9953.280

Path, Line and Section 
A SONET / SDH frame is transmitted from an end user to another end user through nodes in the network.  To ensure the deliverability and the integrity of the signal, overheads are added to the sending signal to be used for network administration purposes.  This overhead information is transparent to end users.  The overhead has been organized hierarchically in three types :  Path  Line  Section

Path, Line and Section 
Path : overhead added at transmitting path-terminating equipment ( PTE ), and it is read by the receiving PTE.  Path information is not checked or altered by intermediate equipment  Line : overhead added by the transmitting line-terminating equipment ( LTE ) to be used by the receiving LTE.  At the edges of the networks, ( where there are no LTEs, ) PTEs play the role of LTEs  Section : overhead added by equipment terminating a physical segment of the transmission facility.  A segment between two repeaters  A segment between an LTE and a repeater  A segment between an PTE and a repeater  A segment between two adjacent LTEs where no repeaters between them.  All these are sections.

Fig. 5.19 Path/Line/Section definitions

Fig. 5.20 Path/Line/Section definitions

5.2 SONET and SDH

SONET/SDH Frames

STS-1 
The smallest SONET frame  Can be visualized as a 2-dimensional matrix of 9 rows by 90 columns, each entry is a byte.  The whole STS-1 has 9 x 90 = 810 bytes  The first 3 columns : transport overhead for section and line.  Synchronous Payload Envelope ( SPE ) : the byte capacity contained from the 4th column to the last column, totally 87 columns or 783 bytes.  Path overhead : the 4th column.  Fixed stuff : columns 30 and 59 of the STS-1.  The actual payload capacity is 84 columns, i.e., 756 bytes or 48.384 Mbps effective bit rate.

Fig. 5.21 SONET STS-1 frame structure

SDH AU-3 
SDH does not specify a frame similar to SONET STS-1.  It has a payload container as small as the SONET SPE, know as virtual container 3 ( VC-3 ).  VC-3 has a column for path overhead, called the VC-3 path overhead ( VC-3 POH ), and two fixed stuff columns ( columns 30, 59 of VC-3 ).  The actual payload capacity in a VC-3 is also 84 columns or 756 bytes, similar to the SONET case.  At the VC-3 and in the 4th row, three additional bytes are added for the VC-3 pointer ( H1, H2, and H3 ).  Administrative unit level 3 ( AU-3 ) : VC-3 + VC-3 pointer  AUG ( administrative unit group ) : 3 AU-3s are bytemultiplexed

Fig. 5.22 SDH AU-3 frame structure

Transmitting an STS-1 
  

To transmit an STS-1 over optical fibers, one starts with the most significant bit ( MSB ) of the byte in column 1, row 1. Then when the byte is serially transmitted, it continues with the byte in column 2, row 1; and so on. After the first row is transmitted, the process will transmit the 2nd row, and so on. In 125 Qs, all 6480 bits are transmitted, so the bit rate is 51.86 Mbps.

Fig. 5.23 SONET STS-1 frame: unfolded

Floating Frames 
SPE = user data + path overhead  STS-1 = SPE + line and section overhead  When an STS-1 is received, the beginning of the frame is not necessary synchronous with the beginning of the frame that the receiving node generates.  The receiving node may need additional time to process the overhead information.  If the SPE waits for receiving node to synchronized, will suffer added delays, which is bad for a high-speed transmission network.  So , the ³floating´ SPE technique is used to minimized delays.

Fig. 5.24 STS-1 frame ± SPE unfolded

Floating Frames 
The process of an SPE is separated from the process of the line and section overhead of an STS-1.  The received SPE may be out of phase ( in increments of bytes ) with the beginning of the STS-1 frame, which is in synchronization with the NE.  This ³out of phase´ can be viewed as an offset in the STS-1 frame, in terms of columns and rows.  Thus, the first byte of the receiving SPE ( this will be the first byte of the path overhead ) is mapped in the SPE of the current STS-1 frame on row N and column K, for example.  As more bytes are received, the received SPE is wrapped around the STS-1 SPE space.

Floating Frames 
Resulting path overhead will be aligned in the Kth column.  Due to the offset, not all bytes of the received SPE will be able to fit in the SPE space of the current STS-1 frame.  And the rest of the SPE will be mapped to the next STS-1 SPE and each SPE are synchronous with the 125-Qs interval.  The end result will be an SPE mapped over two consecutive frames.

SONET STS-1 with floating S

Mapping a floating ( partial ) SPE in an STS1 frame

Mapping a floating ( complete ) SPE in STS1

Section Overhead : SONET 
The first 3 rows of the overhead space in an STS-1 frame, totally 9 bytes, carry synchronization and section overhead information.  A1 and A2 are fixed patterns : 0xF628 or in binary 1111 0110 0010 1000. Receiver will use these to detect the beginning of the frame.  A1, A2 are not scrambled.  C1 : STS-1 ID, defined for each STS-1.  B1 : used for error monitoring.  E1 : a 64-Kbps voice communication channel for craft personnel.  In an STS-N signal, E1 is defined for the 1st STS-1 only. The other N-1 E1¶s are not used.  F1 : used by the section.  D1 to D3 : a 192-Kbps communication channel between STEs.  Used for alarms, maintenance, control, monitoring, administration, and other communicaiton needs.  In an STS-N signal, this channel is defined for the 1st STS-1 only. The other N-1 E1¶s are not used.

STS-1 section overhead

Line Overhead : SONET 
Located in the row 4-9 of the overhead section, totally 45 bytes.  H1, H2 :offset between the pointer and the first SPE byte.  H3 : an action byte for frequency justification.  It carries valid payload if the justification is negative.  BIP-8 : Used for locating errors.  It is calculated using even parity over the STS-1 for the previous frame after scrambling and placed in B2 before scrambling the current frame.  K1, K2 : for automatic protection switching.  In STS-N, this is defined for #1 only.

STS-1 line overhead

Line Overhead : SONET 
D4 to D12 : a 576-Kbps communication channel between LTE for  Alarms  Maintenance  Control  Monitoring  Administration  Other communication needs  In STS-N, this is defined for #1 only  Z1, Z2 : not defined except in STS-N for #3, in which Z2 is only defined as line far-end block error ( FEBE ).  E2 : an express 64-Kbps communications channel between LTE  In STS-N, this is defined for #1 only

Payload Pointers : H1, H2, H3 
NDF ( New Data Found flag ) : first 4 MSB in the H1 byte  ³ normal = 0110 ³ for the following 3 possible situations :
‡ No frequency justification ‡ Positive frequency justification has taken place ‡ Negative frequency justification has taken place 

³ set = 1001 ³: an arbitrary ( and significant ) change of the pointer vaule has occurred due to a change of data position in the SPE  S-bits : the size of the virtual tributary in the payload  I-bit / D-bit word : for incrementing / decrementing the offset  Perform frequency justification in conjunction with the H3 byte  H3 : a payload byte opportunity buffer when positive or negative justifications are necessary

Payload pointers : H1 and H2

Pointer H3 ± frequency justification

Functions of H1, H2 and H3 
Identifies that a change has occurred in the pointer value ( NDF = 1001 ) due to an intermittent synchronization in the node where the new start is ( I + D bits ).  Identifies that a change may have occurred in the pointer value ( 0110 ) due to a frequency difference between node and incoming frequency.  Received frequency may be slightly higher or lower than the node frequency.  Either more or fewer bits are received that what the SPE can fit.  If the received frequency is higher ( lower ), then node will perform a negative ( positive ) frequency justification.  The I-bit / D-bit words value will indicate whether negative or positive frequency justification is needed.

Frequency Justification : No Justification 
The frame rate of the STS SPE = the frame rate of transport overhead ( OH; i.e., NE frame rate)  Then the alignment of the SPE is the same as the previous frame.  NDF = 0110; I-bit / D-bit words are not inverted.  If pointer value = 0, the first byte of the data is located next to the H3 byte;  If pointer value = 10, the first byte starts on the 11th byte after H3;  Example  NDF = 0110  I, D value : 00 0010 1001 = 41  H3 = 00000000

STS-1: no frequency justification

Frequency Justification : Positive Justification 
Frame rate of the STS SPE < the transport OH ( i.e. NE frame rate )  The alignment of the SPE is slipped back by a byte.  H1, H2 incremtns and I-bits are inverted.  NDF = 0110  Example :  I, D : value 00 0010 1010 = 40  First byte in SPE next to the H3 is 0  Positive justification is recognized quickly by examining the bit next to the LSB of the H1 byte
‡ 1 : positive ‡ 0 : negative

STS-1 : positive-frequency justification

Frequency Justification : Negative Justification 
The frame rate of the STS SPE > the transport OH (i.e., NE frame rate ).  The alignment of the SPE is slipped back by a byte  H1, H2 decrements  D-bit are inverted, NDF = 0110  Example :  The I, D value is 00 0010 1000 = 40  H3 byte contains user payload in the current frame.

STS-1 : negative frequency justification

New Data Found ( example ) 
NDF = 1001  A resynchronization has taken place ( not due to a positive or negative frequency justification but because of some system or payload-type change .  I-bit and D-bit are not inverted  New pointer value in H1 and H2  For another case, known as concatenation indicaiton :  NDF=1001  I-bit and D-bit = 11 1111 1111

STS-1 : new data found

Path Overhead 
The first column of the SPE ( 9 bytes )  Path bytes are directional.  J1 : the trace byte , is user programmable  The receiving PTE collects 64 repeating J1 bytes to verify the connectivity with the transmitting PTE.  Default : 0x00  BIP-8 or B3 : error control.  C2 : indicates the construction of SPE, the asynchronous mapping, ATM, etc.  G1 : path status to the originating PTE from the destination PTE.  F2 : for end-user communication use.  H4 : multiframe, is used as an end-to-end generalized multiframe indicator for payloads.  Z3, Z4, Z5 : user bytes, reserved for future use and undefined for now.

SONET STS-1 : path overhead

SDH VC-3/4 : path overhead

5.2 SONET and SDH

Virtual Tributaries

Virtual Tributaries 
VT ( virtual tributary ) : small container that is used to transport user payloads.  In SDH, they are called VC ( virtual containers ).  Each VT arrives within a 125 Qs interval.  Several VTs with OH ( overhead ) will constitute an SPE.

Fig. 5.38 VTs : what are they ?

VT Capacities 
   

VT 1.5 : a 3-column capacity ( 27 bytes ). VT2 : a 4-column capacity ( 36 bytes ). VT3 : a 6-column capacity ( 54 bytes ). VT6 : a 12-column capacity ( 108 bytes ). In SDH :  VT1.5 : TU-11.  VT2 : TU-12.  VT6 : TU-2.

Table 5.7 Virtual Tributaries and Payload Rates
VT Type Columns/VT Bytes/VT VTs/Group VTs/SPE VT payload rates (Mbps) 1.728

VT1.5

3

27

4

28

VT2

4

36

3

21

2.304

VT3

6

54

2

14

3.456

VT6

12

108

1

7

6.912

Fig. 5.39 Virtual tributaries

SPE Groups 
An STS-1 SPE contains 87 columns, the real user payload is 84 columns.  The payload is partitioned in 7 groups, each group has 12 columns in capacity.  Each group can fit in 4 VT1.5 or 3 VT2 or 2 VT3 or 1 VT4.  Different groups can have different types of VTs  Each group can have only one type of VT.  In SDH, groups are called TUG-2 ( tributary unit group ).  7 TUG-2 byte-multiplexed and add two fixed-stuff columns p 1 TUG-3.  7 TUG-3 byte-multiplexed and add one path overhead column p 1 VC-3 ( virtual container level 3 ).

SPE Groups 
  

When VTs form one SPE group, they are byte-interleaved. Each SPE group can have only one kind of VT. The 7 SPE groups form SPE by byte ( or column ) interleaving The fixed stuffs columns and the path overhead column are also byte interleaved to construct the SPE.

Fig. 5.40 SONET ± VTs and group structures tributaries

Fig. 5.41 SONET ± groups and SPE structures

SDH Groups 
TU filled with user data are byte (or column ) interleaved to construct TUG-2.  Each TUG-2 can have only one type of TU.  When all 7 TUGs are constructed, they are byte ( or column ) interleaved and add two columns of fixed stuff to construct TUG-3.

Fig. 5.42 SDH ± TU and TUG structures

Fig. 5.43 SDH -- TUGs

Fig. 5.44 VT1.5 mapping in SPE

VT Superframe 
VTs map data from different end users in the SPE payload.  End user data may also be out of phase with each other and to the STS-N frame.  A pointer and a frequency justification scheme, similar to the SPE level, applies on the VT level.  Each VT is partitioned in a 1-byte VT overhead part and in a VT envelope capacity ( depending on the type of VT ).  VT1.5 : 26 bytes VT envelope capacity  VT6 : 107 bytes VT envelope capacity  4 consecutive VTs from one user¶s data form a VT superframe.  H4 : indicates the beginning of the superframe ( byte overhead V1 )

VT Superframe 
Payloads below DS3 rates are mapped in and trasported by VT structures.  H4 in the SPE path overhead is used to indicate the phase of V1V4 ( bits B7 and B8 ) as well as 6-, 16-, or 24-frame superframe indication ( bits B1 and B2 or B3 and B4 or B5 and B6 )  To comply with Bellcore ( AR-254-CORE ), the following rules apply to H4 bits :
‡ X is set to ³1´ or ‡ Bits B1 and B2 count from 00 through 11 over 24 frames or ‡ Bits B3 and B4 count from 00 through 01 over 6 frames and then repeat or ‡ Bits B5 and B6 countr from 00 through 11 over 16 frames and then repeat the counting process

Fig. 5.45 VT superframe

Fig. 5.46 VT superframe ± H4 byte

H4 BYTE 12345678 XXXXXX 0 0 XXXXXX 0 1 XXXXXX 1 0 XXXXXX 1 1

V1 ± V4

V1 V2 V3 V4

VT Payload Pointer 
VT payload pointer is analogous to the STS payload pointer.  Provides a mechanism for flexible and dynamic alignment of the VT envelope within the VT superframe  VT payload pointer : V1 and V2  Viewed as one word  Obtained in 250 Qs.  First four bits of V1 are N-bits, indicating whether there are new data occurred or not.  Next two bits define the VT size  Remaining 10 bits are I-bit and D-bit words
‡ Point out the beginning of the user frame in the VT superframe and whether a positive, negative, or no frequency justification should be made.

Fig. 5.47 VT payload pointer

Fig. 5.48 VT payload pointer -- NDF

Fig. 5.49 VT payload pointer -- value

VT Payload Pointer 
Frequency adjustment is made by slipping by a byte forwand or back in time the VT envelope.  Similar to the STS SPE frequency justifications.  The increment or decrement amount is indicated by inverting the I-bit or D-bit word, respectively.  When the frame rate of the VT envelope is greater than or less than that of the STS SPE, the alignment of the VT envelope is periodically slipped forward or backward in time by 1 byte shift ( negative or positive stuff bye ) and the pointer value is adjusted by 1.

VT Overhead 
The first byte in the envelope.  Over four frames, or 500 Qs, the four overhead bytes are V5, J2, Z6 and Z7.  The remaining bytes in a VT are user payload.  For VT1.5, the  user payload is 25 bytes.  For VT6, the user payload is 106 bytes.  Each VT envelope contains four bytes of VT POH ( V5, J2, Z6 and Z7 ), the remaining bytes are VT payload capacity.

Fig. 5.50 Virtual tributaries overhead

VT Overhead 
V5 : the first POH byte in the VT superframe  Error checking as a signal label  Indicating the signal status  Bits in V5 are :  BIP-2 (1) : set if parity of all odd-numbered bits of previous VT SPE is even  BIB-2 (2) : set if parity of all eveb-numbered bits of previous VT SPE is even  REI-V : VT remote error indication back to originating VT PTE.  RFI-V : VT path remote failure indication in byte-synchronous DS1 mapping.  Signal label : indicates content of the VT SPE.  RDI-V : VT path remote defect indication.

Fig. 5.51 Virtual tributaries path overhead

Application : DS0 byte synchronous Mapping For DS1 
In a DS1 frame, there are 24 time slots (bytes ).  Mapping one DS1 in a VT1.5.  User payload in a VT1.5 has 25 bytes, the first byte has  The phase of the signaling and the frame bits (bits P0 and P1)  Signaling for the 24 DS0 channels ( bits S1-S4 )  The framing bit ( bit F )  A bit not used ( or fixed stuff )  The remaining 24 bytes map the 24 bytes of the DS1 frame.

Fig. 5.52 VT1.5 DS0 byte ±synchronous mapping in Ds1

Application : DS0 Byte-Asynchronous Mapping for DS1 
A DS1 frame consists of 24 time slots  User payload in a VT1.5 has 25 bytes, the first byte has  Fixed stuff ( bits R )  Information bits ( bit 1 )  Stuff control bits ( bits C1 and C2 )  Overhead bits ( bit O )  Stuff opportunity ( bits S )  The remaining 24 bytes map the 24 bytes of the DS1 frame.

Fig. 5.53 VT1.5 DS0 byte ±synchronous mapping in Ds1

5.2 SONET and SDH

STS-N/STM-N Frames

Single-Step Multiplexer 
A multiplexer that receives N STS-1 SPEs within a time window and then in one single stage to multiplex them to produce an STS-N.  The basic building block ( signal ) for STS-N frames are STS-1 SPE when N > 1.  Multiplexing is on the byte level.

Fig. 5.54 STS-N single-stage interleaving.

Two-Step Multiplexer 
Two-step multiplexing : K STS-1 SPEs may be received and multiplexed in an STS-K and then N/K STS-Ks multiplexed to produce an STS-N.  Multiplexing is on the byte level.  Two-step multiplexing is more complicated than one-step multiplexing.  Two-step multiplexing also creates more delay ( because there are more steps in multiplexing ) than one-step multiplexing.

Fig. 5.55 STS-N two-stage interleaving.

STS-N Frame Structure 
   

Each STS-N frame has Nx90 columns and 9 rows. Overhead ( line, section ) : the first Nx3 columns Each STS-N SPE has Nx783 bytes The first N columns of an STS-N SPE are the path overhead For example  For an STS-3 frame, has a total of 3x90 = 270 columns by 9 rows.  Line & section overhead are the first 3x3 = 9 columns  Each STS-3 SPE has 3x783 = 2249 bytes  The first 3 columns of an STS-3 SPE are path overhead.

Fig. 5.56 STS-N frame structure.

STS-N Floating Frame 
The component STS-1 SPE in an STS-N frame may be floating.  Each component STS-1 SPE may start from a different location in the STS-N SPE.  The pointers H1, H2 and H3 for each component STS-1 SPE are still in the STS-N section overhead and each identifies the starting location of its corresponding STS-1 SPE as well as makes frequency adjustments individually.

Fig. 5.57 STS-N example, N = 3.

Concatenation or Super-Rate 
For example of STS-3  If 3 STS-1 SPEs are from different sources and each one may be in different phases with each other and with the node, each will keep its own ponter H1-H3 and frequency justification.  If 3 SPS-1 SPEs are belong to the SAME signal source, and they arrive in a specific order and in synchronism ( which must be maintained throughout the end-to-end path .), these will be processed into concatenated SPEs.  The 3 SPEs are mapped in a known sequence and thus only one overhead pointer is required to point to the first SPE.  The remaining pointers contain a fixed code set to a value that means concatenation indication ( CI ).

Concatenation or Super-Rate 
Also, only one path overhead is needed.  At the receiving end, all 3 SPEs are received in the same order they were transmitted.  The STS that transports concatenated SPEs is called concatenated STS, denoted by STS-Nc ( in this case, STS-3c ).  Similar case for N > 3.

Fig. 5.58 STS-Nc example, N = 3.

Contiguous Concatenation in SDH 
Allows bit rates in excess of the capacity of the C-4 container.  Concatenation implies that payload should not be split up  A virtually contiguous container within an STM-4 ( for example ).  Then the payload of several consecutive AU-4¶s are mapped in a fixed order.  The first pointer is set to its value and the rest pointers are set to a fixed CI value.  This VC-4 is identified as a VC-4-4c.

Virtual Concatenation 
Some network (switching) nodes are not capable of switching complete STS-3c or VC-4-4c with contiguously concatenated payload in them.  Then these payload will be split into payloads of lower rate, like STS-1 and each STS-1 will be switched independently.  The responsibility of reconstruct the STS-1¶s back to the original STS-3c is on the network elements.

STM-N Frame Structure 
      Section overhead : N x 9 columns Payload : N x 261 columns. The smallest STM ( STM-1 ) has 270 columns ( 9 overhead + 261 payload ). The first three rows of the first three columns of the SOH are the regenerator section overhead ( RSOH ). The 4th row of SOH is the administrative unit pointer ( SUP ). The remaining 5 rows are the multiplex section overhead ( MSOH ). Comparison between SDH and SONET :  SOH vs. section overhead.  AU pointer vs. SPE pointers.  MSHO vs. line overhead Constructing an STM-N : 3 VC-12 p 1 VC-2; 7 VC-2¶s p 1 VC-3; 3 VC3¶s p 1 VC-4; N VC-4¶s p 1 STM-N 

Fig. 5.59 STM-N frame structure.

STS-Nc Frame Structure 
The major difference between STS-N and STS-Nc payload : then number of path overhead column in the SPE.  For STS-N : N columns  For STS-Nc : 1 column  The difference in the fixed-stuff columns  For STS-N : 2N columns  For STS-Nc : N/3 -1 columns  STS-Nc has more bandwidth for dedicated user payload and more efficient.

Fig. 5.60 STS-Nc frame structure.

Pointer Processing for STS-Nc 
The pointer H1-H3 of the first STS-1 in the STS-Nc will be the pointer for the whole SPE.  Pointer processing  New data flag  Frequency justifications  Identify the starting location of the STS-Nc SPE  The rest of the pointer bytes will contain a concatenation indicator value to indicate that the payload is concatenated ( i.e., it is an STS-Nc case ).  H1 : 1001XX11 ( X : undefined )  H2 : 11111111

Fig. 5.61 STS-3c transport overhead.

Transport Overhead : SOH for SDH 
A1, A2 : frame alignment; fixed framing pattern set at the hexadecimal value 0xF628 [ 1111 0110 0010 1000 ].  A1, A2 are NOT scrambled.  B1, B2 : parity bytes for quality monitoring ( or error monitoring in SONET ).  The parity is calculated over ALL bytes of the previous frame before scrambling and is placed in the current frame before scrambling.  D1 ± D3 : for network management for the regenerator section  An 192 Kbps communication channels for alarms, maintenance, control, monitoring, administration, and other needs.  D4 ± D12 : for network management as the D1 ± D3 but for the multiplex section.  E1, E2 : a 64 Kbps voice communication channel for craft personnel.  F1 : a maintenance byte  J0 ( C1 ) : trace identifier

Transport Overhead : SOH for SDH 
K1, K2 : for automatic protection switching ( APS ) control  S1 : clock quality indicator  M1 : transmission error acknowledgement

Fig. 5.62 STM-1 section overhead.

Scrambling 
When the complete frame has been assembled, the bytes in it are scrambled.  Purpose : to assure the receiver that a density of 1¶s is maintained in the signal.  A1, A2, C1 bytes are not scrambled  A1 and A2 are used to locate the beginning of the frame, so no scrambling are applied

Fig. 5.63 Frame scrambled.

STS-N Scrambler 
The scrambling code is generated by the polynomial 1 + x6 + x7  The scrambler is frame synchronous at the line rate ( STS-N ), and it has a sequence length of 127 bits that repeats itself.  The scrambler is set to 1111111 on the most significant bit ( MSB ) of the byte following the Nth STS-1 C1 byte ( of an STSN)  The framing bytes A1, A2 and C1 from the first STS-1 through the Nth STS-1 are NOT scrambled.  The scrambler runs continuously throughout the complete STSN frame.

Fig. 5.64 STS-N scrambler.

Layered Overhead and Transport Function 
How to transport traditional signals, such as DS1, E1, etc. with a SONET signal ?  To transport a DS1 signal to a SONET signal :  The incoming DS1 signal at the path layer is mapped onto a VT.  The VT is mapped onto the SPE, and the SPE path overhead is also constructed.  The SPE is mapped onto the SONET signal, and the line overhead information is added.  The signal is mapped onto the STS-N signal, and the section overhead information is added.
‡ Now the complete STS SONET signal is formed, and the signal is scrambled. 

The signal passed through the electrical-to-optical transducer ( the transmitter ), and the optical signal ( with an NRZ optical coding ) is coupled into the optical fiber and transmitted at the speed of light.

Fig. 5.65 Conversion of a legacy signal to a SONET, layered overhead and transport functions.

Interaction Between Layers 
SONET signal is transmitted between NEs  Similar layers of the SONET signal, as viewed between the transmitting and receiving NEs, interact at the corresponding layers.

Fig. 5.66 Example²from DS3 on OC-N, layered overhead and transport functions.

Fig. 5.67 Interaction between layers, layered overhead and transport functions.

SDH Multiplexing 
SDH follows a very similar hierarchical multiplexing scheme to SONET  Traditional signals are mapped onto containers, the containers are multiplexed into groups, and overhead is added to construct an STM-N signal. ( CCITT Recommendation G.907 )

Fig. 5.68 SONET multiplexing scheme.

Fig. 5.69 SDH multiplexing scheme.

5.2 SONET and SDH

Synchronization and Timing

Network Synchronization 
On a network, from a receiving node point of reference, all signals are received at different phases with its clock and frame synchronization.  The phase difference could be easily engineered if it would remian fixed or bounded;  Received signals can be aligned if the phase difference is fixed.  A fixed phase also implies all nodes are exactly operate at the same frequency.  But NEs are not exactly using the SAME frequency due to NEs have different built-in clocks.

Network Synchronization 
In practical, a small difference between the clocks from node to node may result in a substantial frequency variation as the signal passed through many nodes.  The problem can be solved if a common timing source can be established  The network uses a frequency of extremely high accuracy from a timing service known as Building Information Timing Supply ( BITS ) in North America  In Europe, the similar standard to BITS is the primary reference Source ( PRS).

Fig. 5.70 Network synchronization²a model.

Timing Accuracy : Strata 
The timing architecture in the communications network is hierarchical.  Each node in the network is required to main a frequency accuracy and depends on its major function in the network for timing accuracy ( if other nodes in the network depend on it for timing accuracy).  The highest possible frequency accuracy is 10 x 10-11  Highly accurate atomic clocks  The clock will be the PRS and the accuracy is referred as stratum 1.  Stratum 1 accuracy allows a slip up to 2.523 periods per year

Timing Accuracy : Strata 
Stratum 2 : 1.6 x 10-8  Stratum 3 : 4.6 x 10-6  Stratum 4 : 3.2 x 10-5  For example, the digital PRXs, central office terminals, and digital channel banks.  In SDH, PRS is called primary reference clock ( PRC )  Accuracy 1.0 x 10-11  The clock is distributed throughout the entire network  Regenerated at the NEs by a synchronization supply unit ( SSU ) and also by a synchronous equipment clock ( SEC ).

Table 5.8 Timing Accuracy
Stratum Minimum Accuracy 10×10-11 Skip Rate Notes

1

2.523 / year

PRS

2

1.6×10-8

11.06 / day

e.g., 4ESS/5ESS

3

4.6×10-6

132.48 / hour

e.g. 5ESS/DCS

4

3.2×10-5

15.36 / minute

DCB/COT/DPBX

Timing Stability 
Since signal may pass through parts of network owned by more than one provider  SONET NEs must be synchronized with a stratum 3 or better quality clock.  Or, SONET NEs can equipped with an internal clock ( oscillator ) that has a minimum accuracy of ±20 parts per million ( ppm ).  This is required to support operation, administration, maintenance, and provisioning ( OAM&P ) functionality of all nodes in the network.  Network providers must also supply a timing reference signal that meets the recommended stability requirements.

NE Synchronization 
An NE should be synchronized with the network.  The strategy of synchronization has several models.  When model BITS or SASE is available, NEs are externally timed from the reference clock.  When BITS is not available, NEs are timed from a received OC-N signal.

BITS Timing Model 
Where BITS ( or SASE ) is available, the NEs are externally timed from it.  The same quality reference clock over two paths, the active and the alternate, are supplied by BITS.  The NE receives both signals, but it operates from the active path.  If the active path goes down, the NE selects the alternate clock.  NE¶s time unit ( TU )
‡ ‡ ‡ ‡ Receiving the two clocks from BITS Monitoring the incoming clocks for faults Selecting the correct ones Distributes the clock to all functional units in the NE

Fig. 5.71 BITS network synchronization.

OC-N Single DS1 Line Timing Model 
The NE extracts the clock from an incoming signal, typically a DS1.  The incoming OC-N is used to extract an 8-KHz reference clock.  A DS1 frame occurs 8000 times per second and that DS1 frames are mapped in VT1.5¶s.  The extracted 8 KHz is sent to the TU, which sends it over two separate paths to the BITS  BITS retimed and then sent it back to NE over the active and alternate paths  The TU distributes the retimed reference clock to all functional units in the NE

Fig. 5.72 OC-N Line (DS1) timing model.

OC-M Single DS1 Line Timing Model 
The NE extracts an 8-KHz clock from an incoming signal, typically a DS1 that is embedded in an incoming OC-M ( N > M) signal.  The extracted 8 KHz is sent to TU, then sent to BITS over two separate paths.  The BITS retimed and sent it back to the NE over the active and alternate paths.  The TU distributes the retimed reference clock to all functional units in the NE.

Fig. 5.73 OC-M Line (DS1) timing model.

OC-N Dual-DS1 Line Timing Model 
The NE extracts two independent clocks from two OC-N incoming signals.  The two incoming OC-Ns are used to extract from DS1¶s two 8KHz synchronization signals.  The extracted 8 KHz is sent to TU, then sent to BITS over two separate paths.  The BITS retimed and sent it back to the NE over the active and alternate paths.  The TU distributes the retimed reference clock to all functional units in the NE.

Fig. 5.74 OC-N Line (two DS1) timing model.

Loop Timing Model 
The NE is not connected to a BITS.  The NE is connected to only one OC-N and to lower bit rate OC-M ( N > M ) and DS-N.  The NE extracts the clock from incoming signals of the loop, typically a DS1.  An incoming OC-N is used to extract an 8-KHz synchronization signal.  The extracted 8 KHz is sent to TU, .  The TU retimed the clock based on an accurate phase-locked loop oscillator, located on the TU and then distributes it to all functional units in the NE.

Fig. 5.75 Loop timing model.

Through-Timing Model 
This model applies to regenerator elements.  Each direction has its own TU and timing is extracted from each direction.  This model is not recommended for add-drop multiplexers.

Fig. 5.76 Through-timing model.

Clock Applications 
The minimum free-running accuracy of a SONET minimum clock ( SMC ) shall be ±20 ppm.  An ADM could be timed from a clock extracted from a line or from an external clock sources, or though-timed.  A holdover : when the reference clock fails, then for a specified limited period of time, it enters a free-running state with a specified tolerance.  A switched digital module in terminal mode ( TM ) configuration and looptimed may be required to provide holdover.  A digital cross-connect system with external timing is required with a minimum accuracy of ±4.6 ppm.  Entry into holdover and restoration from holdover shall be error free.

5.2 SONET and SDH

Maintenance

Maintenance 
Anomaly : a discrepancy between the actual and desired characteristics of an item.  Defect : a limited interruption in the ability of an item to perform a required function.  Failure : a persistent defect. An NE enters a failure state when a failure condition has been detected, and it exits a failure state when the condition has been corrected.  Maintenance of a system defines  The functions to be monitored  The requirements  The criteria  Their parameters

Maintenance 
Performance of a function :  Acceptable  Unacceptable ( failed )  Degraded  Main objectives of maintenance :  Trouble detection  Trouble sectionalization  Trouble or repair verification  Trouble isolation  Restoration

Maintenance 
Maintenance requirements  Alarm surveillance  Performance monitoring ( PM )  Control

Alarm Surveillance 
Alarm surveillance deals with the detection and reporting of certain failures and degraded conditions in the network.  NE alarm surveillance takes place on different levels, at :  Section terminating equipment ( STE )  Line terminating equipment ( LTE )  STS path terminating equipment ( STS PTE )  VT path terminating equipment ( VT PTE )  A VT PTE contains all functionality of an STE, LTE and STS PTE.  An STS PTE has all the functionality of an LTE and STE.

Fig. 5.77 Maintenance: layered alarm surveillance.

Alarm Indications 
A change of state may cause an alarm indication immediately as it occurs or after some period of time.  When a failure state is detected, an alarm indication signal ( AIS ) is constructed and it is sent to the next NE on the downstream path.  AIS is to alert downstream equipment of an upstream failure.  The response to a received AIS is a remote defect indication ( RDI ) signal in the upstream direction  AIS and RDI status signals are communicated to the far end over overhead bytes in the SONET/SDH signal

Alarm States, Declarations, and Indications 
Loss of signal ( LOS ) : no light pulses persist for 100 Qs.  If light pulses are lost for less than 2.3 Qs, then no LOS is declared.  However, if they persist for a period longer than 2.5 sec, then an alarm message is sent to the operating system ( OS ).  Loss of frame ( LOF ) : a severely erroneous frame ( SEF ) persists for a period longer than 3 ms.  If at least four consecutive frameshave incorrect framing patterns, then an alarm message is sent to the OS.  Loss of pointer ( LOP ) : pointer failures occur, such as out of range, wrong NDF, or any other failures related to the pointer processing mechanism. An alarm is sent to OS.

Alarm States, Declarations, and Indications 
Equipment failures :  Service affecting ( SA )  Non-service affecting ( NSA )  Classified as critical, major, and minor.  When equipment failure is detected, a message is sent to OS.  Loss of synchronization : loss of primary or secondary timing reference. A message is sent to the OS stating the reason ( LOS, LOF, etc )  Automatic protection switching ( APS ) troubles : related to channel mismatch, K1 with unused code, K2 with APS mode mismatch, etc.

Alarm States, Declarations, and Indications 
DCC failure : any failed hardware or failure to carry a data communications channel ( DCC ).  When this occurs, then the switch moves to standby DCC and a report is sent to the OS.  Signal label mismatch : related to signal labels.  Payload label mismatch ( PLM )  Unequipped ( ENEQ )  Signal label mismatch is performed by monitoring the C2 byte for STS signal labels and for STS payload label mismatch ( PLM-P).  Also performed by monitoring the V5 byte for VT PLM and VT path unequipped.

Alarm States, Declarations, and Indications 
AIS alerts the downstream equipment that a defect or a failure has been detected.  Can be generated for
‡ Line ( AIS-L ) ‡ Path ( AIS-P) ‡ VT path ( AIS-V) 

Remote alarm indication ( RAI ) : alerts the upstream equipment ofa downstream failure  RDI  RFI : remote failure indication  Payload defect indication ( PDI ) : alerts the downstream equipment that there is a defect in one or more of its embedded payloads.

Table 5.9 SONET Alarms

AIS and RDI 
AIS ± L  Generated when a line LOS / LOF is detected  An STE forms an OC-N signal with a valid section overhead and generates a line AIS by sending an all ± 1¶s pattern ( after scrambling ) for the remainder of the OC-N signal.  The signal is sent downstream.  When the LTE detects AIS-L, an RDI signal is sent upstream.  AIS ± P  Generated when a path LOS / LOF / LOP is detected  An LTE forms a STS path AIS by filling the entire STS SPE with all 1¶s including H1 ± H3 bytes ( after scrambling).  The signal is sent downstream to the STS-LTE.  An RDI signal is sent upstream if the AIS is for LOS / LOF only.

AIS and RDI 
AIS ± V  Generated when a VT LOS / LOF / LOP is detected.  An STS PTE generates a VT path AIS by filling the entire VT with all 1¶s ( after scrambling ).  The signal is sent downstream to the VT PTE.  An RDI signal is sent upstream if the AIS is for LOS or LOF.  Embedded DS-N with failures ( LOS, LOF, or LOP ) will cause an AIS depending on the composition.  RDI alarm signals are sent upstream to  Alert the network  Initiate trunk conditioning.

Types of RDI 
STS path RDI : alerts the upstream STS PTE that an AIS has been received in the downstream STS path.  VT path RDI : alerts the upstream VT PTE that an AIS has been received in the downstream VT path.  DS-N RDI : when this signal is detected, it is converted to a VT path RDI signal if a DS-N is mapped into its associated VT.

AIS ± RDI Summary 
When an STE receives an invalid signal, it sends an AIS-L alarm downstream.  When an LTE receives an invalid signal or an AIS-L and it is unable to protect the line, it sends an STS path AIS downstream.  When an STS PTE receives either an invalid signal or an STS path AIS, it propagates the appropriate AIS downstream.  When a VT PTE receives either a VT path AIS or an invalid signal, it generates the appropriate AIS downstream.  Upon detection at a terminal of a service affecting failure, a local RED alarm is declared and an RDI signal is returned upstream to the far-end NE that terminates the service.  Automatic service management processes are initiated at each NE and they are maintained for the duration of th failure.  Alarm states and RDI are coordinated between source and sink to restore service.

Performance Monitoring 
Performance monitoring ( PM ) : a set of rules for the in-service monitoring of the transmission quality.  Difference between SONET and SDH PM philosophy :  SDH : PM is based on counting erroneous blocks within a period of a second.  SONET : based on counting code violations within a period of a second.  SONET PM functions :  Detection of transmission degradation  Detection of performance parameter deviation  Communication of OSs

Performance Monitoring 
NE gathers PM from the values of :  Section BIP ( B1 )  Line BIP ( B2 )  STS path BIP ( B3 )  VT path BIP ( bits 1 and 2 of V5 : BIP ± 2 )  PM stores information collected over a period of time in some registers. Information related to  The current period  The previous period  The recent period  Threshold value  Two types of registers  Current-second register ( CSR ) : contains defects or anomalies that have occurred within a second.  Current-period register ( CPR ) : contains cumulative defects or anomalies detected and stored in the CSR.

PM Requirements 
A SONET NE shall provide the following accumulation and storage registers :  One current 15 min  One current day  One previous 15 min  One previous day  31 recent 15 min

PM Inhibition Registers 
Inhibition registers are positive and negative adjustment registers that are used to add or subtract one of the following to a current period register  Coding violation ( CV )  Erroneous second ( ES )  Severely erroneous second ( SES )  Unavailable second ( UAS )  Line pointer justification ( PJ ), STS path PJ, and VT path PJ  All of the above have a negative adjustment register, a positive adjustment register, or both

PM at the Physical Layer 
At the (photonic) physical layer, only the current value ( event ) is required  Laser bias current ( LBC ) LBCnormal = LBC/LBCo, where LBCo is the initial/nominal value provided by the NE supplier  Optical power transmitted ( OPT ) OPTnormal = OPT/OPTo, where OPTo is the initial/nominal value provided by the NE supplier.  Optical power received ( OPR ) OPRnormal = OPR/OPRo, where OPRo is the initial/nominal value provided by the NE supplier.

PM at the Section Layer 
Severely erroneous framing seconds ( SEFS-Ss ) : a count of seconds during which an SEF defect was present  Coding violations ( CV-Ss ) : a count of BIP errors detected at the section layer ( using the B1 byte ).  Erroneous seconds ( ES-Ss ) : a count of the number of seconds during which  at least one section layer BIP error was detected or  an SEF or  LOS defect was present  Severely erroneous seconds ( SES-Ss ) : a count of the seconds during which  K or more section layer BIP errors was detected or  an SEF or  LOS defect was present.

PM at the Line Layer 
The line layer performance parameters are divided into near end ( NE ) and the far end ( FE )  For the near-end line layer, performance parameters are :  NE line coding violations ( CV-Ls ) : a count of BIP errors detected at the line layer.  NE line erroneous seconds ( ES-Ls ) : a counter in seconds during which at least one line layer BIP error was detected or an AIS-L defect was present  NE line severely erroneous seconds ( SES-Ls ) : a counter in seconds during whichk K or more line layer BIP error ware detected or an AIS-L defect was present

PM at the Line Layer 
NE line unavailable seconds ( UAS-Ls ) : a count in seconds during which the line was considered unavailable.  NE line failure counts ( FC-L ) : a count of the number of near-end line failure events.
‡ A failure event begins when an AIS-L is declared and ends when an AIS-L is cleared. 

Protection switching count ( PSC ) : relates to systems that are equipped with two switching fabrics, the working and the protection
‡ For working line : a count of the times that service switched from the monitored line to the protection line + the times it switched back to the working line. ‡ For protection line : a count of the times that service switched from any working line to the protection line + the times it switched back to the working line.

PM at the Line Layer 
Protection switching duration ( PSD ) : a count in seconds
‡ For working line : a count in seconds that indicates the duration of time service was carried on the protection line. ‡ For protection line : a count in seconds that indicates the duration of time the protection line was used to carry service. 

STS pointer justification ( STS-PJ ) : a count of the STS pointer adjustments created or absorbed by an NE due to differences in the frame rates of incoming and outgoing SONET signals.
‡ The STS-PJ parameter is accumulated for a non-terminated STS path. 

For the far-end line layer performance parameters are :  FE line coding violations ( CV-LFEs ) : a count of BIP errors detected by the FE LTE and reported back to the NE LTE using the REI-L in the line OH.

PM at the Line Layer 
FE line erroneous seconds ( ES-LFEs ) : a count in seconds during which at least one line BIP error was reported by the FE LTE using the REI-L or an RDI-L defect was present.  FE line severely erroneous seconds ( SES-LFEs ) : a count of the seconds during which K or more line BIP errors were reported by the FE LTE or an RDI-L defect was present.  FE line unavailable seconds ( UAS-LFE ) : a count in seconds during which the line was considered unavailable at the FE.  FE line failure counts ( FC-LFEs ) : a count of the number of FE line failure events.
‡ A failure event begins when an RFI-L failure is declared and it ends when an RFI-L is cleared.

PM STS Path Layer 
STS path layer performance parameters are divided into STS and VT.  Each failure divided into near end and far end  The near end STS path layer performance parameters are :  NE STS path coding violations ( CV-Ps ) : a count of BIP errors detected at the STS path layer.  NE STS path erroneous seconds ( ES-Ps ) : a count in seconds during which at least one STS path BIP error was detected or an AIS-P defect was present.  NE STS path severely erroneous seconds ( SES-Ps ) : a count of the seconds during K or more STS path BIP errors were detected or an AIS-P or LOP-P was present.  NE STS path unavailable seconds ( UAS-P ) : a count of the seconds during which the STS path was unavailable.  NE STS path failure counts ( FC-Ps ) : a count of the number of NE STS path failure events.
‡ A failure event begins when an AIS-P or LOP-P is declared and it ends when it is cleared.

Fig. 5.78 Performance monitoring: intermediate path.

Fig. 5.79 Maintenance: facility loopbacks.

Fig. 5.80 Maintenance: DS-N loopbacks.

5.2 SONET and SDH

IP over SONET

IP over SONET 

Mapping IP over SONET takes 4 steps 1. The IP packet consists of a header and a variable-length data field, known as a datagram 2. There is a protocol known as point-to-point protocol ( PPP ). The IP datagram is encapsulated ina PPP packet; that is , a header is attached to it. 3. The PPP-encapsulated IP datagram is framed using highlevel data link control ( HDLC ). 4. The end result is mapped row-wise and byte-wise in the SPE. That is, VTs and byte multiplexing is not applicable in this case.

IP over SONET 
PPP provides multiprotocol encapsulation, error control, and link initialization control features.  HDLC is primarily used for frame delineation. That is, to identify the start and the end of the IP/PPP/HDLC frame.  Each HDLC frame stars and ends with the hexadecimal byte 7E or 0x7E.  Other 0x7E, if not in the start or the end position of the frame, it is replaced by the escape sequence 0x7D followed by 0x5E.  Similarly, the escape sequence 0x7D is replaced by 0x7D 0x5D.

Sign up to vote on this title
UsefulNot useful