TDM PWs Yaakov J Stein Chief Scientist RAD

TDM PWs Yaakov (J) Stein Chief Scientist RAD Data Communications

Outline 1) Pseudowires 2) Emulating TDM 3) TDMo. IP encapsulation formats 4) TDM signaling transport 5) Timing recovery 6) Packet loss and mis-ordering 2

Pseudowires Pseudowire (PW): A mechanism that emulates the essential attributes of a native service while transporting over a packet switched network (PSN) 3

The old model (X. 200, OSI) Once upon a time networks were exclusively described by the OSI model However n few networks actually work only that way n highly inflexible (always need more layers!) n some features only in one place (security, mux) n missing features (OAM) n doesn’t help to design transport networks APPLICATION PRESENTATION SESSION TRANSPORT NETWORK LINK PHYSICAL 4

Simple telephony counter-example voice channel E 1 (TDM) E 3 (PDH) OSI application layer ? STM 1 (SDH) OC 3 (OTN) voice channel E 1 (TDM) E 3 (PDH) STM 1 (SDH) OSI physical layer OC 3 (OTN) n this type of scenario important to carriers, and thus to ITU-T n not captured by ISO layering model n there can be an arbitrary large number of intervening layers n all intermediate layers fulfill the same function -- transport 5

The new model (G. 805) A more general and applicable model for transport networks n Layer network and trail n Layering and partition n Basic network modes n Interworking n Diagrammatic technique References: G. 805 generic G. 806 CO networks G. 809 CL networks G. 705 PDH G. 781 timing G. 783 SDH G. 732 ATM G. 8010 Ethernet G. 8110 MPLS 6

Layer networks Layering Network may be decomposed (vertically) into layer networks Client-server relationship between adjacent layer networks Layer network Basic topological component for information transfer Link in layer network supported by network below Layer network provides link connection to layer above Layers are completely independent Trail Transport entity in layer network Contains client payload and OAM Partitioning Network may be decomposed (horizontally) into subnetworks connected by links Recursively, each subnetwork is similarly decomposed Peer-peer relationship between adjacent subnetworks 7

Network Modes Circuit Switched Packet Switched (CS) (PSN) Connection Oriented Connectionless (CO) (CL) n Many native network types (technologies) for each mode – CS: TDM, PDH, SDH, OTN – CO: ATM, FR, MPLS, TCP/IP, SCTP/IP – CL: UDP/IP, IPX, Ethernet, CLNP n Can layer any mode over any mode – BUT some layerings may involve performance loss – CL over CO over CS is EASY – CO over CL, or CS over CO is harder – CS over CL is very hard 8

Network interworking (tunneling) Network interworking may be provided by tunneling (edge to edge) network Native Service edge Service Interworking requires more complex mechanisms Native Service A network Native Service B 9

Pseudo. Wire Emulation Edge to Edge PWE 3 Customer Edge Pseudowire (PW): mechanism that emulates essential attributes of a native service while transporting over a PSN (CE) Customer Edge Provider’s PSN Provider Edge (CE) (PE) Customer Edge native service Pseudo. Wires (PWs) native service (CE) 10

Emulating TDM From PSTN to PSN 11

Classic Telephony Access Network Core (Backbone) Network analog lines CO SWITCH T 1/E 1 extensions P B X T 1/E 1 or AAL 1/2 n SONET/SDH NETWORK CO SWITCH P B X Synchronous Non-packet network Circuit switched ensures signal integrity n Very High Reliability (“five nines”) n Low Delay and no noticeable echo n Timing information transported over the network n Mature Signaling Protocols (over 3000 features) 12

A few G. XXX sayings … n G. 114 (One-way transmission time) – – n delay < 150 ms acceptable 150 ms < delay < 400 ms conditionally acceptable delay > 400 ms unacceptable G. 126/G. 131 echo control may be needed G. 823/G. 824 (timing) – primary vs. secondary clocks – jitter masks – wander masks n G. 826 (error performance) – BER better than 2 * 10 -4 – strict limitation on errored seconds 13

TDM PWs Access Network analog lines T 1/E 1/T 3/E 3 extensions P B X T 1/E 1 or AAL 1/2 Packet Switched Network P B X Asynchronous network No timing information transfer TDMo. IP replaces CS core with a PSN The access networks and their protocols remain ! TDM Pseudowire Can G. xxx compliance be maintained? 14

Network Comparison TDM PSN Circuit switched Connection oriented / connectionless Guaranteed BW Shared BW Low overhead High overhead Minimal delay Delay (introduced by forwarding) Constant arrival rate Packet delay variation (and bursts) Timing transport No physical layer clock No information loss Packet loss (congestion, errors) 15

TDMo. IP Protocol Processing TDM frames IP Packets TDM frames PSN Steps in TDMo. IP n The synchronous bit stream is segmented n The TDM segments are adapted n TDMo. IP control word is prepended n PSN (IP/MPLS) headers are prepended (encapsulation) n Packets are transported over PSN to destination n PSN headers are utilized and stripped n Control word is checked, utilized and stripped n TDM stream is reconstituted (using adaptation) and played out 16

TDMo. IP vs. Vo. IP Two ways to integrate TDM services into PSNs Vo. IP n n n Revolution - complete (forklift) CPE replacement New signaling protocols (translation needed) New functionality (e. g. video-phone, presence) TDMo. IP n n Evolution - CPE unchanged, IWF added at edge No change to signaling protocols (network IW) No new functionality Migration path 17

TDMo. IP encapsulation formats 18

TDMo. IP layering structure PSN / multiplexing Optional RTP header TDMo. IP Control Word higher layers Adapted TDM payload 19

PW Multiplexing to reduce resources in core network PWs are sent inside PSN tunnels we often wish to send several PWs in same tunnel to demux we use a PW label n for application muxing, IANA has assigned to TDMo. IP UDP port number 0 x 085 E (2142) n in IP networks we use UDP source port number as bundle ID n in MPLS networks we use an inner label n for L 2 TPv 3 we could use L 2 TP multiplexing 20

Packet Components PSN headers • ensure packet transported to destination RTP header • contains timestamp that may help in timing recovery Control Word • enables detection of out-of-order and lost packets • indicates critical alarm conditions TDM payload may be adapted • to assist in timing recovery and recovery from packet loss • to ensure proper transfer of TDM signaling • to provide an efficiency vs. latency trade-off 21

TDM over IP and MPLS IP header UDP header (20 bytes) (PW label) Optional RTP header (8 bytes) (12 bytes) TDMo. IP Control Word (4 bytes) TDM payload PSN label PW label control word TDM payload 22

TDMo. IP Control Word PID flags FRG Length Sequence Number PID (4 b) special uses flags (4 b) – L bit (Local failure) – R bit (Remote failure) FRG (2 bits) indicates fragmentation (only for special uses) Length (6 b) used when packet may be padded Sequence Number (16 b) used to detect packet loss / misordering 23

TDM Payload What needs to be transported from end to end? n n n Voice (telephony quality, low delay, echo-less) Tones (for dialing, PIN, etc. ) Fax and modem transmissions Signaling (there are 1000 s of PSTN features!) Timing “timeslots” T 1/E 1 frame SYNC TS 1 TS 2 TS 3 … signaling bits … TSn (1 byte) 24

Why not N bytes? Why don’t we simply encapsulate N bytes frame? IP UDP RTP? N TDM octets because a single lost packet would cause service interruption n need constant N (else don’t know how many TDM bytes were lost) n n need to conceal lost packet by proper amount of AIS “all ones” TDM synchronization would be lost SATo. P is good for well-engineered networks n essentially no packet loss n very low PDV (see below) 25

Why not one frame? Why don’t we simply encapsulate the T 1/E 1 frame? 24 or 32 bytes IP UDP RTP? T 1/E 1 frame because it is inefficient - however N frames is reasonable (structure-locking) because a single lost packet could cause service interruption n n and for CAS, signaling uses a superframe (16/24 frames) so superframe integrity must be respected too because we want to efficiently handle fractional T 1/E 1 because we want a latency vs. efficiency trade-off 26

TDM Structure handling of TDM depends on its structure unstructured TDM (TDM = arbitrary stream of bits) … structured TDM framed S Y N C (8000 frames per second) S Y N C channelized SYNC S Y N C (single byte timeslots) TS 2 TS 1 (1 byte) TS 3 … signaling bits … TSn multiframed frame … multiframe 27

TDM transport types Structure-agnostic transport (SATo. P) • for unstructured TDM • even if there is structure, we ignore it • simplest way of making payload • OK if network is well-engineered Structure-aware transport (TDMo. IP, CESo. PSN) • take TDM structure into account • must decide which level of structure (frame, multiframe, …) • can overcome PSN impairments (PDV, packet loss, etc) 28

Structure aware encapsulations Structure-locked encapsulation (CESo. PSN) headers TDM structure Structure-indicated encapsulation (TDMo. IP – AAL 1 mode) headers AAL 1 subframe Structure-reassembled encapsulation (TDMo. IP – AAL 2 mode) headers AAL 2 minicell 29

Structure indication - AAL 1 For robust emulation: n n adding a packet sequence number adding a pointer to the next superframe boundary only sending timeslots in use allowing multiple frames per packet UDP/IP seqnum ptr T 1/E 1 frames (only timeslots in use) (with CRC) for example 7 @ TS 1 TS 2 TS 5 TS 7 30

Structure reassembly - AAL 2 TDM frame 1 1 1 2 PSN hdrs TDM frame 3 4 4 4 5 TDM frame CW 2 2 hdr 1 2 3 4 TS 1 3 5 3 hdr 1 2 3 TS 2 TDM frame 4 hdr 5 1 2 3 5 4 5 5 TS 3 AAL 1 is inefficient when timeslots are dynamically allocated n each minicell consists of a header and buffered data n minicell header contains: – CID (Channel IDentifier) – LI (Length Indicator) = length-1 – UUI (User-User Indication) counter + payload type ID 31

TDM Signaling 32

Signaling? signaling is used for network control – call setup/tear-down (including routing) – OAM – billing in TDM networks there may be different types: – Subscriber - CO – CO - CPE (e. g. PBX) there are four common PSTN signaling techniques: – Analog * (E&M, ground-start/loop-start, ring-voltage, etc) – In-band (dial-tone, ring-back, DTMF, etc) – CAS – Channel Associated Signaling – CCS – Common Channel Signaling * we needn’t discuss the analog techniques 33

In-band signaling n in-band signaling is transferred in the audio (200 -3600 Hz) band n for example: – call progress tones (dial tone, ring back) – DTMF tones, – FSK for caller identification, – MFR 1 in North America or MFCR 2 in Europe, audible tones in TDM time slot automatically forwarded n n this is not the case for Vo. IP! – speech compression may not pass (need tone relay) – Vo. IP protocols replace legacy signaling with its own l l l SIP H. 323 Megaco 34

CAS is carried in the same T 1/E 1 as payload – but not in the audio – T 1 uses robbed bits – E 1 uses a dedicated time slot (usually TS 16) Readily handled by TDMo. IP (even for fractional T 1/E 1 links) Vo. IP systems need to – detect the CAS bits, – interpret them according to the appropriate protocol – transport them through PSN using a relay protocol – finally regenerate and recombine them at the far end 35

CCS Examples: ISDN PRI signaling, SS 7 if occupy a TDM timeslot (trunk associated) then forwarded by TDMo. IP (see HDLCo. IP) if not trunk associated, then forwarded by signaling network or signaling gateway employed l encapsulate (relay) the native signaling l forward as additional traffic through the PSN 36

HDLCo. IP intended to operate in port mode Data / control messages transparently transported Assume messages shorter than the MTU (no fragmentation) Only use when has potential to significantly compress BW Transmission : – monitor flags until frame detected – test FCS – if incorrect - discarded – if correct - l perform unstuffing l flags and FCS removed l send frame 37

TDM Timing Recovery 38

TDM Jitter and Wander Jitter = short term timing variation * Wander = long term timing variation * (i. e. fast jumps - frequency > 10 Hz) (i. e. slow moving- frequency < 10 Hz) Jitter amplitude in UIpp Unit Interval pk-pk Measure in MTIE(t) or TDEV(t) E 1 : 1 UIpp = 1/2 MHz = 488 ns Mask as function of t MTIE - max pk-pk error TDEV expected deviation * compared to reference clock Note: requirements for E 1 given in G. 823 for T 1 given in G. 824 39

PSN - Delay and PDV n n PSNs do not carry timing n clock recovery required for TDMo. IP PSNs introduce delay and packet delay variation (PDV) n Delay degrades perceived voice quality n PDV makes clock recovery difficult E 1/T 1 VOICE TDMo. IP DATA GW PSN TDMo. IP GW DATA The arrival time is not constant!!! 40

Jitter Buffer Arriving TDMo. IP packets written into jitter buffer Once buffer filled 1/2 can start reading from buffer Packets read from jitter buffer at constant rate How do we know the right rate? How do we guard against buffer overflow/underflow? E 1/T 1 VOICE TDMo. IP DATA GW PSN TDMo. IP GW DATA Jitter Buffer 41

Clock Recovery The packets are injected into network ingress at times Tn For TDM the source packet rate R is constant Tn = n / R The network delay Dn can be considered to be the sum of typical delay d and random delay variation Vn The packets are received at network egress at times tn tn = Tn + Dn = Tn + d + Vn By proper averaging/filtering <tn > = Tn + d = n / R + d and the packet rate R has been recovered 42

FLL We can estimate the rate R by counting the number of arrivals N per unit time T the longer the averaging the better the estimate R = N / T Open loop frequency setting Better method is closed loop Measure reception rate Fn = 1 / (tn - tn-1) Correct present rate F according to filtered Fn F = F + a < Fn - F > tn TDMo. IP GW F 44

PLL Phase difference between write (arrival) clock and read (present local) clock Number of packets written into the jitter buffer minus the number of packets read from the jitter buffer 360 o write events read events 270 o counter 45

Packet Loss and Misordering 46

Reasons for packet loss In a perfect network all packets should reach their destination In real networks, some packets are lost Loss is caused by bit errors invalidating the data (detected by ECC) intentional dropping by forwarder because of congestion intentional dropping by forwarder due to policy (e. g. (W)RED) router 47

Handling of packet loss In order to maintain timing SOMETHING must be output towards the TDM interface when a packet is lost PSN Packet Loss Concealment methods: n fixed n replay n interpolation 48

Voice Quality Comparison See draft-stein-pwe 3 -tdm-packetloss-00. txt 49

Mis-ordering In a perfect network all packets should arrive in proper order In real networks, some packets are delayed (or even duplicated!) Misordering is caused by parallel paths – aggravated by load balancing mechanisms 1 2 3 4 5 2 1 router 3 4 5 router 1 2 4 3 Misordering can be handled by n Reordering (from jitter buffer) n Handling as packet loss and dropping later 50 5