Question? Leave a message!




SONET: Broadband Convergence

SONET: Broadband Convergence
SONET: Broadband Convergence at Layer 1 Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1Telephony: Multiplexing  Telephone Trunks between central offices carry hundreds of conversations: Can’t run thick bundles  Send many calls on the same wire: multiplexing  Analog multiplexing  bandlimit call to 3.4 KHz and frequency shift onto higher bandwidth trunk  Digital multiplexing: convert voice to samples  8000 samples/sec = call = 64 Kbps Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 2Telephony: Multiplexing Hierarchy  PreSONET:  Telephone call: 64 kbps  T1 line: 1.544 Mbps = 24 calls (aka DS1)  T3 line: 45 Mbps = 28 T1 lines (aka DS3)  Multiplexing and demultiplexing based upon strict timing (synchronous)  At higher rates, jitter is a problem  Have to resort to bitstuffing and complex extraction = costly “plesiochronous” hierarchy  SONET developed for higher multiplexing aggregates  Use of “pointers” like C to avoid bitstuffing Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 3Digital Telephony in 1984 Fiber Optic DS1 M13 M13 Transmission DS1 DS1 Cross Systems Connect • Switches • Leased Line Fiber DS3 Central Office M13 Central Office DS3 No Guaranteed Key System Aspects: Timing • M13 Building Blocks Synchronization DS1 • Asynchronous Operation • Electrical DS3 Signals • Proprietary Fiber Systems • Brute Force Cross Connect Central • ATT Network/Western Office Electric Equipment Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 4Digital Carrier Hierarchy (contd)  Multiplexing trunk networks: called “carrier” systems (eg: Tcarrier):  allowed fast addition of digital trunk capacity without expensive layout of new cables  Time frames (125 us) and a perframe bit in the Tcarrier for synchronization = TDM  Each phone call (DS0) occupies same position in the frame  Overhead bits: error control  “robbed” bits in voice call for OAM information  Too many 0s = synch loss (max number = 15)  “yellow alarm”. 1s density etc = usable b/w = 7bits/frame = 56 kbps  Europe: E1; more streamlined framing 2.048 Mbps  Variants: Concatenated T1, Unchannelized (raw) T1 Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 5Digital Hierarchy (Contd)  1980s: demand for bandwidth. But T3s not available except in proprietary form  Fiberoptic interface for T3 was proprietary  Primitive online OAMP capabilities (eg: robbed bits…)  Fewer operators: interoperability/midspan meet not critical  Changed dramatically after 1984 deregulation  Public vs Private Networks:  Private: Customer operates n/w (eg: w/ private leased lines): developed from PBX SNA  Public: Provider operates n/w for subscribers  More public networks (eg: X.25) outside US  Drivers of SONET:  IBM SNA/mainframes = hubandspoke networking  Increase of PCs = clientserver p2p computing = more demands on longdistance trunks  Tcarrier evolution rate much slower than computing trends Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 6Digital Hierarchy (Contd)  Digital streams organized as bytes (eg: voice samples, data)  Byte interleaving: (eg: 24 DS0 DS1)  service one byte from each input port into a transmission frame  Simple device: T1 mux a.k.a channel bank  Very convenient for processing, adddrop multiplexor (ADM) or Digital Crossconnect System (DCS) functions (fig 3.8/3.10)  ADM/DCS does both mux (“add”) and demux (“drop”) functions = need to do this with minimal buffering, fast/scalable processing  Bitinterleaving (eg: DS1 DS2 etc)  Cant use buffers to mask jitter = bit stuffing  Partly because high speed memory was costly then  “Plesiochronous hierarchy” = harder to ADM/DCS because full destuffing/demultiplexing necessary before these functions  DS3s used to be muxed using proprietary optical methods (eg: M13 mux): SONET solves all these problems Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 7US Telephone Network Structure (after 1984 divestiture) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 8PostATT Divestiture Dilemmas • Switches • Leased Line • LAN Services Different • Data Services Carriers, DS1 M13 Vendors Internal DS3 Cross Connect Needs: • Support Faster Fiber Support • Support New Services Other Topologies, • Allow Other Topologies Protect Fibers • Standardize Redundancy • Common OAMP • Scalable Cross Connect Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 9The SONET Standards Process Divestiture CCITT Expresses Interest in SONET SONET/SDH Exchange Carriers British and Japanese Standards Standards Associate (ECSA) Participation in T1X1 Approved T1 Committee Formed ANSI T1X1 CCITT XVIII CEPT Proposes Bellcore Proposed Approves Begins Study Merged ANSI/CCITT SONET Principles Project Group Standard To ANSI T1X1 1984 1985 1986 1987 1988 SONET Concept Developed By Bellcore US T1X1 Accepts 400 Technical Proposals Modifications • Rate Discussions ATT vs. Bellcore (resolved w/ virtual tributary concept) ANSI Approves • International Changes For Byte/Bit SYNTRAN Interleaving, Frames, Data Rates • Phase I, II, III Separate APS, etc. Shivkumar Kalyanaraman • ITU’s SDH initiative… Rensselaer Polytechnic Institute 10SONET Standards Story  SYNTRAN: predivestiture effort, no pointer concept.  SONET: primarily US (divestiture) driven  ATT vs Bellcore debate: 146.432 Mbps vs 50.688 Mbps: compromise at 49.94 Mbps  Virtual tributary concept to transport DS1 services  1986: CCITT (ITU) starts own effort (SDH)  June 1987: change SONET from bitinterleaved to byte interleaved; and rate from 49.92 to 51.84 Mbps  Phased rollouts:  1988 = Phase 1: signal level interoperability  Phase II: OAMP functions: embedded channel electrical I/f specification, APS work initiated  Phase III: OSI network management adopted  Seamless worldwide connectivity (allowed Europe to merge its Ehierarchy into SDH) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 11SONET: Achievements  1. Standard multiplexing using multiples of 51.84 Mbps (STS1 and STSN) as building blocks  2. Optical signal standard for interconnecting multiple vendor equipment  3. Extensive OAMP capabilities  4. Multiplexing formats for existing digital signals (DS1, DS2 etc)  5. Supports ITU hierarchy (E1 etc)  6. Accomodates other applications: BISDN etc Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 12SONET Lingo  OCN: Optical carrier Nx51.84 Mbps  Approximate heuristic: bit rate = N/20 Gbps (eg: OC48 = 48/20 = 2.4 Gbps)  Overhead percentage = 3.45 for all N (unlike PDH)  OC signal is sent after scrambling to avoid long string of zeros and ones to enable clock recovery  STSN: Synchronous Transport Signal (electronic equivalent of OC)  Envelope: Payload + endsystem overhead  Synchronous payload envelope (SPE): 9 rows, 87 columns in STS1  Overhead: management OAMP portion  Concatenation: “unchannelized” (envelope can carry “superrate” data payloads: eg: ATM): Eg: OC3c  Method of concatenation different from that of Tcarrier hierarchy… Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 13SONET Multiplexing Possibilities •Asynchronous DS3 •Virtual Tributaries for DS1 etc •STS3c for CEPT4 and B ISDN STS1s are mutually synchronized irrespective of inputs Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 14STS1 Frame Format 90 Bytes Or “Columns” 9 Rows Small Rectangle =1 Byte Twodimensional frame representation (90 bytes x 9 bytes)… Frame Transmission: Top Row First, Sent Left To Right • Timeframe: 125 ms/Frame • Frame Size Rate: 810 Bytes/Frame 8000 Frames/s 8 b/byte= 51.84 Mbps • For STS3, only the number of columns changes (90x3 = 270) STS = Synchronous Transport Signal Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 15STS1 Headers Section Overhead (SOH) 90 Bytes Or “Columns” 9 Rows Path Overhead (POH): Line Overhead (LOH) Floating = can begin anywhere Line + Section overhead = Transport Overhead (TOH) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 16SONET Equipment Types Path Sections PTE Repeaters • Section Termination (STE) Line SONET End • Line Termination (LTE) Device I.e. Telephony Switch, Router • Path Termination (PTE) PTE Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 17SONET Overhead Processing Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 18Headers: Section Overhead (SOH) A1 A2 J0/Z0 Rcv Xmt =0x28 STSID =0xF6 SOH SOH B1 F1 E1 Orderwire BIP8 User D1 D2 D3 Data Com Data Com Data Com Selected Fields: Section Overhead •A1,A2 Framing Bytes • 9 Bytes Total •BIP8 Bit Interleaved • Originated And Terminated By All Parity Section Devices (Regenerators, • F1 User Proprietary Multiplexers, CPE) OAM Management • Other Fields Pass Unaffected Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 19Headers: Line Overhead (LOH) H1 H2 H3 Pointer Pointer Pointer Act K1 K2 B2 BIP8 APS APS Xmt Rcv LOH LOH D4 D5 D6 Data Com Data Com Data Com Rcv Xmt Xmt Rcv D7 D8 D9 SOH SOH SOH SOH Data Com Data Com Data Com D10 D11 D12 Data Com Data Com Data Com S1 M0 E1 Sync REI Orderwire Line Overhead Selected Fields: • 18 Bytes Total •H13 Payload Pointers • Originated And Terminated By All •K1, K2 Automatic Line Devices (Multiplexers, CPE) Protection Switching • LOH+SOH=TOH (Transport OH) • D4D12 576 kbps OSI/CMIP Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 20Floating Payload: SONET LOH Pointers SPE is not framealigned: overlaps multiple frames Avoids buffer management complexity artificial delays Allows direct access to bytesynchronous lowerlevel signals (eg: DS1) with just one frame recovery procedure Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 21SPE: Synchronous Payload Envelope Defined Payloads Synchronous Payload Envelope • Virtual Tributaries • Contains POH + Data (For DS1, DS2) • First Byte Follows First Byte Of POH • DS3 • Wraps In Subsequent Columns • SMDS • May Span Frames • ATM • Up To 49.536 Mbps for Data: • PPP … •Enough for DS3 Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 22Headers: Path Overhead (POH) J1 STE PTE PTE Trace B3 BIP8 Frame N Selected Frame N Frame N+1 C2 Frame N+1 fields: Sig Label •BIP8 Parity G1 • C2 Payload Path Stat Path Overhead Type Indicator F2 User • H1,H2 fields of LOH points to • G1 End End H4 Beginning of POH Path Status Indicator •POH Beginning Floats Within Frame Z3 Growth • 9 Bytes (1 Column) Spans Frames Z4 • Originated And Terminated By All Growth Path Devices (I.e. CPE, Switches) Z5 Tandem • Endtoend OAM support Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 23STS1 Headers: Putting it Together Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 24Accommodating Jitter Positive Stuff Negative Stuff • To Shorten/Lengthen Frame: • Byte After H3 Ignored; Or H3 Holds Extra Byte • H1, H2 Values Indicate Changes Maximum Every 4 Frames • Requires Close (Not Exact) Clock Synch Among Elements Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 25Clock Synchronization BITS BITS 11 •Level 1: 10 8 •Level 2: 1.6x10 6 •Level 3: 4.6x10 6 •Level 4: 32x10 PTE Primary Backup Reference Reference Building Integrated Timing System • Hierarchical Clocking Distribution BITS • Normally All Synch’d To Stratum 1 (Can Be Cesium/Rubidium Clock) • Dedicated Link Or Recovered PTE • Fallback To Higher Stratum In Failure (Temperature Controlled Crystal) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 26STSN Frame Format 90xN Bytes Or “Columns” N Individual STS1 Frames Examples Composite Frames: STS1 51.84 Mbps • Byte Interleaved STS1’s STS3 155.520 Mbps • Clock Rate = Nx51.84 Mbps STS12 622.080 Mbps • 9 colns overhead STS48 2.48832 Gbps STS192 9.95323 Gbps Multiple frame streams, w/ independent payload pointers Note: header columns also interleaved Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 27STSN: Generic Frame Format STSN STS1 Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 28Example: STS3 Frame Format Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 29STSNc Frame Format 90xN Bytes Or “Columns” Transport Overhead: SOH+LOH Concatenated mode: • Same TOH Structure And Data Rates As STSN • Not All TOH Bytes Used • First H1, H2 Point To POH • Single Payload In Rest Of SPE • Accommodates FDDI, E4, data Current IP over SONET technologies use concatenated mode: OC3c (155 Mbps) to OC192c (10 Gbps) rates a.k.a “superrate” payloads Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 30Virtual Tributaries (Containers)  Opposite of STSN: submultiplexing  STS1 is divided into 7 virtual tributary groups (12 columns ea), which can be subdivided further  VT groups are byteinterleaved to create a basic SONET SPE  VT1.5: most popular quickly access T1 lines within the STS1 frame  SDH uses the word “virtual containers” (VCs) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 31Virtual Tributaries: Pointers  VT payload (a.k.a VT SPE) floats inside the VT  One more level of pointer used to access it.  Can access a T1 with just two pointer operations  Very complex to do the same function in DS3  Eg: accessing DS0 within DS3 requires FULL de multiplexing: a.k.a stacked multiplexing or mux mountains Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 32SONET Transmission Encoding Electrical Transmission Scrambling Standard • Ensures Ones Density • STS1: B3ZS (BPV), 450’ • Does Not Include A1, A2, C1 Bytes • STS3: Coded Mark • Output Is NRZ Encoded Inversion, 225’ • Useful IntraOffice Connection 6 7 1+x +x E O OCN Is Optical Carrier STSN Long Reach: 40 km 1310 or 1550 nm SM Intermediate Reach: 15 km 1310 or 1550 nm SM Short Reach:Long Reach 2 km 1310 nm MM Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 33SONET Scrambling Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 34Packet Over SONET (POS) Special Data Scrambler • 1+ x43 Polynomial Standard PPP Encapsulation • Protects Against Transmitted • Magic Number Recommended Frames Containing Synch Bytes • No Address and Control Compression Or Insufficient Ones Density • No Protocol Field Compression Byte SONET PPP FCS Scrambling Stuff Framing Standard CRC Computation SONET Framing • OC3 May Use CRC16 • OC3, OC12, OC48, OC192 Defined • Other Speeds Use CRC32 • C2 Byte = 0x16 With Scrambling • C2 Byte = oxCF Without (OC3) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 35Practical SONET Architectures Today: multiple “stacked” rings over DWDM (different s) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 36SONET Network Elements D+R MN MN DS1s TM ADM DCC D+R MN MN D+R DS1s Nonstandard, Functional Names TM: Terminal Mux: (aka LTE: ends of ptpt links) ADM: AddDrop Mux DCC: Digital Cross Connect (Wideband and Broadband) MN: Matched Node D+R: Drop and Repeat Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 37Digital Cross Connects (DCS)  Crossconnects thousands of streams under software control (replaces patch panel)  Handles perf monitoring, PDH/SONET streams, and also provides ADM functions  Grooming:  Grouping traffic with similar destinations, QoS etc  Muxing/extracting streams also  Narrow/wide/broadband and optical crossconnects Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 38Topology Building Blocks ADM ADM 2 Fiber Ring 4 Fiber Ring DCC ADM DCC ADM Each Line Is Each Line Is ADM ADM Full Duplex Full Duplex ADM ADM DCC ADM DCC ADM ADM ADM Uni vs. Bi Directional All Traffic Runs Clockwise, vs Either Way Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 39APS ADM ADM ADM ADM ADM ADM Line Protection Switching Path Protection Switching Uses TOH Uses POH Trunk Application Access Line Applications Backup Capacity Is Idle Duplicate Traffic Sent On Protect Supports 1:n, N=114 1+1 Automatic Protection Switching • Line Or Path Based • Revertive vs. NonRevertive • Mechanism For Intentional Cutover • Restoration Times 50 ms • K1, K2 Bytes Signal Change • Common Uses: 2 Fiber UPSR or ULSR, 4 Fiber BPSR Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 40
Website URL
Comment
sharer
Presentations
Free
Document Information
Category:
Presentations
User Name:
Dr.NeerajMittal
User Type:
Teacher
Country:
India
Uploaded Date:
19-07-2017