SONET

SONET/SDHSONET/SDH

Yaakov (J) Stein Chief ScientistRAD Data Communications

Y(J)S SONET Slide 2

Course OutlineCourse Outline

Background (analog telephony, TDM, PDH)SONET/SDH history and motivationArchitecture (path, line, section)Rates and frame structurePayloads and mappingsProtection and rings VCAT and LCASHandling packet data

Y(J)S SONET Slide 3

BackgroundBackground

Y(J)S SONET Slide 4

The PSTN circa 1900The PSTN circa 1900

pair of copper wires

“local loop”

manual routing at local exchange office (CO)

• Analog voltage travels over copper wire end-to-end • Voice signal arrives at destination severely attenuated and distorted

• Routing performed manually at exchanges office(s)• Routing is expensive and lengthy operation• Route is maintained for duration of call

Y(J)S SONET Slide 5

Telephony MultiplexingTelephony Multiplexing1900: 25% of telephony revenues went to copper mines standard was 18 gauge, long distance even heavier two wires per loop to combat cross-talk needed method to place multiple conversations on a single trunk

1918: “Carrier system” (FDM) 5 conversations on single trunk later extended to 12 (group) still later supergroups (60), master groups (60)), …

fchannels

8 kHz

12 kHz

4 kHz

16 kHz

20 kHz

Y(J)S SONET Slide 6

The Digitalization of the PSTNThe Digitalization of the PSTN

Shannon (Bell Labs) proved thatDigital communicationsis always better than

Analog communicationsand the PSTN became digital

Better means More efficient use of resources (e.g. more channels on trunks) Higher voice quality (less noise, less distortion) Added features

After the invention of the transistor, in 1963 T-carrier system (TDM) 1 byte per sample – 8000 samples per second T1 = 24 conversations per trunk 2 groups per cable! t

timeslots

Y(J)S SONET Slide 7

and switching became easier tooand switching became easier too

Complexity increases rapidly with size

1 2 4 5 6 7 83

1234567

Analog Crossbar switch Digital Cross-connect (DXC)

processor

t1 2 3 4 5

t2 1 5 4 3

Y(J)S SONET Slide 8

Optimized Telephony RoutingOptimized Telephony Routing

Circuit switching (route is maintained for duration of call)

Route “set-up” is an expensive operation, just as it was for manual switching

Today, complex least cost routing algorithms are used

Call duration consists of set-up, voice and tear-down phases

Y(J)S SONET Slide 9

The PSTN circa 1960The PSTN circa 1960

local loop

subscriber line

automatic routing through universal telephone network

• Analog voltages used throughout, but extensive Frequency Division Multiplexing • Voice signal arrives at destination after amplification and filtering to 4 KHz

• Automatic routing• Universal dial-tone• Voltage and tone signaling• Circuit switching (route is maintained for duration of call)

trunks

circuits

Y(J)S SONET Slide 10

The Present PSTNThe Present PSTN

subscriber line

• Analog voltages and copper wire used only in “last mile”, but core designed to mimic original situation• Voice signal filtered to 4 KHz at input to digital network

• Time Division Multiplexing of digital signals in the network• Extensive use of fiber optic and wireless physical links• T1/E1, PDH and SONET/SDH “synchronous” protocols

• Signaling can be channel/trunk associated or via separate network (SS7)

• Automatic routing• Circuit switching (route is maintained for duration of call)• Complex routing optimization algorithms (LP, Karmarkar, etc)

PSTN Network

class 5 switchclass 5 switch

tandem switch

last mile


TDM timingTDM timingTime Domain Multiplexing relies on all channels (timeslots)

having precisely the same timing (frequency and phase)

In order to enforce thisthe TDM device itself frequently performs the digitization

analog

signals

digital

signals


if the inputs are already digitalif the inputs are already digital

If the TDM switch does not digitize the analog signalsthen there can be a problemthe clocks used to digitize do not have identical frequencies

we get byte slips! (well, actually, we can get bit slips first …)

exaggerated pictorial example

Numerical example:

clock derived from 8000 Hz. quartz crystal

typical crystal accuracy = 50 ppm

So 2 crystals can differ by 100 ppm

i.e. 0.8 samples / second

So difference is 1 sample after 1 ¼ seconds

1

1

1

1

1

1

2

2

2

2

2

2

3

3

3

3

3

3

4

4

4

4

4

4

5

5

5

6

6

6

5

5

5

7

7

7

6

6

6

8

8

8

9

9

9

7

7

7

9

8

8

componentsignals

TDM


The fixThe fixWe must ensure that all the clocks have the same frequency

Every telephony network has an accurate clock called

a “stratum 1” or “Primary Reference Clock”

All other clocks are directly or indirectly locked to it (master – slave)

A TDM receiving device can lock onto the source clock based on the incoming data (FLL, PLL)

For this to work, we must ensure that the data has enough transitions(special line coding, scrambling bits, etc.)

1

0transitions no transitions


Comparing clocksComparing clocks

A clock is said to be isochronous (isos=equal, chronos=time)

if its ticks are equally spaced in time

2 clocks are said to be synchronous (syn=same chronos=time)

if they tick in time, i.e. have precisely the same frequency

2 clocks are said to be plesiochronous (plesio=near chronos=time)

if they are nominally if the same frequency but are not locked


PDH principlePDH principleIf we want yet higher rates, we can mux together TDM signals (tributaries)We could demux the TDM timeslots and directly remux them

– but that is too complex

The TDM inputs are already digital, so we must– insist that the mux provide clock to all tributaries (not always possible, may already be locked to a network)

OR– somehow transport tributary with its own clock

across a higher speed network with a different clock (without spoiling remote clock recovery)


PDH hierarchiesPDH hierarchies

64 kbps

2.048 Mbps 1.544 Mbps 1.544 Mbps

6.312 Mbps 6.312 Mbps8.448 Mbps

34.368 Mbps

139.264 Mbps

44.736 Mbps 32.064 Mbps

97.728 Mbps274.176 Mbps

CEPT N.A. Japan

4

3

2

1

0

level

* 30* 24 * 24

* 4

* 4

* 4

* 4

* 7

* 6

* 4

* 5

* 3

E1

E2

E3

E4

T1

T2

T3

T4

J1

J2

J3

J4


Framing and overheadFraming and overheadIn addition to locking on to bit-rate

we need to recognize the frame structureWe identify frames by adding Frame Alignment Signal

The FAS is part of the frame overhead (which also includes "C-bits", OAM, etc.)

Each layer in PDH hierarchy adds its own overhead

For example E1 – 2 overhead bytes per 32 bytes – overhead 6.25 % E2 – 4 E1s = 8.192 Mbps out of 8.448Mbps

so there is an additional 0.256 Mbps = 3 % altogether 4*30*64 kbps = 7.680 Mbps out of 8.448 Mbps

or 9.09% overhead

What happens next ?


PDH overheadPDH overhead

Overhead always increases with data rate !

digital signal

data rate

(Mbps)

voice

channels

overhead percentage

T1 1.544 24 0.52 %

T2 6.312 96 2.66 %

T3 44.736 672 3.86 %

T4 274.176 4032 5.88 %

E1 2.048 30 6.25 %

E2 8.448 120 9.09 %

E3 34.368 480 10.61 %

E4 139.264 1920 11.76 %


OAMOAManalog channels and 64 kbps digital channels

do not have mechanisms to check signal validity and quality

thus major faults could go undetected for long periods of time hard to characterize and localize faults when reported minor defects might be unnoticed indefinitely

Solution is to add mechanisms based on overhead

as PDH networks evolved, more and more overhead was dedicated toOperations, Administration and Maintenance (OAM) functions

including: monitoring for valid signal defect reporting alarm indication/inhibition (AIS)


PDH JustificationPDH JustificationIn addition to FAS, PDH overhead includes

justification control (C-bits) and justification opportunity “stuffing” (R-bits)Assume the tributary bitrate is B TPositive justification

payload is expected at highest bitrate B+Tif the tributary rate is actually at the maximum bitrate

then all payload and R bits are filledif the tributary rate is lower than the maximum

then sometimes there are not enough incoming bitsso the R-bits are not filled and C-bits indicate this

Negative justificationpayload is expected at lowest bitrate B-Tif the tributary rate is actually the minimum bitrate

then payload space suffices if the tributary rate is higher than the minimum

then sometimes there are not enough positions to accommodateso R-bits in the overhead are used and the C-bits indicate this

Positive/Negative justificationpayload is expected at nominal bitrate Bpositive or negative justification is applied as required


SONET/SDH SONET/SDH

motivation and historymotivation and history


First stepFirst stepWith the disvestiture of the US Bell system a new need aroseMCI and NYNEX couldn’t directly interconnect optical trunksInterexchange Carrier Compatibility Forum requested T1 to solve problem

Needed multivendor/ multioperator fiber-optic communications standardThree main tasks: Optical interfaces (wavelengths, power levels, etc)

proposal submitted to T1X1 (Aug 1984)T1.106 standard on single mode optical interfaces (1988)

Operations (OAM) systemproposal submitted to T1M1T1.119 standard

Rates, formats, definition of network elementsBellcore (Yau-Chau Ching and Rodney Boehm) proposal (Feb 1985)proposed to T1X1term SONET was coinedT1.105 standard (1988)


PDH limitationsPDH limitations

Rate limitations Copper interfaces defined Need to mux/demux hierarchy of levels (hard to pull out a single timeslot) Overhead percentage increases with rate

At least three different systems (Europe, NA, Japan)– E 2.048, 8.448, 34.348, 139.264– T 1.544, 3.152, 6.312, 44.736, 91.053, 274.176– J 1.544, 3.152, 6.312, 32.064, 97.728, 397.2

So a completely new mechanism was needed


Idea behind SONETIdea behind SONET

Synchronous Optical NETwork Designed for optical transport (high bitrate) Direct mapping of lower levels into higher ones Carry all PDH types in one universal hierarchy

– ITU version = Synchronous Digital Hierarchy– different terminology but interoperable

Overhead doesn’t increase with rate OAM designed-in from beginning


Standardization !Standardization !The original Bellcore proposal: hierarchy of signals, all multiple of basic rate (50.688) basic rate about 50 Mbps to carry DS3 payload bit-oriented mux mechanisms to carry DS1, DS2, DS3

Many other proposals were merged into 1987 draft document (rate 49.920)

In summer of 1986 CCITT express interest in cooperation needed a rate of about 150 Mbps to carry E4 wanted byte oriented mux

Initial compromise attempt byte mux US wanted 13 rows * 180 columns CEPT wanted 9 rows * 270 columns

Compromise! US would use basic rate of 51.84 Mbps, 9 rows * 90 columns CEPT would use three times that rate - 155.52 Mbps, 9 rows * 270 columns


SONET/SDH SONET/SDH

architecturearchitecture


LayersLayersSONET was designed with definite layering concepts

Physical layer – optical fiber (linear or ring)– when exceed fiber reach – regenerators– regenerators are not mere amplifiers, – regenerators use their own overhead– fiber between regenerators called section (regenerator section)

Line layer – link between SONET muxes (Add/Drop Multiplexers)– input and output at this level are Virtual Tributaries (VCs)– actually 2 layers

lower order VC (for low bitrate payloads) higher order VC (for high bitrate payloads)

Path layer – end-to-end path of client data (tributaries)– client data (payload) may be

PDH ATM packet data


SONET architectureSONET architecture

SONET (SDH) has at 3 layers: path – end-to-end data connection, muxes tributary signals path section

– there are STS paths + Virtual Tributary (VT) paths

line – protected multiplexed SONET payload multiplex section section – physical link between adjacent elements regenerator section

Each layer has its own overhead to support needed functionality

SDH terminology

PathTermination

PathTermination

LineTermination

LineTermination

SectionTermination

path

line line line

ADM ADMregenerator

section section sectionsection


STS, OC, etc.STS, OC, etc.

A SONET signal is called a Synchronous Transport Signal

The basic STS is STS-1, all others are multiples of it - STS-N

The (optical) physical layer signal corresponding to an STS-N is an OC-N

SONET Optical rate

STS-1 OC-1 51.84M

STS-3 OC-3 155.52M

STS-12 OC-12 622.080M

STS-48 OC-48 2488.32M

STS-192 OC-192 9953.28M

* 3

* 4

* 4

* 4


rates rates

and and

frame structureframe structure


SONET / SDH framesSONET / SDH frames

Synchronous Transfer Signals are bit-signals (OC are optical)

Like all TDM signals, there are framing bits at the beginning of the frame

However, it is convenient to draw SONET/SDH signals as rectangles

framing


SONET STS-1 frameSONET STS-1 frame

Each STS-1 frame is 90 columns * 9 rows = 810 bytes

There are 8000 STS-1 frames per secondso each byte represents 64 kbps (each column is 576 kbps)

Thus the basic STS-1 rate is 51.840 Mbps

90 columns

9 ro

ws

framing


SDH STM-1 frameSDH STM-1 frame

Synchronous Transport Modules are the bit-signals for SDHEach STM-1 frame is 270 columns * 9 rows = 2430 bytesThere are 8000 STM-1 frames per secondThus the basic STM-1 rate is 155.520 Mbps

3 times the STS-1 rate!

270 columns

9 ro

ws

…


SONET/SDH ratesSONET/SDH rates

STS-N has 90N columns STM-M corresponds to STS-N with N = 3M

SDH rates increase by factors of 4 each time

STS/STM signals can carry PDH tributaries, for example: STS-1 can carry 1 T3 or 28 T1s or 1 E3 or 21 E1s STM-1 can carry 3 E3s or 63 E1s or 3 T3s or 84 T1s

SONET SDH columns rate

STS-1 90 51.84M

STS-3 STM-1 270 155.52M

STS-12 STM-4 1080 622.080M

STS-48 STM-16 4320 2488.32M

STS-192 STM-64 17280 9953.28M


SONET/SDH tributariesSONET/SDH tributaries

E3 and T3 are carried as Higher Order Paths (HOPs)

E1 and T1 are carried as Lower Order Paths (LOPs) (the numbers are for direct mapping)

SONET SDH T1 T3 E1 E3 E4

STS-1 28 1 21 1

STS-3 STM-1 84 3 63 3 1

STS-12 STM-4 336 12 252 12 4

STS-48 STM-16 1344 48 1008 48 16

STS-192 STM-64 5376 192 4032 192 64


Synchronous Payload Envelope

STS-1 frame structureSTS-1 frame structure9

row

s

TransportOverhead

TOH

6 ro

ws

3 ro

ws

Section overhead is 3 rows * 3 columns = 9 bytes = 576 kbpsframing, performance monitoring, management

Line overhead is 6 rows * 3 columns = 18 bytes = 1152 kbpsprotection switching, line maintenance, mux/concat, SPE pointer

SPE is 9 rows * 87 columns = 783 bytes = 50.112 Mbps

Similarly, STM-1 has 9 (different) columns of section+line overhead !

90 columns

9 ro

ws


STM-1 frame structureSTM-1 frame structure

SectionOverhead

SOHSTM-1 has 9 (different) columns of transport overhead !

RS overhead is 3 rows * 9 columns

Pointer overhead is 1 row * 9 columns

MS overhead is 5 rows * 9 columns

SPE is 9 rows * 261 columns

…

270 columns

RSOH

MSOH


Even higher ratesEven higher rates

3 STS-1s can form an STS-34 STM-1s (STS-3s) can form an STM-4 (STS-12)4 STM-4s (STS-12s) can form an STM-16 (STS-48)etc. for STM-N (STS-3N)The procedure is byte-interleaving

9 rows

9*N columns

270*N columns


Byte-interleavingByte-interleaving

. . .


ScramblingScramblingSONET/SDH receivers recover clock based on incoming signal

Insufficient number of 0-1 transitions causes degradation of clock performance

In order to guarantee sufficient transitions, SONET/SDH employ a scrambler All data except first row of section overhead is scrambled Scrambler is 7 bit self-synchronizing X7 + X6 + 1 Scrambler is initialized with ones

A short scrambler is sufficient for voice databut NOT for data which may contain long stretches of zeros

When sending data an additional payload scrambler is used modern standards use 43 bit X43 + 1 run continuously on ATM payload bytes (suspended for 5 bytes of cell tax) run continuously on HDLC payloads

Z-43

Xn Yn = Xn + Yn-43


STS-1 OverheadSTS-1 Overhead

The STS-1 overhead consists of 3 rows of section overhead

– frame sync (A1, A2)– section trace (J0)– error control (B1)– section orderwire (E1)– Embedded Operations Channel (Di)

6 rows of line overhead– pointer and pointer action (Hi)– error control (B2)– Automatic Protection Switching signaling (Ki)– Data Channel (Di)– Synchronization Status Message (S1)– Far End Block Error (M0)– line orderwire (E2)

A1 A2 J0

B1 E1 F1

D1 D2 D3

H1 H2 H3

B2 K1 K2

D4 D5 D6

D7 D8 D9

D10 D11 D12

S1 M0 E2

sectionoverhead

lineoverhead


STM-1 OverheadSTM-1 Overhead

A1 A1 A1 A2 A2 A2 J0 res res

B1 m m E1 m F1 res res

D1 m m D2 m D3

B2 B2 B2 K1 K2

D4 D5 D6

D7 D8 D9

D10 D11 D12

S1 M1 E2

RSOH

MSOH

SOH

m – media dependent(defined for SONET radio)

res – reserved for national use

AU pointers


A1, A2, J0 A1, A2, J0 (section overhead)(section overhead)

A1, A2 - framing bytes A1 = 11110110 A2 = 00101000SONET/SDH framing always uses equal numbers of A1 and A2 bytes

J0 - regenerator section trace (in early SONET - a counter called C1)

enables receiver to be sure that the section connection is still OKenables identifying individual STS/STMs after muxing

J0 goes through a 16 byte sequenceMSBs are J0 framing (1000…00)

Cs are CRC-7 of previous frameS are 15 7-bit characters

section access point identifierSSSSSSS0

SSSSSSS0

C7C6C5C4C3C2C11

…


B1, E1, F1, D1-3 B1, E1, F1, D1-3 (section overhead)(section overhead)

B1 – Byte Interleaved Parity-8 byteeven parity of bits of bytes of previous frame after scrambling

only 1 BIT-8 for multiplexed STS/STM

E1 – section orderwire 64 kbps voice link for techniciansfrom regenerator to regenerator

F1 – 64 kbps link for user purposes

D1 + D2 + D3 – 192 kbps messaging channelused by section termination as Embedded Operations Channel (SONET)

or Data Communications Channel (SDH)


Pointers Pointers (line overhead)(line overhead)

In SONET, pointers are considered part of line overhead

For STS-1, H1+H2 is the pointer, H3 is the pointer action

H1+H2 indicates the offset (in bytes) from H3 to the SPE(i.e. if 0 then J1 POH byte is immediately after H3 in the row)

4 MSBs are New Data Flag, 10 LSBs are actual offset value (0 – 782)

When offset=522 the STS-1 SPE is in a single STS-1 frameIn all other cases the SPE straddles two frames

When offset is a multiple of 87, the SPE is rectangular

To compensate for clock differenceswe have pointer justification

When negative justification H3 carries the extra data

When positive justificationbyte after H3 is stuffing byte


SONET JustificationSONET JustificationIf tributary rate is above nominal, negative justification is needed

When less than 8 more bits than expected in buffer NDF is 0110 offset unchanged

When 8 extra bits accumulate NDF is set to 1001 extra byte placed into H3 offset is decremented by 1 (byte)

If tributary rate is below nominal, positive justification is neededWhen less than 8 fewer than expected bits in buffer

NDF is 0110 offset unchanged

When 8 missing bits NDF is set to 1001 byte after H3 is stuffing offset is incremented by 1 (byte)

H1 H2 extra …

H1 H2 H3 stuff …


B2, K1, K2, D4-D12 B2, K1, K2, D4-D12 (line overhead)(line overhead)

B2 – BIP-8 of line overhead + previous envelope (w/o scrambling)

N B2s for muxed STM-N

K1 and K2 are used for Automatic Protection Switching (see later)

D4 – D12 are a 576 Kbps Data Communications Channelbetween multiplexersusually manufacturer specific OAM functions


S1, M0, E2 S1, M0, E2 (line overhead)(line overhead)

S1 – Synchronization Status Messageindicates stratum level (unknown, stratum 1, …, do not use)

M0 – Far End Block Errorindicates number of BIP violations detected

E2 – line orderwire64 kbps voice link for techniciansfrom line mux to line mux


Payloads Payloads

andand

MappingsMappings


STS-1 HOP SPE structureSTS-1 HOP SPE structure

We saw that the pointer the line overhead points to the STS path overhead POH(after re-arranging) POH is one column of 9 rows (9 bytes = 576 kbps)


STS-1 HOPSTS-1 HOP

1 column of SPE is POH

2 more (“fixed stuffing”) columns are reserved

We are left with84 columns = 756 bytes = 48.384 Mbps for payload

This is enough for a E3 (34.368M) or a T3 (44.736M)

1 875930


STS-1 Path overheadSTS-1 Path overhead

1 column of overhead for path (576 Kbps)

POH is responsible for – path type identification– path performance monitoring– status (including of mapped payloads)– virtual concatenation– path protection– trace

J1

B3

C2

G1

F2

H4

F3

K3

N1

POH


J1, B3, C2 J1, B3, C2 (path overhead)(path overhead)

J1 – path traceenables receiver to be sure that the path connection is still OK

B3 – BIP-8 even bit parity of bytes (without scrambling)

of previous payload

C2 – path signal labelidentifies the payload type(examples in table)

C2 (hex)

Payload type

00 unequipped

01 nonspecific

02 LOP (TUG)

04 E3/T3

12 E4

13 ATM

16 PoS – RFC 1662

18 LAPS X.85

1A 10G Ethernet

1B GFP

CF PoS - RFC1619


G1, F2, H4, F3, K3, N1 G1, F2, H4, F3, K3, N1 (path overhead)(path overhead)

G1 – path statusconveys status and performance back to originator 4 MSBs are path FEBE, 1 bit RDI, 3 unused

F2 and F3 – user specific communications

H4 – used for LOP multiframe sync and VCAT (see later)

K3 (4 MSBs) – path APS

N1 – Tandem Connection Monitoring Messaging channel for tandem connections


LOPLOP

To carry lower rate payloads, divide the 84 available columns into 7 * 12 interleaved columns, i.e. 7 Virtual Tributary (VT) Groups

VT group is 12 columns of 9 rows, i.e. 108 bytes or 6.912 MbpsVT group is composed of VT(s) there are different types of VT in order to carry different types of payload all VTs in VT group must be of the same type (no mixing) but different VT groups in same SPE can have different VT typesA VT can have 3, 4, 6 or 12 columns

1 875930 1 2 3 4 5 6 77 VTGs


SONET/SDH : VT/VC typesSONET/SDH : VT/VC types

VT/STS VC column rate

payload

VT 1.5 VC-11 3 1.728 DS1 (1.544)

VT 2 VC-12 4 2.304 E1 (2.048)

VT 3 6 3.456 DS1C (3.152)

VT 6 VC-2 12 6.912 DS2 (6.312)

STS-1 VC-3 48.384 E3 (34.368)

STS-1 VC-3 48.384 DS3 (44.736)

STS-3c VC-4 149.760 E4 (139.264)

LOP

HOP

standard PDH rates map efficiently into SONET/SDH !

4 per group

3 per group

2 per group

1 per group

Y(J)S APS Slide 57

LO Path overheadLO Path overhead

LOP OH is responsible for timing, PM, REI, …

LO Path APS signaling is 4 MSBs of byte K4

V5

J2

N2

K4

V1 pointer

V2 pointer

V3 pointer

V4 pointer

VC11 – 25BVC12 – 34B

125 sec

500 sec

H4=XXXXXX00

H4=XXXXXX01

H4=XXXXXX10

H4=XXXXXX11VC11 – 27BVC12 – 36B


Payload capacityPayload capacity

VT1.5/VC-11 has 3 columns = 27 bytes = 1.728 Mbps

but 2 bytes are used for overhead (V1/V2/V3/V4 and V5/J2/N2/K4)

so actually only 25 bytes = 1.6 Mbps are available

Similarly

VT2/VC-12 has 4 columns = 36 bytes = 2.304 Mbps

but 2 bytes are used for overhead

So actually only 34 bytes = 2.176 Mbps are available


LOP overheadLOP overhead

V5 consists of BIP (2b) REI (1b) RFI (1b) Signal label (3b) (uneq, async, bit-sync, byte-sync, test, AIS) RDI (1b)

J2 is path trace

N2 is the network operator byte – may be used for LOP tandem connection monitoring (LO-TCM)

K4 is for LO VCAT and LO APS


SDH ContainersSDH Containers

Tributary payloads are not placed directly into SDH

Payloads are placed (adapted) into containers

The containers are made into virtual containers (by adding POH)

Next, the pointer is used – the pointer + VC is a TU or AU

Tributary Unit adapts a lower order VC to high order VC

Administrative Unit adapts higher order VC to SDH

TUs and AUs are grouped together until they are big enough

We finally get an Administrative Unit Group

To the AUG we add SOH to make the STM frame


Formally …Formally …

C-n n = 11, 12, 2, 3, 4

VC-n = POH + C-n

TU-n = pointer + VC-n (n=11, 12, 2, 3)

AU-n = pointer + VC-n (n=3,4)

TUG = N * TU-n

AUG = N * AU-n

STM-N = SOH + AUG


MultiplexingMultiplexingAn AUG may contain a VC-4 with an E4

or it may contain 3 AU-3s each with a VC-3s with an E3

In the latter case, the AU pointer points to the AUGand inside the AUG are 3 pointers to the AU-3s

J1B3C2G1F2H4F3K3N1

H1 H1H1 H2 H2H2 H3 H3H3


More multiplexingMore multiplexing

Similarly, we can hierarchically build complex structures

Lower rate STMs can be combined into higher rate STMs

AUGs can be combined into STMs

AUs can be combined into AUGs

TUGs can be combined into high order VCs

Lower rate TUs can be combined into TUGs

etc.

But only certain combinations are allowed by standards


All SDH mappingsAll SDH mappingsSTM-N

AU-3 VC-3 C3

VC-3TU-3TUG-3

C-4VC-4AU-4AUG…

AUG

AUG

C2

C12

C11

TUG-2 VC-2TU-2

VC-12TU-12

VC-11TU-11

STM-0

ATM 2.144 M

E4 139.264 M

ATM 1.6 M

ATM 149.760M

ATM 48.384 M

ATM 6.874M

E3 34.368 MT3 44.736 M

T2 6.312 M

E1 2.048 M

T1 1.544 M

* 3

*7

* 3

*7

* 4

* 3


All SONET mappingsAll SONET mappingsSTS-N STS-3 SPESTS-3c

STS-1

VT6 SPE

VT2 SPE

VT1.5 SPE

VT6

VT-2

VT1.5

ATM 2.144 M

E4 139.264 M

ATM 1.6 M

ATM 149.760M

ATM 48.384 M

ATM 6.874M

E3 34.368 MT3 44.736 M

T2 6.312 M

E1 2.048 M

T1 1.544 M

*N STS-1 SPE

VTG

*7

pointer processing

* 3

* 4


Tributary mapping typesTributary mapping types

When mapping tributaries into VCs, PDH-like bit-stuffing is used

For E1 and T1 there are several options Asynchronous mapping (framing-agnostic)

Bit synchronous mapping Byte synchronous mapping (time-slot aligned)

E4 into VC-4, E3/T3 into VC-3 are always asynchronous

T1 into VC-11 may be any of the 3 (in byte synchronous the framing bit is placed in the VC overhead)

E1 into VC-12 may be asynchronous or byte synchronous


WAN-PHY WAN-PHY (10 GbE in STM-64)(10 GbE in STM-64)

There is a special case where the bit-rates work out relatively wellGbE 10GBASE-R (64B/66B coding) can be directly mapped

into a STM-64 (with contiguous concatenation - see later) without need for GFPMAC creates "stretched InterPacket Gap" to compensate for rate being < 10GThis is the fastest connection commonly used for Internet trafficComplication: SDH clock accuracy is 4.6 ppm, GbE accuracy is 20 ppm

64*(270-9) = 16704 columns

J1

63 columns of fixed stuff

10GBASE-W 802.3-2005 Clause 50


Protection Protection

and and

Rings Rings


What is protection ?What is protection ?SONET/SDH need to be highly reliable (five nines)Down-time should be minimal (less than 50 msec)So systems must repair themselves (no time for manual intervention)

Upon detection of a failure (dLOS, dLOF, high BER)the network must reroute traffic (protection switching)from working channel to protection channel

The Network Element that detects the failure (tail-end NE)initiates the protection switching

The head-end NE must change forwarding or to send duplicate traffic

Protection switching is unidirectionalProtection switching may be revertive (automatically revert to working channel)

head-end NE tail-end NE

working channel

protection channel


How does it work?How does it work?

Head-end and tail-end NEs have bridges (muxes)Head-end and tail-end NEs maintain bidirectional signaling channel

Signaling is contained in K1 and K2 bytes of protection channel K1 – tail-end status and requests K2 – head-end status

head-end bridge tail-end bridgeworking channel

protection channel signaling channel


Linear 1+1 protectionLinear 1+1 protectionSimplest form of protectionCan be at OC-n level (different physical fibers)

or at STM/VC level (called SubNetwork Connection Protection)or end-to-end path (called trail protection)

Head-end bridge always sends data on both channelsTail-end chooses channel to use based on BER, dLOS, etc.

No need for signalingIf non-revertive

there is no distinction between working and protection channels BW utilization is 50%

channel A

channel B


Linear 1:1 protectionLinear 1:1 protectionHead-end bridge usually sends data on working channelWhen tail-end detects failure it signals (using K1) to head-endHead-end then starts sending data over protection channel

When not in useprotection channel can be used for (discounted) extra traffic (pre-emptible unprotected traffic)

May be at any layer (only OC-n level protects against fiber cuts)

working channel

protection channel

extra traffic


Linear 1:N protectionLinear 1:N protection

In order to save BWwe allocate 1 protection channel for every N working channels

N limited to 144 bits in K1 byte from tail-end to head-end – 0 protection channel – 1-14 working channels – 15 extra traffic channel

working channels

protection channel


Two fiber vs. Four-fiber ringsTwo fiber vs. Four-fiber ringsRing based protection is popular in North America (100K+ rings)Full protection against physical fiber cutsSimpler and less expensive than mesh topologiesProtection at line (multiplexed section) or path layerFour-fiber rings

fully redundant at OC levelcan support bidirectional routing at line layer

Two-fiber ringssupport unidirectional routing at line layer

2 fibers in opposite directions


Unidirectional vs. bidirectionalUnidirectional vs. bidirectionalUnidirectional routing

working channel B-A same direction (e.g. clockwise) as A-Bmanagement simplicity: A-B and B-A can occupy same timeslotsInefficient: waste in ring BW and excessive delay in one direction

Bidirectional routingA-B and B-1 are opposite in directionboth using shortest routespatial reuse: timeslots can be reused in other sections

A

BA-B

B-A

A

BB-A

A-B

C

B-C

C-B


UPSR vs. BLSR UPSR vs. BLSR (MS-SPRing)(MS-SPRing)

Of all the possible combinations, only a few are in use

Unidirectional Path Switched Ringsprotects tributariesextension of 1+1 to ring topology

Bidirectional Line Switched Rings (two-fiber and four-fiber versions)called Multiplex Section Shared Protection Ring in SDHsimultaneously protects all tributaries in STMextension of 1:1 to ring topology

Path switching

Line switching

Two-fiber

Four-fiber

Unidirectional

Bidirectional

UPSR

BLSR


UPSRUPSRWorking channel is in one direction

protection channel in the opposite direction

All traffic is added in both directions decision as to which to use at drop point (no signaling)

Normally non-revertive, so effective two diversity paths

Good match for access networks1 access resilient ring

less expensive than fiber pair per customer

Inefficient for core networksno spatial reuse

every signal in every spanin both directions

node needs to continuously monitorevery tributary to be dropped


BLSRBLSR

Switch at line level – less monitoring

When failure detected tail-end NE signals head-end NE

Works for unidirectional/bidirectional fiber cuts, and NE failures

Two-fiber versionhalf of OC-N capacity devoted to protectiononly half capacity available for traffic

Four-fiber versionfull redundant OC-N devoted to protectiontwice as many NEs as compared to two-fiber

Examplerecovery from unidirectional fiber cut


VCATVCAT

and and

LCASLCAS


ConcatenationConcatenationPayloads that don’t fit into standard VT/VC sizes can be accommodated

by concatenating of several VTs / VCs

For example, 10 Mbps doesn’t fit into any VT or VCso w/o concatenation we need to put it into an STS-1 (48.384 Mbps)the remaining 38.384 Mbps can not be used

We would like to be able to divide the 10 Mbps among 7 VT1.5/VC-11 s = 7 * 1.600 = 11.20 Mbps or5 VT2/VC-12 s = 5 * 2.176 = 10.88 Mbps


Concatenation Concatenation (cont.)(cont.)There are 2 ways to concatenate X VTs or VCs: Contiguous Concatenation (G.707 11.1)

– HOP – STS-Nc (SONET) or VC-4-Nc (SDH)or LOP – 1-7 VC-2-Nc into a VC-3– since has to fit into SONET/SDH payload

only STS-Nc : N=3 * 4n or VC-4-Nc : N=4n

– components transported together and in-phase– requires support at intermediate network elements

Virtual Concatenation (VCAT G.707 11.2) – HOP – STS-1-Xv or STS-Nc-Xv (SONET) or VC-3/4-Xv (SDH)or LOP – VT-1.5/2/3/6-Xv (SONET) or VC-11/12/2-Xv (SDH)– HOP: X ≤ 256 LOP: X ≤ 64 (limitation due to bits in header)– payload split over multiple STSs / STMs– fragments may follow different routes– requires support only at path terminations– requires buffering and differential delay alignment


Contiguous Concatenation: STS-3cContiguous Concatenation: STS-3c270 columns

9 ro

ws …

9 columns of section and

line overhead

3 columns of path overhead

258 columns of SPE STS-3

270 columns

9 ro

ws …

9 columns of section and

line overhead

1 column of path overhead

260 columns of SPE STS-3c

258 columns * 0.576 = 148.608 Mbps

260 columns * 0.576 = 149.760 Mbps


STS-N vs. STS-NcSTS-N vs. STS-Nc

Although both have raw rates of 155.520 Mbps

STS-3c has 2 more columns (1.152Mbps) available

More generally, For STS-Nc gains (N-1) columnse.g. STS-12c gains 11 columns = 6.336Mbps vis a vis STS-12STS-48c gains 47 columns = 27.072 MbpsSTS-192c gains 191 columns = 110.016 Mbps !

However, an STS-Nc signal is not as easily separablewhen we want to add/drop component signals


Virtual ConcatenationVirtual Concatenation

VCAT is an inverse multiplexing mechanism (round-robin)VCAT members may travel along different routes in SONET/SDH networkIntermediate network elements don’t need to know about VCAT

(unlike contiguous concatenation that is handled by all intermediate nodes)

…

H4


SDH virtually concatenated VCsSDH virtually concatenated VCs

So we have many permissible rates

1.600, 2.176, 3.200, 4.352, 4.800, 6.400, 6.528, 6.784, 8.000, …

VC Capacity (Mbps) if all members in one VC

VC-11-Xv 1.600, 3.200, … 1.600X

in VC-3 X ≤ 28 C ≤ 44.800

in VC-4 X ≤ 64 C ≤ 102.400

VC-12-Xv 2.176, 4.352, … 2.176X

in VC-3 X ≤ 21 C ≤ 45.696

in VC-4 X ≤ 63 C ≤ 137.088

VC-2-Xv 6.784, 13.568, …, 6.784X

in VC-3 X ≤ 7 C ≤ 47.448

in VC-4 X ≤ 21 C ≤ 142.464


SONET virtually concatenated VTsSONET virtually concatenated VTsVT Capacity (Mbps) If all members in one STS

VT1.5-Xv 1.600, 3.200, … 1.600X in STS-1 X ≤ 28 C ≤ 44.800

in STS-3c X ≤ 64 C ≤ 102.400

VT2-Xv 2.176, 4.352, … 2.176X in STS-1 X ≤ 21 C ≤ 45.696

in STS-3c X ≤ 63 C ≤ 137.088

VT3-Xv 3.328, 6.656, … 3.328X in STS-1 X ≤ 14 C ≤ 46.592

in STS-3c X ≤ 42 C ≤ 139.776

VT6-Xv 6.784, 13.568, … 6.784X in STS-1 X ≤ 7 C ≤ 47.448

in STS-3c X ≤ 21 C ≤ 142.464

So we have many permissible rates

1.600, 2.176, 3.200, 3.328, 4.352, 4.800, 6.400, 6.528, 6.656, 6.784, …


Efficiency comparisonEfficiency comparison

Using VCAT increases efficiency to close to 100% !

rate w/o VCAT efficiency with VCAT efficiency

10 STS-1 21% VT2-5v

VC-12-5v

92%

100 STS-3c

VC-4

67% STS-1-2v

VC-3-2v

100%

1000 STS-48c

VC-4-16c

42% STS-3c-7v

VC-4-7v

95%


PDH VCATPDH VCAT

Recently ITU-T G.7043 expanded VCAT to E1,T1,E3,T3Enables bonding of up to 16 PDH signals to support higher ratesOnly bonding of like PDH signals allowed (e.g. can’t mix E1s and T1s)

Multiframe is always per G.704/G.832 (e.g. T1 – ESF 24 frames, E1 16 frames)

1 byte per multiframe is VCAT overhead (SQ, MFI, MST, CRC)

Supports LCAS (to be discussed next)

TS0

1st frameof4 E1s

VCAToverhead

octet

timeeach E1


PDH VCAT overhead octetPDH VCAT overhead octet

There is one VCAT overhead octet per multiframe, so net rate isT1: (24*24-1=) 575 data bytes per 3 ms. multiframe = 191.666 kB/sE1: (16*30-1=) 495 data bytes per 2 ms multiframe = 247.5 kB/sT3 and E3 can also be usedWe will show the overhead octet format later

(when using LCAS, the overhead octet is called VLI)

TS0

frames of an E1

VCAToverhead

octet

…


Delay compensationDelay compensation802.1ad Ethernet link aggregation cheats

– each identifiable flow is restricted to one link– doesn’t work if single high-BW flow

VCAT is completely general– works even with a single flow

VCG members may travel over completely separate pathsso the VCAT mechanism must compensate for differential delay

Requirement for over ½ second compensation

Must compensate to the bit level

but since frames have Frame Alignment Signalthe VCAT mechanism only needs to identify individual frames


VCAT bufferingVCAT buffering

Since VCAT components may take different paths

At egress the members are no longer in the proper temporal relationship

VCAT path termination function buffers membersand outputs in proper order (relying on POH sequencing)(up to 512 ms of differential delay can be tolerated)

VCAT defines a multiframe to enable delay compensation– length of multiframe determines delay that can be accommodated

H4 byte in member’s POH contains : sequence indicator (identifies component) (number of bits limits X) MFI multiframe indicator (multiframe sequencing to find differential delay)


Multiframes and superframesMultiframes and superframesHere is how we compensate for 512 ms of differential delay512 ms corresponds to a superframe is 4096 TDM frames (4096*0.125m=512m)

For HOP SDH VCAT and PDH VCAT (H4 byte or PDH VCAT overhead)

The basic multiframe is 16 framesSo we need 256 multiframes in a superframe (256*16=4096)The MultiFrame Indicator is divided into two parts: MFI1 (4 bits) appears once per frame

– and counts from 0 to 15 to sequence the multiframe MFI2 (8bits) appears once per multiframe

– and counts from 0 to 255

For LOP SDH (bit 2 of K4 byte)– a 32 bit frame is built and a 5-bit MFI is dedicated– 32 multiframes of 16 ms give the needed 512 ms


LLinkink C Capacityapacity A Adjustmentdjustment S Schemecheme

LCAS is defined in G.7042 (also numbered Y.1305)LCAS extends VCAT by allowing dynamic BW changesLCAS is a protocol for dynamic adding/removing of VCAT members

– hitless BW modification– similar to Link Aggregation Control Protocol for Ethernet links

LCAS is not a “control plane” or “management” protocol– it doesn’t allocate the members– still need control protocols to perform actual allocation

LCAS is a “handshake” protocol– it enables the path ends to negotiate the additional / deletion – it guarantees that there will be no loss of data during change– it can determine that a proposed member is ill suited– it allows automatic removal of faulty member


LCAS – how does it work?LCAS – how does it work?LCAS is unidirectional (for symmetric BW need to perform twice)

LCAS functions can be initiated by source or sink

LCAS assumes that all VCG members are error-free

– LCAS messages are CRC protected

LCAS messages are sent in advance – sink processes messages after differential compensation– message describes link state at time of next message– receiver can switch to new configuration in time

LCAS messages are in the upper nibble of– H4 byte for HOS SONET/SDH– K4 byte for LOS SONET/SDH– VCAT overhead octet for PDH – VCAT and LCAS Information

LCAS messages employ redundancy– messages from source to sink are member specific– messages from sink to source are replicated

J1

B3

C2

G1

F2

H4

F3

K3

N1

POH


LCAS control messagesLCAS control messages

LCAS adds fields to the basic VCAT ones

Fields in messages from source to sink:– MFI MultiFrame Indicator– SQ SeQuence indicator (member ID inside VCAT group)– CTRL ConTRoL (IDLE, being ADDed, NORMal, End of Sequence, Do Not Use)– GID Group Identification (identifies VCAT group)

Fields in messages from sink to source (identical in all members):– MST Member Status (1 bit for each VCG member)– RS-Ack ReSequence Acknowledgement

Fields in both directions– CRC Cyclic Redundancy Code

The precise format depends on the VCAT type (H4, K4, PDH)

Note: for H4 format SQ is 8 bits, so up to 256 VCG members for PDH SQ is only 4 bits, so up to 16 VCG members


H4 formatH4 format

MFI2 bits 1-4 0 0 0 0 MFI2 bits 5-8 0 0 0 1

CTRL 0 0 1 0 0 0 0 GID 0 0 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 1

CRC-8 bits 1-4 0 1 1 0CRC-8 bits 5-8 0 1 1 1

MST bits 1 0 0 0more MST bits 1 0 0 1

0 0 0 RS-ACK 1 0 1 0 0 0 0 0 1 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 1 1 0 1

SQ bits 1-4 1 1 1 0SQ bits 5-8 1 1 1 1

16 frame m

ultiframeMFI1

rese

rved

fiel

dsre

serv

ed fi

elds


H4 format – some commentsH4 format – some commentsCRC-8 (when using K4 it is CRC-3)

– covers the previous 14 frames (not sync’ed on multiframe)– polynomial x8 + x2 + x + 1

MST– each VCG member carries the status of all members– so we need 256 bits of member status– this is done by muxing MST bits– there are MST bits per multiframe– and 32 multiframes in an MST multiframe– no special sequencing, just MFI2 multiframe mod 32

GID– single bit indentifier– all members of VCG share the same bit– cycles through 215-1 LFSR sequence– different VCGs use different phase offsets of sequence


LCAS – adding a member (1)LCAS – adding a member (1)When more/less BW is needed, we need to add/remove VCAT members

Adding/removing VCAT members first requires provisioning (management)

LCAS handles member sequence numbers assignment

LCAS ensures service is not disrupted

Example: to add a 4th member to group “1”

Initial state:

Step 1: NMS provisions new member

source sends CTRL=IDLE for new member

sink sends MST=FAIL for new member

GID=g SQ=1 CTRL=NORM


GID=g SQ=3 CTRL=EOS



GID=g SQ=3 CTRL=EOS

GID=g SQ=FF CTRL=IDLE


LCAS – adding a member (2)LCAS – adding a member (2)Step 2: source sends CTRL=ADD and SQ

sink sends MST=OK for new member if it has been provisioned if receiving new member OK if it is able to compensate for delay

otherwise it will send MST=FAILand source reports this to NMS

Step 3: source sends CTRL=EOS for new member

new member starts to carry traffic

sink sends RS-ACK

Note 1: several new members may be added at onceNote 2: removing a member is similar

Source puts CTRL=IDLE for member to be removed and stops using it All member sequence numbers must be adjusted



GID=g SQ=3 CTRL=EOS

GID=g SQ=4 CTRL=ADD




GID=g SQ=4 CTRL=EOS


LCAS – service preservationLCAS – service preservationTo preserve service integrity if sink detects a failure of a VCAT memberLCAS can temporarily remove member (if service can tolerate BW reduction)

Example: Initial state

Step 1: sink sends MST=FAIL for member 2 source sends CTRL=DNU (special treatment if EoS) and ceases to use member 2Note: if EoS fails, renumber to ensure EoS is active

Step 2: sink sends MST=OK indicating defect is cleared source returns CTRL to NORM and starts using the member again Note: if NMS decides to permanently remove the member, proceed as in previous slide




GID=g SQ=4 CTRL=EOS


GID=g SQ=2 CTRL=DNU


GID=g SQ=4 CTRL=EOS


HandlingHandling

PacketPacket

DataData


Packet over SONETPacket over SONET

Currently defined in RFC2615 (PPP over SONET) obsoletes RFC1619

SONET/SDH can provide a point-to-point byte-oriented full-duplex synchronous link

PPP is ideal for data transport over such a link

PoS uses PPP in HDLC framing to provide a byte-oriented interfaceto the SONET/SDH infrastructure

POH signal label (C2) indicates PoS as C2=16 (C2=CF if no scrambler)


PoS architecturePoS architecture

PoS is based on PPP in HDLC framing

Since SONET/SDH is byte oriented, byte stuffing is employed

A special scrambler is used to protect SONET/SDH timing

PoS operates on IP packets

If IP is delivered over Ethernet– the Ethernet is terminated (frame removed)– Ethernet must be reconstituted at the far end– require routers at edges of SONET/SDH network

IPPPP

HDLCSONET/SDH


PoS DetailsPoS Details

IP packet is encapsulated in PPP– default MTU is 1500 bytes– up to 64,000 bytes allowed if negotiated by PPP

FCS is generated and appendedPPP in HDLC framing with byte stuffing43 bit scrambler is run over the SPEbyte stream is placed octet-aligned in SPE

– (e.g. 149.760 Mbps of STM-1)– HDLC frames may cross SPE boundaries


POS problemsPOS problems

PoS is BW efficient

but POS has its disadvantages BW must be predetermined HDLC BW expansion and nondeterminacy BW allocation is tightly constrained by SONET/SDH capacities

– e.g. GBE requires a full OC-48 pipe POS requires removing the Ethernet headers

– so lose RPR, VLAN, 802.1p, multicasting, etc POS requires IP routers


LAPSLAPS

In 2001 ITU-T introduced protocols for transporting packets over SDH X.85 IP over SDH using LAPS X.86 Ethernet over LAPS

Built on series of ITU “LAPx” HDLC-based protocols

Use ISO HDLC format

Implement connectionless byte-oriented protocols over SDH

X.85 is very close to (but not quite) IETF PoS


GFP architectureGFP architectureA new approach, not based on HDLC

Defined in ITU-T G.7041 (also numbered Y.1303)originally developed in T1X1 to fix ATM limitations(like ATM) uses HEC protected frames instead of HDLC

Client may be PDU-oriented (Ethernet MAC, IP) or block-oriented (GBE, fiber channel)

GFP frames– are octet aligned– contain at most 65,535 bytes– consist of a header + payload area

Any idle time between GFP frames is filled with GFP idle frames

Ethernet IP otherGFP – client specific part

GFP – common partSDH OTN other

HDLC


GFP frame structureGFP frame structure

Every GFP frame has a 4-byte core header– 2 byte Payload Length Indicator PLI = 01,2,3 are for control frames

– 2 byte core Header Error Control X16 + X12 + X5 + 1– entire core header is XOR’ed with B6AB31E0

Idle GFP frames – have PLI=0 – have no payload area

Non-idle GFP frames – have ≥ 4 bytes in payload area– the payload has its own header– 2 payload modes : GFP-F and GFP-T– optionally protect payload with CRC-32

PLI (2B)cHEC (2B)

payload header (4-64B)

payload

optional payloadFCS (4B)

coreheader

payloadarea


GFP payload headerGFP payload headerGFP payload header has

– type (2B)– type HEC (CRC-16)– extension header (0-60B)

either null or linear extension (payload type muxing)– extension HEC (CRC-16)

type consists of– Payload Type Identifier (3b)

PTI=000 for client data PTI=100 for client management (OAM dLOS, dLOF)

– Payload FCS Indicator (1b) PFI=1 means there is a payload FCS

– Extension Header ID (3b)– User Payload Identifier (8b)

values for Ethernet, IP, PPP, FC, RPR, MPLS, etc.

type (2B)tHEC (2B)

extension header (0-60B)

eHEC (2B)

UPI (8b)

PTI (3b) EXI (3b)PFI


GFP modes GFP modes GFP-F - frame mapped GFPGood for PDU-based protocols (Ethernet, IP, MPLS)

or HDLC-based ones (PPP)

Client PDU is placed in GFP payload field

GFP-T – transparent GFPGood for protocols that exploit physical layer capabilities

In particular8B/10B line code used in fiber channel, GbE, FICON, ESCON, DVB, etc

Were we to use GFP-F would lose control info, GFP-T is transparent to these codes

Also, GFP-T needn’t wait for entire PDU to be received (adding delay!)

SONET

Documents

timeslotss sonet siide7and

phases s sonet siide9the

tdm switch

tdm timeslots

ior duration

complexthe tdm

cost routing algorithms

tdm signals tributariese