Chapter 1

Network Architecture

© 2001, Cisco Systems, Inc.

 Objectives

Upon completion of this chapter, you will

be able to perform the following tasks:
• Identify the main terms used to describe
network components
• Describe the link structure and network
elements
• Describe the interface options and interface
layers

© 2001, Cisco Systems, Inc. Network Architecture-2

 Agenda

1.1 - Link Structure and Line Interfaces

1.2 - Network Elements
Summary, Information Resources

© 2001, Cisco Systems, Inc. Network Architecture-3

 Section 1.1
Link Structure and
Line Interfaces
© 2001, Cisco Systems, Inc.
 Objectives
Upon completion of this section, you will be
able to perform the following tasks:
• Describe the functionality and interaction of the
interface layers
• Define the three overhead layers
• Describe the topology concepts related to the
overhead layers
• Describe the main features of electrical and optical
interfaces

© 2001, Cisco Systems, Inc. Network Architecture-5

 Overhead Layer Concepts

path
multiplex section multiplex section

regenerator regen. regen. regenerator

section section section section

ADM
PTE REG or REG PTE
DCS
path regen. section multipl. section regen. section path
termination termination termination termination termination

PTE = path terminating element

service (E1, E4..) MUX = terminal multiplexer
mapping service (E1, E4..)
demapping REG = regenerator mapping
ADM = add/drop multiplexer demapping
DCS = digital cross-connect system
© 2001, Cisco Systems, Inc. Network Architecture-6
 Regenerator Section
• Regeneration section layer is the lowest level of link
components in a SDH network
• Deals with the transport of an STM-N frame across the
physical medium
• Point-to-point connection between two regeneration
section termination points with direct optical or
electrical domain connectivity
• Terminated by Regenerator Section Terminating
Equipment (RSTE)
• The Regeneration section is mainly designed to
overcome physical limitations of the transport
technology

© 2001, Cisco Systems, Inc. Network Architecture-7

 Multiplex Section
• One or more consecutive regenerator sections might
compose a multiplex section
– Main element to build different topologies (e.g. ring)
• Deals with the transport of path layer payloads across
the physical medium
• Multiplex section is a point-to-point logical link that
connects to ADM, MUX, or DCS devices
– These devices might not include a path termination
• Overhead is interpreted and modified by Multiplex
Section Terminating Equipment (MSTE)
– Multiplex section (MS) overhead is accessed only
after the section overhead has been first terminated
© 2001, Cisco Systems, Inc. Network Architecture-8
 Path
• One or more connected multiplex sections may provide
a transport service for a path
– Multiplex section may carry multiple paths by
multiplexing
• Deals with the transport of various payloads between
SDH terminal multiplexing equipment
• Path layer maps payloads into the format required by
the MS Layer
• Communicates end-to-end via the Path Overhead (POH)
• POH is terminated and modified by Path Terminating
Equipment (PTE)
– Regenerator and multiplex section overhead must be
terminated to access the overhead
© 2001, Cisco Systems, Inc. Network Architecture-9
 HO and LO Paths
• In SDH the PDH payload multiplexing is done at 2
different layers
• High-order (HO) path carries E3/E4 or similar payloads
– Organized into administrative units (AU) including
higher order tributaries
• Low-order (LO) path carries E1/E2 or similar payloads
– Organized into tributary units (TU) including lower
order tributaries

© 2001, Cisco Systems, Inc. Network Architecture-10

 Topology Concepts
• SDH topologies are designed for providing a flexible and
reliable transport for required paths
• Main issues:
– Capacity planning, bandwidth provisioning
– Redundancy, automatic fail-over
– Delay and jitter control
• Typical topology concepts:
– Point-to-point links (with protection) and DCS/MUX
• Arbitrary complex topology may be built
– Interconnected protected rings with ADM/DCS
• Minimum resource usage (physical media) for avoiding
single point of failures
© 2001, Cisco Systems, Inc. Network Architecture-11
 Physical Layer - I.

Services (E1, E2, E3,

E4, Video, etc.)
Layers
Map Payload and Payload and Path Overhead
Path OH into VC Path

Map VC and MS VC and MS Overhead

OH into internal Multiplex
signal Section
STM-N
Map internal signal
and RS OH Signal Regen. Regen.
Section
into STM-N signal Section
Light
Optical Conversion Pulse Physical Photonic

Terminal Regenerator Terminal

© 2001, Cisco Systems, Inc. Network Architecture-12

 Physical Layer - II.
• Line coding applied:
– CMI for electrical interfaces
• Guarantees transmit-receive clock
synchronization
– Binary NRZ for optical interfaces
• May change for very high speeds (STM-256 or
higher) into RZ solitons
• Does not guarantee enough 1-0 or 0-1 changes,
and thus clock transmit-receive synchronization
– Depends on the frame content
– Scrambling is needed for a guarantee

© 2001, Cisco Systems, Inc. Network Architecture-13

 Electrical Interfaces
• Defined to be as compatible as possible with
existing PDH physical interfaces
– Same hardware should be used
• For intra-office applications only
– Maximum 150 m 75 Ohm coax for STM-1
• 155.520 Mbit/s, CMI line coding

© 2001, Cisco Systems, Inc. Network Architecture-14

 Optical Interfaces
• Intra-office (application code: I-<n>)
– LED or MLM laser at 1310 nm or 1550 nm
– Up to 2 km, max. loss 7-12 dB
• Inter-office, short-haul (application code: S-<n.w>)
– Low power SLM or MLM laser at 1310 nm or 1550 nm
– Up to 15 km, max. loss 12 dB
• Inter-office, long-haul (application code: L-<n.w>)
– High power SLM or MLM laser at 1310 nm or 1550 nm
(zero-dispersion or dispersion-shifted fiber)
– Up to 40-60 km, loss: 10-28 dB up to STM-1, 10-24 dB
up to STM-16
© 2001, Cisco Systems, Inc. Network Architecture-15
 Summary

• Describe the functionality and

interaction of the interface layers
• Define the three overhead layers
• Describe the topology concepts related
to the overhead layers
• Describe the main features of electrical
and optical interfaces

© 2001, Cisco Systems, Inc. Network Architecture-16

 Review Questions
• How many layers are used to build up a SDH
network?
• What is the purpose of the Multiplex Section layer?
• What is the purpose of the HO Path overhead and
the LO Path overhead ?
• Why do electrical and optical interfaces have
different line coding?
• Is there a Regenerator Section termination in a
Terminal Multiplexer?
• Is a usual add-drop multiplexer also a Path
terminating equipment?
© 2001, Cisco Systems, Inc. Network Architecture-17
 Section 1.2

Network Elements

© 2001, Cisco Systems, Inc.

 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify main network concepts

• Describe the functions of typical network

elements

© 2001, Cisco Systems, Inc. Network Architecture-19

 Network Concepts
• Networks should be designed by
decomposition
– Service needs into layers
– Logical connectivity needs into subnetworks
• Devices might be categorized by functionality
and role in the network layers and topology
• Multiple set of functionality may be integrated
into a single device if it is economically
feasible

© 2001, Cisco Systems, Inc. Network Architecture-20

 Terminal Multiplexer
• Terminal multiplexer is at the edge of the SDH
network
– Provides connectivity to the PDH network
devices and certain end-user equipment
• It includes a regenerator section, multiplex
section, and path termination in one link

© 2001, Cisco Systems, Inc. Network Architecture-21

 Regenerator
• A regenerator simply extends the possible
distance and quality of a line by decomposing
it into multiple sections
– Replaces regenerator section overhead
– Multiplex section and path overhead is not
altered

© 2001, Cisco Systems, Inc. Network Architecture-22

 Add-drop Multiplexer - I.
• Add/drop multiplexer (ADM)
– Main element for configuring paths on top of line
topologies (point-to-point or ring)
– Multiplexed channels may be dropped and added
– Special drop and repeat mode for broadcast and
survivability
– An ADM has at least 3 logical ports: 2 core and 1 or
more add-drop
• Ports have different roles
• No switching between the core ports
• Switching only between the add-drop and the core
ports
© 2001, Cisco Systems, Inc. Network Architecture-23
 Add-drop Multiplexer - II.
• ADM always includes regenerator and multiplex
section termination. However paths might not be
terminated, but only switched from one multiplex
section’s channel to another multiplex section’s
channel
• ADM may be integrated with terminal multiplexer
functionality for direct interfacing to non-SDH network
elements
• ADM always processes and replaces the multiplex
section overhead
– ADM may not change the path overhead (POH)
• POH is changed only if terminal multiplexer
function is included

© 2001, Cisco Systems, Inc. Network Architecture-24

 SDH Cross-connect
• ADM concept is extended to have many similar capacity
ports with any-to-any channel connectivity: the
resulting device is called a Digital Cross-connect (DCS)
– SDH DCS may have only 2 logical ports
• Pure SDH DCS may connect only STM-1 or higher
channels with each other
– Cross-connects are named after historical patch
panels interconnecting regenerator or multiplex
section termination devices
• Pure SDH DCS may not include path termination,
switching of channels is typically done at the multiplex
section layer

© 2001, Cisco Systems, Inc. Network Architecture-25

 Wideband Digital Cross-
connect
• SDH wideband digital cross-connect (WDCS) is
designed for interconnecting a large number of
channels at the LO path (e.g. E1 basic PDH)
level
• SDH WDCS has only SDH ports carrying a
large number of LO path payloads
• Interconnection of LO paths may be done
virtually, without a physical LO path
termination
– Provides an economical alternative to legacy
physical E1 cross-connects

© 2001, Cisco Systems, Inc. Network Architecture-26

 Broadband Digital Cross-
connect
• Broadband digital cross-connect (BDCS) uses
a transparent switch matrix for HO path
speeds to interconnect a large number of
channels
• HO path BDCS has a similar architecture to a
WDCS in an integrated device
• Many SDH ports carrying HO path payloads
– Interconnect without physical HO path
termination

© 2001, Cisco Systems, Inc. Network Architecture-27

 Subscriber Loop Access
System
• Subscriber loop access system (SLAS) is a
concentrator of low speed services
– Provides an efficient feed of subscribers into the
central office by using multiplexing to create a higher
speed trunk line
• By using SLASs at the edge of the network, the cross-
connects and ADMs can be optimized by having ports
with concentrated end-user services
• SLAS does not provide a local switching function
between the input channels, only aggregation

© 2001, Cisco Systems, Inc. Network Architecture-28

 Summary

• Identify main network concepts

• Describe the functions of typical

network elements

© 2001, Cisco Systems, Inc. Network Architecture-29

 Review Questions
• What is the purpose of a regenerator
equipment?

• What is a Subscriber Loop Access System

(SLAS)?

© 2001, Cisco Systems, Inc. Network Architecture-30

 Summary
Information Resources

© 2001, Cisco Systems, Inc.

 Questions

© 2001, Cisco Systems, Inc.

? Network Architecture-32
 Information Resources
• Books
– Mike Sexton, Andy Reid: “Broadband
Networking: ATM, SDH, and SONET”
• Artech House, 1997. ISBN 0-89006-578-0
• Web
– ITU-T standards
• http://www.itu.int
– ETSI standards
• http://www.etsi.org
© 2001, Cisco Systems, Inc. Network Architecture-33
 Summary

After completing this chapter, you should

be able to perform the following tasks:
• Identify the main terms used to describe
network components
• Describe the link structure and network
elements
• Describe the interface options and interface
layers

© 2001, Cisco Systems, Inc. Network Architecture-34

 Chapter 2

Frame Structure
 Objectives

Upon completion of this chapter, you will

be able to perform the following tasks:
• Identify the main frame concepts
• Describe the basic structure of frames at
various hierarchy levels
• Make basic computations for bit rates at
various hierarchy levels
• Describe the internal details of payloads and
overheads

© 2001, Cisco Systems, Inc. Network Architecture-36

 Agenda

2.1 - Frame Concept

2.2 - STM-1 Frames
2.3 - STM-n Frames
2.4 - Frames and Rates
2.5 - Payload Internals
2.6 - Overhead Internals
Summary, Information Resources

© 2001, Cisco Systems, Inc. Network Architecture-37

 Section 2.1

Frame Concept
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Describe the notation for electrical and optical
signals

• Describe the two-dimensional frame model

• Describe the main components of the frame

© 2001, Cisco Systems, Inc. Network Architecture-39

 Electrical and Optical Signals
• STM-<N> is electrical, STM-<N>O is optical
• STM-<N>c means concatenated
– Not multiplexed signal
– Originates at that speed
– Administrative overhead optimized
compared to real multiplexed signal
• Frame format is independent from electrical or
optical signals
– For simplicity we will always refer to STM-
<N> only

© 2001, Cisco Systems, Inc. Network Architecture-40

 Two-dimensional Frame Model
• Fixed 125 microseconds frame time length
– To support 8 KHz sampled voice applications
• Bytes organized into rows and columns
– Administrative channels are rate decoupled
for easier processing
• STM-1 frame is organized into 270 (3 x 90)
columns by 9 rows
– Frame size is 2430 bytes
– 9 x 270 bytes/frame x 8 bits/byte x 8000
frame/s = 155.52 Mbit/s

© 2001, Cisco Systems, Inc. Network Architecture-41

 SONET Compatibility
• STM-0 frame is defined to be compatible with STS-1 of
SONET
– Originally it was only a virtual tool to show SONET
compatibility (not part of ITU-T specifications in 1993)
– Recently it has become a valid real-life frame format for
microwave links (proposed by ETSI, included into new
merged ITU-T G.707 standard in 1996)
• STM-1 has a compatibly structure with STS-3 of SONET
• Although STM-0 is already defined, for historical reasons
most SDH discussions are based on STM-1
– In many places specifications use a multiplication by the
hierarchy level, so 0 cannot be used

© 2001, Cisco Systems, Inc. Network Architecture-42

 Overheads
• 3x3=9 columns section overhead (SOH) for STM-1
– Includes a complex set of OAM information
– 3 rows (27 bytes) for regenerator section (RS) overhead (RSOH)
– 1 row (9 bytes) for administrative unit (AU) pointer
– 5 rows (45 bytes) for multiplex section (MS) overhead (MSOH)
• Path overhead (POH)
– Provides framing for payload, not part of SOH

Path

ADM
Path Terminator REG REG Path Terminator
or DCS

R-Section R-Section R-Section R-Section

M-Section M-Section

© 2001, Cisco Systems, Inc. Network Architecture-43

 Payloads - I.
• Two main types of payloads:
– Multiplexed voice channels originated in PDH based
devices
– Transparent bit stream services
• May be used for data packet transport or ATM
• Payloads are organized into paths over the network
– Different path types based on the content
• STM path, HO or LO path etc.
• Payloads are managed by using the multiplex section
overhead (MSOH) and the path overhead (POH)

© 2001, Cisco Systems, Inc. Network Architecture-44

 Payloads - II.
• Payloads are put into a so called payload or VC (virtual
container) capacity
– Frames provide a higher bit rate than the payload
• Required to be able to compensate for frequency
differences
– Similar concept to PDH stuffing
• Payloads may arrive with very different phases
– Varying line delays cannot be avoided in a WAN with
long distances
• Light needs some time to reach from the
transmitter to the receiver that is much bigger than
the bit timing interval or the frame cycle time
© 2001, Cisco Systems, Inc. Network Architecture-45
 Pointers
• Pointers were included into SDH design to
provide tools to compensate for incoming
payload phase differences
– Without extensive buffering
• So not too much delay and jitter
• Advantage over PDH network where this
problem was solved by asynchronous
multiplexing and a lot of bad consequences
• Pointers make it possible to create ring
topologies (efficient fail-over redundancy) for
SDH
© 2001, Cisco Systems, Inc. Network Architecture-46
 Summary

• Describe the notation for electrical and

optical signals

• Describe the two-dimensional frame

model

• Describe the main components of the

frame

© 2001, Cisco Systems, Inc. Network Architecture-47

 Review Questions
• Why are SDH frames repeated every 125 micro-
seconds?

• Why are overhead channels distributed evenly in the

bit stream?

• Why are pointers needed to implement ring

topologies?

© 2001, Cisco Systems, Inc. Network Architecture-48

 Section 2.2

STM-1 Frames
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Describe the overall structure of the frame

• Describe the payload pointer solution for

timing differences

© 2001, Cisco Systems, Inc. Network Architecture-50

 General Structure
Order of transmission
1st 270 columns

2nd

Section
overhead VC Capacity
(SOH) (for AUG)

9 columns 261 columns

© 2001, Cisco Systems, Inc. Network Architecture-51
 Synchronous Payload
Envelope - I.
• SPE = Synchronous Payload Envelope is the common sense
name (also used for SONET)
• AUG = Administrative Unit Group is the official ITU-T name
for the SPE + 4. row of SOH (AU pointer)
– 261 columns x 9 rows
– POH is 1 column
– Fixed stuffing depends on internal structure of AUG
– 260 columns = 2340 bytes for AUG-4 payload capacity
• AUG may be composed by different type of administrative
units (AU)
– 1 x AU-4 (specific to carry PDH E4, not compatible with
SONET frame structure)
– Or 3 x AU-3 (compatible with SONET frame structure)
© 2001, Cisco Systems, Inc. Network Architecture-52
 Synchronous Payload
Envelope - II.
• Administrative unit is composed of a pointer and a
virtual container (VC)
– Pointer for VC-4 inside an AU-4 is the full 4. row of
SOH
– Pointer for VC-3 inside an AU-3 is determined by a
1:3 demultiplexing of the 4. row of SOH
• VC inside AU may begin anywhere in STM-1 VC
capacity
• AU (payload) pointer in SOH designates the first byte of
VC inside AU (SPE)

© 2001, Cisco Systems, Inc. Network Architecture-53

 STM-1 VC-3 Capacity
Structure
87 columns

Payload Capacity

Path overhead Path overhead

Fixed stuff
(POH) (POH)

1 column 30. column 59. column

© 2001, Cisco Systems, Inc. Network Architecture-54

 STM-1 VC-4 Capacity
Structure
261 columns

Payload Capacity

Path overhead
(POH)

1 column

© 2001, Cisco Systems, Inc. Network Architecture-55

 Payload Pointer
Payload Pointer marks
start of STM-1 VC-3 or
VC-4
90 (VC-3) or 270 (VC-4) Columns
STM-1 Frame #1

H1 H2 H3...

9
Rows STM-1
VC-3 or VC-4
STM-1 Frame #2 125 μsec

9
Rows
STM-1 VC-3 or VC-4
POH column
250 μsec
Section
© 2001, Cisco Systems, Inc. Overhead Network Architecture-56
 Summary

• Describe the overall structure of the

frame

• Describe the payload pointer solution

for timing differences

© 2001, Cisco Systems, Inc. Network Architecture-57

 Review Questions
• What is the size of a basic STM-1 frame?

• Which type of frame has been defined to show

compatibility with SONET ?

• What are the overheads carried by a STM-1 frame?

• What is the purpose of pointers in a SDH frame?

© 2001, Cisco Systems, Inc. Network Architecture-58

 Section 2.3

STM-n Frames
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Describe the overall structure of multiplexing
frames
• Describe the main steps of the multiplexing
process
• Identify the concept of concatenated payloads
• Determine the details of frame structures for a
common hierarchy level

© 2001, Cisco Systems, Inc. Network Architecture-60

 General Structure
• STM-N structure
– Byte-interleaving STM-1 modules
• No extra overhead introduced
• Overhead of multiplexed signals taken over, but section
overhead (SOH) should be replaced with new
information for the STS-N multiplex section
• Overhead is growing in absolute number of bits, but
relative size is the same
• New overhead is bigger than necessary for regenerator
and multiplex section overheads, so some bytes are
unused
– Section overhead (SOH) is frame aligned
– SPE (multiplexed VC-3 or VC-4 channels) is not frame
aligned

© 2001, Cisco Systems, Inc. Network Architecture-61

 STM-N frame

270 x N Columns

9xN
Columns

STM-N VC capacity
9
Rows

125 μsec
Section
Overhead

© 2001, Cisco Systems, Inc. Network Architecture-62

 Multiplexing Processes
• Multiplexing is composed of various processes:
– Mapping
• Tributaries adapted into Virtual Containers (VC) by
adding stuffing and POH
– Aligning
• Pointer is added to locate the VC inside an AU or
TU
– Multiplexing
• Interleaving the bytes of multiple paths
– Stuffing
• Adding up the fixed stuff bits to compensate for
frequency variances

© 2001, Cisco Systems, Inc. Network Architecture-63

 Concatenated Frames

X-1 Columns SDH terminology is using

X instead of N (X = N)

X x 260 Columns
STM
POH
9 bytes
9 Rows STS-Xc Payload Capacity
(AU-4-Xc)
Fixed
Stuff
(9X-9 STM-4c = 599.040 Mbit/s
bytes) STM-16c = 2396.160 Mbit/s

125 μsec

X x 261 Columns

© 2001, Cisco Systems, Inc. Network Architecture-64

 Frame Structures for Each
Common Hierarchy Level
270 Columns
STM-1

9 Rows 155.52 Mbit/s

1,080 Columns
STM-4

9 Rows 622.08 Mbit/s

4,320 Columns
STM-16

9 Rows
2488.32 Mbit/s

STM-64 9 rows x 17280 columns, 9953.28 Mbit/s

© 2001, Cisco Systems, Inc. Network Architecture-65

 Summary
• Describe the overall structure of multiplexing
frames
• Describe the main steps of the multiplexing
process
• Identify the concept of concatenated
payloads
• Determine the details of frame structures for
a common hierarchy level

© 2001, Cisco Systems, Inc. Network Architecture-66

 Review Questions
• What is the big advantage of the SDH multiplexing
concept over PDH in terms of overhead?

• Why is it important to have more fixed stuffing at

higher line rates?

• What is the concept behind allocating the amount

of fixed stuffing?

© 2001, Cisco Systems, Inc. Network Architecture-67

 Section 2.4

Frames and Rates

 Objectives
Upon completion of this section, you will
be able to perform the following tasks:
• Identify the various bit rates used to
characterize TDM networks

• Describe how the multiplexing process affects

the bit rates

• Make the computation to summarize the rate

hierarchy

© 2001, Cisco Systems, Inc. Network Architecture-69

 Line, SPE and Payload Rates
• Line rate = SOH + SPE
• SPE rate = POH + payload capacity + fixed
stuffing
• VC payload capacity rate = line rate - SOH -
POH - fixed stuffing
• Transparent bit-stream capacity rate = line rate
- SOH - POH
• Example for STM-1 frame line rate:
– 270 columns x 9 rows = 2430 bytes
– 8000 fps x 19440 bits = 155.52 Mbit/s

© 2001, Cisco Systems, Inc. Network Architecture-70

 Multiplexing Effect on Rates
• Because of synchronous TDM byte-multiplexing
the line rate is simply multiplied
• Example for STM-4 frame line rate:
– 4 x STM-1 byte-multiplexing
– 4 x 155.52 Mbit/s = 622.08 Mbit/s
• With higher level concatenated frames the
transparent bit-stream capacity rate increases
slightly as a relative value
– Since the POH is a fixed absolute value in this
case
– However, fixed stuff may use up this extra
capacity
© 2001, Cisco Systems, Inc. Network Architecture-71
 Rate Hierarchy

SONET
SDH Line Rate SPE Rate Optical Electrical
Level (Mbit/s) (Mbit/s) Level Level
STM-1 155.52 150.336 OC-3 STS-3
STM-4 622.08 601.344 OC-12 STS-12
STM-16 2488.32 2405.376 OC-48 STS-48
STM-64 9953.28 9621.504 OC-192 STS-192
STM-256 39813.12 38486.016 OC-768 STS-768

© 2001, Cisco Systems, Inc. Network Architecture-72

 Summary
• Identify the various bit rates used to characterize TDM
networks

• Describe how the multiplexing process affects the bit rates

• Make the computation to summarize the rate hierarchy

© 2001, Cisco Systems, Inc. Network Architecture-73

 Review Questions
• Is the SPE rate a good measure of the capacity
available to a data transport client?

• What is the VC payload capacity rate of a

concatenated 40 Gbit/s SDH signal? Demonstrate
the computation steps!

• How much frequency variation is allowed by the

fixed stuffing in a 10 Gbit/s SDH signal?
Demonstrate the computation steps!

© 2001, Cisco Systems, Inc. Network Architecture-74

 Section 2.5

Payload Internals
 Objectives
Upon completion of this section, you will
be able to perform the following tasks:
• Identify the need for special structures to
support a mixture of payloads

• Describe the grouping options of various

payloads

• Describe the pointer processing associated

with the individual payloads

© 2001, Cisco Systems, Inc. Network Architecture-76

 Administrative and Tributary
Units
• Multiplexing of PDH signals inside SDH is organized
using administrative and tributary units
– One more multiplexing level than SONET
• Administrative unit (AU) is a special construct for SDH,
which allows two alternative ways of multiplexing and
thus supporting PDH E4 mapping and SONET
compatibility at the same time
• Tributary unit is a construct which accommodates
various PDH signals at the lower level
• AU and TU are both called units, because both are
composed from a virtual container and a pointer
• Byte-multiplexed AUs and TUs are called an AU group
(AUG) and a TU Group (TUG) respectively

© 2001, Cisco Systems, Inc. Network Architecture-77

 Virtual Containers - I.
• Virtual containers (VC-x) encapsulate a PDH payload
with a special framing and a POH
• In SDH terminology, the original PDH payload with
special framing is called a container (C-x)
• Various container sizes with some space for stuffing
are defined
– C-11 for DS1 (25 bytes = 1.600 Mbit/s)
– C-12 for E1 (34 bytes = 2.176 Mbit/s)
– C-2 for DS2 (106 bytes = 6.784 Mbit/s)
– C-3 for DS3 or E3 (84 columns = 48.384 Mbit/s)
– C-4 for E4 (260 columns = 149.760 Mbit/s)

© 2001, Cisco Systems, Inc. Network Architecture-78

 Virtual Containers - II.
• Various VC sizes defined:
– With 1 byte allocated for POH
• VC-11 for DS1 (26 bytes = 1.664 Mbit/s)
• VC-12 for E1 (35 bytes = 2.240 Mbit/s)
• VC-2 for DS2 (107 bytes = 6.848 Mbit/s)
– With 1 column allocated for POH
• VC-3 for DS3 or E3 (85 columns = 48.960 Mbit/s)
• VC-4 for E4 (261 columns = 150.336 Mbit/s)

© 2001, Cisco Systems, Inc. Network Architecture-79

 Tributary Unit Structure
• TUs are defined to fit into a number of columns
– This requirement determines the size of virtual
containers and containers
– TU-3 adds up 3-byte pointer plus stuffing to VC-3
– Lower TUs add up 1 byte for pointer storage
• Organized into 4 frames (500 μs multi-frame)
• This provides V1, V2, V3, V4 TU pointer bytes
• Lower TUs also organize POH along the multi-frame
– This provides V5, J2, N2, K4 POH bytes

© 2001, Cisco Systems, Inc. Network Architecture-80

 Mapping Hierarchy - I.

STM-N

STM-1
STS-1
STS-1
STS-1
Frame
Frame
Frame
Frame

AU
AU SPE-Nc
AU
AU

DS1
DS1 E1
E1 TU
TUDS1C
DS1C DS2
DS2
DS1
DS1 E1
E1 TUDS1C
TU DS1C DS2
DS2

DS1 E1 DS2 DS3/E3 DS3/E3 E4 IP/ATM/Video

© 2001, Cisco Systems, Inc. Network Architecture-81

 Mapping Hierarchy - II.

STS-3N
STS-3N STS-3c
xN x1 STS-3cBULK
BULK
139 Mbit/s
STM-N AUG AU-4 VC-4 C-4
ATM
x3
x1
x3
TUG-3 TU-3 VC-3
x1
44 Mbit/s
STM-0 AUG AU-3 VC-3 C-3 34 Mbit/s
x7 DS3
DS3BULK
BULK
STS-1
STS-1 STS-1 x7
STS-1 SPE
SPE
x1
TU-2 VC-2 C-2 6.3 Mbit/s
TUG-2
x3
VT
VT
group TU-12 VC-12 C-12 2 Mbit/s
group
xN
Multiplexing x4
TU-11 VC-11 C-11 1.5 Mbit/s
Aligning VT-1.5
VT-1.5
Mapping
© 2001, Cisco Systems, Inc. Network Architecture-82
 Pointer Processing - I.
• Pointer processing compensates for phase differences and
small frequency variations (lasting for a number of frames)
• 10-bit pointer offset is stored in the H1, H2 overhead bytes
(normal range is 0-782)
– Specifies the byte position after the last H3 byte in the VC
capacity
– For AU-4 the byte position is 3 x (pointer offset) + 1
• TU pointer processing has a similar concept but different
implementation details
– TU-3 uses H1, H2, H3 bytes inside the TU payload capacity
– Lower TUs use V1, V2, V3, V4 bytes in 500 μs multi-frame

© 2001, Cisco Systems, Inc. Network Architecture-83

 Pointer Processing - II.
• AU-4 positive frequency justification
– If the tributary has a lower speed than its nominal
rate, then 3 stuffed bytes are inserted just after the
last H3 overhead byte
– Indicated by an inversion of the I-bits in the H1, H2
overhead bytes
• AU-4 negative frequency justification
– If the tributary has a higher speed than its nominal
rate, then H3 overhead bytes are used to carry extra
payload bytes
– Indicated by an inversion of the D-bits in the H1, H2
overhead bytes
© 2001, Cisco Systems, Inc. Network Architecture-84
 Low- and High-order Paths
• 2 mapping levels may be clearly identified in SDH:
– AU for high-order (HO) paths
• Direct termination of DS3, E3, E4, ATM
• May carry multiple LO paths by multiplexing
– TU for low-order (LO) paths
• Termination of DS1, DS2, DS3, E1, E3
• 2 special layered networks on top of STM-1 SDH
network
– HO and LO networks
– Managed by HO-POH and LO-POH information
• DS3/E3 may be part of both the HO or the LO network
© 2001, Cisco Systems, Inc. Network Architecture-85
 Summary
• Identify the need for special structures to support a mixture of
payloads

• Describe the grouping options of various payloads

• Describe the pointer processing associated with the individual

payloads

© 2001, Cisco Systems, Inc. Network Architecture-86

 Review Questions
• What are the two pointer options used by an
AUG?

• Does an E3 client signal belong to a High-order

Path or to a Low-order Path?

• What is the main difference between positive

and negative frequency justification?

© 2001, Cisco Systems, Inc. Network Architecture-87

 Section 2.6

Overhead Internals
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Describe the multiplexing process of overhead
bytes

• Describe the meaning of overhead bytes at

various hierarchy levels

© 2001, Cisco Systems, Inc. Network Architecture-89

 STM-1 Overheads
• STM-1 overheads were designed to be
compatible with SONET overheads
– Thus STM-1 overhead looks very similar to
STS-3 overhead
• Easiest to understand by drawing STM-0
section overhead first
• Then creating STM-1 section overhead by
multiplexing STM-0 3 times, and leaving out
unnecessary bytes
• POH is not changed by multiplexing

© 2001, Cisco Systems, Inc. Network Architecture-90

 STM-0 Overheads
HO Path
Section Overhead Overhead
Framing Framing RS Trace Path Trace
A1 A2 J0 J1
R-Section BIP-8 Orderwire User Channel BIP-8
Overhead B1 E1 F1 B3
Data Com Data Com Data Com Signal Label
D1 D2 D3 C2
Pointer Path Status
AU pointer Pointer Pointer
G1
H1 H2 H3
BIP-8 APS APS User Channel
B2 K1 K2 F2
Multiframe
Data Com Data Com Data Com Indicator
M-Section D4 D5 D6 H4
Overhead Data Com Data Com Data Com User Channel
D7 D8 D9 F3
Data Com Data Com Data Com APS
D10 D11 D12 K3

Sync (REI) Orderwire Tandem

S1 (M1) E2 N1

© 2001, Cisco Systems, Inc. Network Architecture-91

 STM-1 Section Overhead

A1 A1 A1 A2 A2 A2 J0 Δ - media
dependent
R-Section
B1 Δ Δ E1 Δ F1
Overhead
D1 Δ Δ D2 Δ D3

AU pointer H1 H1* H1* H2 H2* H2* H3 H3 H3 H1* = 10010011

B2 B2 B2 K1 K2 H2* = 11111111

D4 D5 D6
M-Section
Overhead D7 D8 D9

D10 D11 D12

national use

S1 M1 E2

© 2001, Cisco Systems, Inc. Network Architecture-92

 STM-N Section Overhead
• Created by multiplexing STM-1 section
overheads
• Only a few bytes are extended into each
position
– A1/A2, B2, national use
• Most bytes are not multiplied by multiplexing,
only the first appearance is used in the higher
level (bigger) frames

© 2001, Cisco Systems, Inc. Network Architecture-93

 Regenerator Section Overhead
Bytes
• Framing bytes (A1, A2)
– Identify the start of each STM-0 frame
• R-Section Trace (J0)
– Used to trace the origin of the STM-1 frame
• BIP-8 = Bit Interleaved Parity (B1)
– Checks even-parity on previous STM-N
frame after scrambling
• Orderwire (E1)
– 64 Kbit/s voice path used for communication
• User (F1) R-Section OH

– Optional, vendor specific A1 A2 J0

B1 E1 F1
• DCC = Data Communications Channel (D1-D3) D1 D2 D3

© 2001, Cisco Systems, Inc. Network Architecture-94

 Data Communication Channel
(DCC)
• Formed by the D1, D2 and D3 bites in the Section Overhead
• Creates a 192 Kbit/s link
• Used for monitoring, alarms, provisioning, and software download
• Protocol is point-to-point between ADMs
• Today protocol is proprietary
• Moving to IS-IS, CLNS and ES-IS
R-Section 1 2 3 4 5 6 7 8 9 10 268 269 270

271 273
D1 D2 D3
STM-1
M-SectionH1 c c H2 c c
K1 K2 SDH
Frame
SPE

2430

Section
Overhead (SOH) Path
Overhead (POH)

© 2001, Cisco Systems, Inc. Network Architecture-95

 AU Pointers
• Pointer (H1, H2)
– Two bytes used to indicate the offset between the
pointer bytes and the first byte of the SPE
– Also indicates concatenation
• Pointer Action (H3)
– Used to compensate for the SPE timing variations
• Positive/negative stuff bytes
– Pointer bytes tell when H3 is being used
M-Section OH

H1 H2 H3

B2 K1 K2

D4 D5 D6

D7 D8 D9

D10 D11 D12

S1 M1 E2

© 2001, Cisco Systems, Inc. Network Architecture-96

 Pointer Bytes (H1, H2) for AU-
3 Based Frames
• STM-1 pointer bytes usage:
– 3 x AU-3 bit streams should be located

10
1 2
271 273
3 4 5 6 7 8 9 268 269 270
STM-1

H1 H1 H1 H2 H2 H2

2430

Section
Overhead (SOH) 3 x AU-3's
1 2 3 87 89 90

H1 H2

SPE

810
Path
Overhead (POH)
© 2001, Cisco Systems, Inc. Network Architecture-97
 Multiplex Section Overhead
Bytes - I.
• BIP-24 (B2) interleaved parity
– Used for STM-N multiplex section error monitoring
• Parity check on MSOH and previous STM-N frame
before scrambling
• Provided for each STM-1 inside STM-N
• APS = Automatic Protection Switching (K1, K2)
– APS commands and error conditions between line
termination equipment M-Section OH

H1 H2 H3

B2 K1 K2

D4 D5 D6

D7 D8 D9

D10 D11 D12

S1 M1 E2

© 2001, Cisco Systems, Inc. Network Architecture-98

 Multiplex Section Overhead
Bytes - II.
• DCC (D4-D12)
– Uses same protocols and procedures as the RS-DCC
– For OAM&P messages between OSS and SDH multiplex-
section-level equipment
• Synchronization Status (S1)
– Allows the SDH equipment to choose the best clocking
source from many candidates
• STM-N REI-L (M1)
– Multiplex Section Level Remote Error Indicator M-Section OH

H1 H2 H3

• Orderwire (E2) B2 K1 K2

– 64 Kbit/s voice channel D4 D5 D6

D7 D8 D9
• Defined only in the first STS-N signal D10 D11 D12

S1 M1 E2

© 2001, Cisco Systems, Inc. Network Architecture-99

 Path Overhead Bytes - I.
• STS Path Trace (J1)
– Fixed length Access Point ID enabling the path
terminator to verify connection
Path OH
• 15-byte E.164 address plus 1 byte CRC-7 Path Trace
J1

• A 64-byte version permitted for SONET BIP-8

B3

compatibility Signal Label

C2
Path Status
• Path BIP-8 (B3) G1
User Channel
F2
– Parity check of previous VC Indicator
H4
before scrambling User Channel
F3
APS
K3
Tandem
N1

© 2001, Cisco Systems, Inc. Network Architecture-100

 Path Overhead Bytes - II.
Path OH
• Path Signal Label (C2) Path Trace
J1
– VC content type (mapping) BIP-8
B3

• Path Status (G1) Signal Label

C2
Path Status
– Allows the entire path to be monitored end G1
User Channel
to end F2
Indicator
– Used to notify the originating end of the path H4
User Channel
F3
• Performance and status of the entire APS

duplex path K3
Tandem
N1
• Carries the Remote Error Indicator (REI)
and the path Remote Defect Indicator (RDI)

© 2001, Cisco Systems, Inc. Network Architecture-101

 Path Overhead Bytes - III.
• Path User Channel (F2)
– Used by the network provider for internal
network communications
Path OH
• Position and Sequence Indicator (H4) Path Trace
J1

– Used when the frame is organized into BIP-8

B3

various mappings like Virtual Tributaries Signal Label

C2
or ATM cells Path Status
G1

• User Channel (F3) User Channel

F2
Indicator

• APS (K3) H4
User Channel
F3
• Network Operator Byte (N1) APS
K3
Tandem
N1

© 2001, Cisco Systems, Inc. Network Architecture-102

 Summary
• Describe the multiplexing process of
overhead bytes

• Describe the meaning of overhead bytes at

various hierarchy levels

© 2001, Cisco Systems, Inc. Network Architecture-103

 Review Questions
• What are the H1 and H2 bytes in the Multplex Section
overhead used for?

• What are the A1 and A2 bytes in the Regenerator

Section overhead used for?

• Why is the parity computed before scrambling?

• Is there a difference between SDH frames in various

countries?

© 2001, Cisco Systems, Inc. Network Architecture-104

 Summary
Information Resources
 Questions

© 2001, Cisco Systems, Inc.

? Network Architecture-106
 Information Resources
• Books
– Stamatios V. Kartalopoulos: “Understanding
SONET/SDH and ATM”
• IEEE, 1999; ISBN 0780347455

© 2001, Cisco Systems, Inc. Network Architecture-107

 Summary

After completing this chapter, you should

be able to perform the following tasks:
• Identify the main frame concepts
• Describe the basic structure of frames at
various hierarchy levels
• Make the basic computation for bit rates at
various hierarchy levels
• Describe the internal details of payloads and
overheads

© 2001, Cisco Systems, Inc. Network Architecture-108

 Chapter 3

Topology and
Protection
 Objectives

Upon completion of this chapter, you will

be able to perform the following tasks:
• Identify the main issues in topology design
• Define the main topologies
• Identify the main protection switching
concepts
• Describe the operations of typical topology
configurations

© 2001, Cisco Systems, Inc. Network Architecture-110

 Agenda

3.1 - Topology Basics

3.2 - Protection Switching
3.3 - USHR Topology
3.4 - BSHR Topology
Summary, Information Resources

© 2001, Cisco Systems, Inc. Network Architecture-111

 Section 3.1

Topology Basics
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify main topology alternatives

• Describe routing and provisioning concepts

© 2001, Cisco Systems, Inc. Network Architecture-113

 Topology Alternatives
• Point-to-point
– Used for SDH island trunks in old asynchronous
networks, or data services as POS or ATM links
• Linear point-to-multipoint
– Adds up ADM in the middle
– Max. 16 nodes
• Hub network
– A DCS interconnects ADMs
• Ring
– ADMs are put into a ring
– Redundant, multiple connected rings USHR
• Automatic protection switching (APS)
© 2001, Cisco Systems, Inc. Network Architecture-114
 Uni- and Bi-directional
Routing
A A
A-C A-C

F B F B
C-A
C-A

E C E C

D D

Uni-directional Ring Bi-directional Ring

(1 fiber) (2 fibers)
• Only working traffic is shown
• Subnetwork (path) or multiplex section switching for
protection
© 2001, Cisco Systems, Inc. Network Architecture-115
 Add-drop Provisioning
• Transport connections over a SDH
infrastructure are created by add-drop
provisioning
– A path is built up by specifying hop-by-hop
which channels should be added to a ring
and which channels should be dropped from
the ring
• Add-drop provisioning is typically done by the
network management system
– There is no signaling protocol

© 2001, Cisco Systems, Inc. Network Architecture-116

 Add and Drop Example
• STM-4 Ring
Add 1-3
• 4 x STM-1 channels
• Uni-directional routing ADM
1
• Provisioning:
Add 4-2
– add 1-3 (drop 3-1)
ADM OC-12 ADM
– add 3-4 (drop 4-3) Drop 2 4
Drop
– add 4-2 (drop 2-4)
ADM
• 2 channels occupied 3

Add 3-4
Drop

© 2001, Cisco Systems, Inc. Network Architecture-117

 Drop and Continue Example
• STM-4 Ring
Add 1-2,3
• 4 x STM-1 channels Drop
• Uni-directional routing ADM
1
• Provisioning: Drop & Drop &
Continue Continue
– add 1-2,3
ADM OC-12 ADM
– add 2-4,1 Add 2-4,1 2 4

• 2 channels occupied
ADM
3

Drop

© 2001, Cisco Systems, Inc. Network Architecture-118

 Uni- and Bi-directional
Example
Provisioning:
• add 1-3
• add 3-1
Uni-directional routing Bi-directional routing

ADM ADM
1 1

ADM ADM ADM ADM

2 4 2 4

ADM ADM
3 3

© 2001, Cisco Systems, Inc. Network Architecture-119

 Summary

• Identify main topology alternatives

• Describe routing and provisioning

concepts

© 2001, Cisco Systems, Inc. Network Architecture-120

 Review Questions
• Which types of topologies does the SDH
network support?

• Is there a signaling protocol that allows the

addressing of different SDH nodes within a
network?

© 2001, Cisco Systems, Inc. Network Architecture-121

 Section 3.2

Protection Switching
 Objectives
Upon completion of this section, you will
be able to perform the following tasks:
• Identify the main alternatives in creating
protection switching solutions

• Describe the main features of various

configuration options

• Describe the operational steps in the activation

of protection switching

© 2001, Cisco Systems, Inc. Network Architecture-123

 Multiplex Section Protection
Switching
R-Section
Overhead
information
controlling
protection Payload
switching
M-Section
Overhead

• Conditions resulting in a protection switch:

– Loss of signal, loss of frame
LOS AIS
– Line AIS (all 1’s) upstream
down
stream
– Signal degrade REI OCN

• Excessive BIP-24 errors in MS overhead

© 2001, Cisco Systems, Inc. Network Architecture-124
 Path Protection Switching

R-Section Payload
Overhead
VC
Path
Overhead
STM Info
Path controlling
Overhead protection
M-Section switching
Overhead VC
Payload

• Conditions resulting in a protection switch:

– Loss of pointer, STM or VC AIS
– Excessive BIP errors for STM path, BIP errors for VC
path
© 2001, Cisco Systems, Inc. Network Architecture-125
 Automatic Protection
Switching - I.
• APS = Automatic Tributary
Channels

Protection Switching STM-N Mux

– Allows network to MSTE K1K2

Read/Sel
K1K2
Write

react to failed lines,

interfaces, or
poor signal quality
• Performed over the Working Protect
STM-N STM-N
entire STM-N payload
• Uses K1 and K2 bytes of
MS Overhead
MSTE K1K2
Write
K1K2
Read/Sel

STM -N Mux

Tributary
Channels

© 2001, Cisco Systems, Inc. Network Architecture-126

 Automatic Protection
Switching - II.
• K1 byte: Tributary
Channels

– Type of request (bits 1-4) STM-N Mux

– Channel requested (bits MSTE K1K2

Read/Sel
K1K2
Write

5-8)
• K2 byte:
– Channel selected (bits 1-
4)
Working Protect
– Architecture (bit 5) STM-N STM-N

– Mode of operation (bits

6-8)
• e.g. Alarm Indication
Signal (AIS), Remote MSTE K1K2
Write
K1K2
Read/Sel

Defect Indicator (RDI)

STM -N Mux

Tributary
Channels

© 2001, Cisco Systems, Inc. Network Architecture-127

 Uni- and Bi-directional APS
• Uni-directional APS
– Only traffic on the affected fiber is switched
to the protect line
• Bi-directional APS
– TX and RX are both switched when channel
is affected

© 2001, Cisco Systems, Inc. Network Architecture-128

 Revertive and Non-revertive
APS
• Revertive switching
– Will restore to the working channel when
WTR timer expires
• Non-revertive switching
– Will not move to working channel after
failure unless requested

© 2001, Cisco Systems, Inc. Network Architecture-129

 1+1 Protection
• Bi- or unidirectional
• Non-revertive
• Transmits traffic on both channels

Working facility

Protection facility

ADM/Router ADM/Router
© 2001, Cisco Systems, Inc. Network Architecture-130
 1:n Protection - I.
• 1:1 protection (special case of 1:n)
– Bi- or unidirectional
– Revertive
– Typically dedicated protection
– May transmit traffic on both channels, or use protect for
low priority traffic

Working facility

Protection facility

ADM/Router ADM/Router
© 2001, Cisco Systems, Inc. Network Architecture-131
 1:n Protection - II.
• 1:n protection
– Bi- or unidirectional
– Revertive
– Shared protection facility

Working facility

Protection facility
ADM/Router ADM/Router

© 2001, Cisco Systems, Inc. Network Architecture-132

 APS Operations
• 1:n Bi-directional switching:
– Switch when transmitted bits K1:5-8 equals
received bits K2:1-4
• 1:n Unidirectional switching
– Same as 1:n
• 1+1 Switching
– Bi-directional - as above
– Unidirectional - each end operates independently
• Switch occurs immediately without capability to
reset

© 2001, Cisco Systems, Inc. Network Architecture-133

 Summary
• Identify the main alternatives in creating protection switching
solutions

• Describe the main features of various configuration options

• Describe the operational steps in the activation of protection

switching

© 2001, Cisco Systems, Inc. Network Architecture-134

 Review Questions
• What is APS used for?
• What is the difference between a two-fiber uni-
and bi-directional link?
• What is the difference between revertive and
non-revertive protection switching?
• Which two bytes of the Multiplex Section
overhead are used to initiate an APS?

© 2001, Cisco Systems, Inc. Network Architecture-135

 Section 3.3

USHR Topology
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify the main concepts in constructing
protected ring topologies
• Describe the main features and application
areas
• Describe the typical operational scenarios
• Describe the standardization efforts for
interoperability

© 2001, Cisco Systems, Inc. Network Architecture-137

 USHR Concepts
• USHR/P = Unidirectional Self-Healing Ring / Path Switched
• 2-fiber ring topology
– Head-end bridge, tail-end switch logical topology
• 1+1 protection with uni-directional routing on each fiber
• Traffic is sent in both directions on the ring on separate fibers
• The better signal is selected
by the receiver

© 2001, Cisco Systems, Inc. Network Architecture-138

 Application Areas
• Used in the access network or MAN
– All traffic homing into a central node
• e.g. CO
• Typical for STM-1, STM-4 rings

© 2001, Cisco Systems, Inc. Network Architecture-139

 Features
• Simplicity at the expense of capacity
– Bandwidth used, cannot be reused
• VC-1/2 and/or STM visibility
• Quick local fail-over independent from the rest
of the network
– No signaling protocol needed

© 2001, Cisco Systems, Inc. Network Architecture-140

 Operations – Traffic Flow
• One direction of duplex working
traffic between any two traffic
nodes goes through
each ring link for both A
working and protection
traffic F B
• Reverse direction of
protection
transmission is traffic
dedicated to protection
E C
• Therefore, the maximum
capacity of this ring
D
equals the line rate, i.e.
STM-4, STM-16 etc.

© 2001, Cisco Systems, Inc. Network Architecture-141

 Operations – Fiber Cut - I.
• Protection dedicated -
head end bridge
• Failure interrupts A-C A
working traffic
• Receiver at C detects
F B
failure
Protection
Traffic
Working
Traffic
E C

Working
D traffic
selected
© 2001, Cisco Systems, Inc. Network Architecture-142
 Operations – Fiber Cut - II.
• Fiber cut recovery steps:
– Tail end (receiver)
switches to protection A
traffic
– Only the
F B
receiving node
knows about Protection
the protection Traffic
switch Working
Traffic
• No traffic lost E C

Protection
D traffic
selected
© 2001, Cisco Systems, Inc. Network Architecture-143
 Operations – Node Failure - I.
• Protection bandwidth
dedicated - head end
bridge
A Node Failure
• Failure interrupts
A-C working traffic
F B
• Receiver at C
detects failure Protection
Traffic
Working
Traffic
E C

Working
D traffic
selected
© 2001, Cisco Systems, Inc. Network Architecture-144
 Operations – Node Failure - II.
• Node recovery steps:
– Tail end (receiver)
switches to protection A
traffic Node Failure

– Only receiving
F B
node knows
about the Protection
protection Traffic
switch Working
Traffic
• Traffic to/from E C
failed node is lost
Protection
D traffic
selected
© 2001, Cisco Systems, Inc. Network Architecture-145
 Standardization
• Basic APS operations are defined in ITU-T
G.783
• USHR/P is originally not fully defined by ITU-T
• Later defined in ITU-T G.841 as general VC trail
protection switching independent of the
underlying topology
– USHR/P is called 1+1 unidirectional VC trail
switching (ring topology is only a special
case) with dedicated protection
• USHR/MS and other variants are more a
theoretical possibility than real products
© 2001, Cisco Systems, Inc. Network Architecture-146
 Summary
• Identify the main concepts in constructing
protected ring topologies
• Describe the main features and application
areas
• Describe the typical operational scenarios
• Describe the standardization efforts for
interoperability

© 2001, Cisco Systems, Inc. Network Architecture-147

 Review Questions
• Why is the USHR/P based APS very fast?

• Why is it natural to use only uni-directional APS

in a USHR/P configuration?

• Why USHR/MS is not a practical idea?

© 2001, Cisco Systems, Inc. Network Architecture-148

 Section 3.4

BSHR Topology
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify the main concepts in constructing
protected ring topologies
• Describe the main features and application
areas
• Describe the typical operational scenarios
• Describe the standardization efforts for
interoperability

© 2001, Cisco Systems, Inc. Network Architecture-150

 BSHR Concepts - I.
• BSHR/MS = Bi-directional Self-Healing Ring / Multiplex Section Switched
• 1:1, or 1:N redundancy options
• 2 fibers with shared protection configuration
– Half the bandwidth in each direction in a link is reserved for the shared protection of all traffic
in that reverse direction of the link
• An even number of
STM-1s are required
• 4 fibers for dedicated protection configuration
– Bi-directional routing on 2 fibers (working line)
– Each direction has a working and a protect fiber

© 2001, Cisco Systems, Inc. Network Architecture-151

 BSHR Concepts - II.
• Multiple fail-over options for 4-fiber BSHR/MS
– In normal operation traffic is sent only in the required direction
– During fiber interruption, the traffic is routed around the break in
opposite direction (long path)
• Ring switching
– Optionally if the other 2 fibers are still available, then traffic
might be routed onto the parallel 2 fibers (short path)
• Span switching

© 2001, Cisco Systems, Inc. Network Architecture-152

 Traffic to/from an NE
dedicated protection
STM-1 #1-12 all paths STM-1 #1-12 all paths
working traffic working traffic
Network
STM-1 #1-12 all paths Element STM-1 #1-12 all paths
dedicated protection dedicated protection
traffic traffic

shared protection
STM-1 #1-6 working traffic STM-1 #1-6 working traffic
STM-1 #7-12 shared STM-1 #7-12 shared
protection traffic Network protection traffic
STM-1 #1-6 working traffic Element STM-1 #1-6 working traffic
STM-1 #7-12 shared STM-1 #7-12 shared
protection traffic protection traffic

Both rings have an add/drop capability of up to 12 STM-1s at any node

© 2001, Cisco Systems, Inc. Network Architecture-153

 Application Areas
• Used in the WAN backbone
– Neighboring traffic pattern is the best fit
• BSHR/MS rings are interconnected in a
hierarchy
– Number of hops should be minimized
– Capacity of rings increasing as moving up in
the ring hierarchy
• Rings might be classified into aggregation and
core

© 2001, Cisco Systems, Inc. Network Architecture-154

 Features
• More complexity, but more flexible capacity
– Bandwidth used can be reused
– Requires signaling between ADMs
• STM visibility
• Might restore service in less than 50
milliseconds on a 1200 km or less ring

© 2001, Cisco Systems, Inc. Network Architecture-155

 Operations – Traffic Flow
• Duplex traffic between
two nodes goes through
a subset of ring links
• Minimum capacity
equals line rate (same as
A
USHR/P maximum)
• Line rate must be an F B
even integer of STM-1 for
2-fiber configurations
– Automatically fulfilled
with newer standards E C
only working
D traffic shown

© 2001, Cisco Systems, Inc. Network Architecture-156

 Maximum Bandwidth Capacity
• Each link represents half
of the line rate of STM-1s
(i.e. 8 STM-1s for an A-B
F-A A
STM-16)
• All traffic from a node A-F B-A
goes to adjacent F B
nodes Only
• Max. capacity = working
E-F F-E traffic C-B B-C
0.5 (line rate) x shown
number of nodes
E C
E-D D-C

D C-D
D-E
© 2001, Cisco Systems, Inc. Network Architecture-157
 Extra Traffic
• Extra traffic utilizes
shared protection
bandwidth Working
A Traffic
• Extra traffic is not
protected when a
failure occurs F B
• Extra traffic could
be lost when a Extra Traffic
in Protection
failure of working bandwidth
traffic occurs
• Extra traffic is ONLY E C
available on a
BSHR/MS D
© 2001, Cisco Systems, Inc. Network Architecture-158
 Operations – Fiber Cut - I.
• Failure interrupts A-C
and C-A traffic Fiber cut
• A and B detect failure
A
STM-1#4

STM-1#4
F B

Working
Traffic

E C

D
© 2001, Cisco Systems, Inc. Network Architecture-159
 Operations – Fiber Cut - II.
STM-1#10 into STM-1#4
• No dedicated protection Fiber cut
bandwidth - only Loops
used when protection A
required
STM-1#4
STM-1#4 into into
• Only nodes next F STM-1#10
B STM-1#10

to the failure know STM-1#10 into

STM-1#4

about the Working

Traffic
protection switch
• No traffic lost
E C

D
Protection
Traffic
© 2001, Cisco Systems, Inc. Network Architecture-160
 Operations – Node Failure - I.
• Failure interrupts D-F
and F-D traffic
• A and C detect failure A STM-1#4

Node Failure
STM-1#4

F B

Working
Traffic

E C

© 2001, Cisco Systems, Inc. Network Architecture-161

 Operations – Node Failure - II.
• No dedicated STM-1#10 into STM-1#4

protection Loops
bandwidth - only A
Node
used when Failure
protection STM-1#4 into
F STM-1#10 B
required
• Only nodes next Working
Traffic Loops
to the failure
know about the STM-1#10 into
STM-1#4 STM-1#4
into
protection switch E C STM-1#10

• Traffic to/from
failed node lost
Protection D
Traffic
© 2001, Cisco Systems, Inc. Network Architecture-162
 Squelching Problem
• Traffic terminating on
nodes cut off by failures
STM-1#1
could be misconnected
A
to other nodes on the Node
ring in case of STM-1#1
Failure
using a local F B
fail-over decision
Working
Traffic

STM-1#1 STM-1#1
E C

© 2001, Cisco Systems, Inc. Network Architecture-163

 Squelching Misconnections
• Node F now talking to
STM-1#1
Node E instead of Node
B A Node
Failure
• Misconnection
would occur F B

Working
Traffic

E C
Protection
Traffic
D
STM-1#7 STS-1#1
© 2001, Cisco Systems, Inc. Network Architecture-164
 Squelching - Path AIS
Insertion
• STM Path AIS STM-1#1 Path
is inserted AIS inserted
by Node A
instead of the A
looped STM-1#1
STM-1#7 Node
F B Failure
• No mis-
connections
Working
Traffic

STM-1#1
Protection E C Path AIS
Traffic inserted
by Node C
D
STM-1#7 STS-1#1
© 2001, Cisco Systems, Inc. Network Architecture-165
 Squelching - Summary
• Squelching is required to assure that misconnections are not made
– Only required for bidirectional line switched rings since it is the only ring to
provide a reuse capability of STM-1s around the ring
– Only required when nodes are cut off from the ring
– Only required for traffic terminating on the cut off nodes
• A ring map that includes all STM and VC Paths on the ring is available at every
node on the ring
• Squelching is also required for extra traffic since the extra traffic may be dropped
when a protection switch is required

© 2001, Cisco Systems, Inc. Network Architecture-166

 Standardization
• Basic APS operations are defined in ITU-T G.783
• BSHR/MS is first defined in ITU-T G.803 (1993), but exact details are referred to as for
further study
• Later ITU-T G.841 (1995) defines BSHR/MS, but only for shared protection
– Dedicated protection is referred to as for further study (1998)
• Conflict with common sense terminology
• 4-fiber BSHR is called shared (!) protection since extra traffic might use the
protection fibers
• BSHR/P and other variants are more a theoretical possibility than real products

© 2001, Cisco Systems, Inc. Network Architecture-167

 Summary
• Identify the main concepts in constructing protected ring topologies

• Describe the main features and application areas

• Describe the typical operational scenarios

• Describe the standardization efforts for interoperability

© 2001, Cisco Systems, Inc. Network Architecture-168

 Review Questions
• Why is signaling required to ensure proper fail-over
in a BSHR/MS protection switching?

• What is the difference between shared protection in

a 2-fiber and a 4-fiber ring?

• Is 1:n protection available in a 4-fiber ring?

• Why is there a potential misconnection problem at

node failures?

© 2001, Cisco Systems, Inc. Network Architecture-169

 Summary
Information Resources
 Remember...
• Unidirectional ring: all working traffic travels around the ring in the same
direction for both A to B and B to A traffic; i.e. clockwise or
counterclockwise
• Bidirectional ring: all working traffic between two nodes travels the two
directions on the same set of fiber links between the two nodes; i.e. A to B
clockwise and B to A counterclockwise
• MS switching: APS is based on received MS signal status and MS layer
performance parameters
• Path switching: APS is based on received path layer signal status and
path layer performance parameters

© 2001, Cisco Systems, Inc. Network Architecture-171

 Remember...
• Ring switching: alternative path used is in the other direction on the
ring
• Span switching: alternative path used is in parallel with the failed path
• Squelching: insertion of AIS for looped signals during protection
switching to avoid misconnections
• Extra traffic: the utilization of the the protection bandwidth in a MS
switched ring for traffic that can and may be disrupted when a
protection switch is established

© 2001, Cisco Systems, Inc. Network Architecture-172

 Remember...
• + BSHR/MS provides higher bandwidth capacity when
internodal traffic exists between ring nodes
• ø BSHR/MS provides the same bandwidth capacity as
USHR/P when all traffic homes on a single node, as in
some access configurations
• + BSHR/MS evolves to support new service types like ATM
easily, path switched rings require new paths to be defined
for new services that don’t fit into STSs or VTs
• - USHR/P is perceived to be simpler since it can be thought
of as diverse routing
• - BSHR/MS is perceived to be complex because of
protection loops and squelching
• ø Both BSHR/MS and USHR/P require synchronization
protection switching which in a line switched protection
scheme is not needed
© 2001, Cisco Systems, Inc. Network Architecture-173
 Questions

© 2001, Cisco Systems, Inc.

? Network Architecture-174
 Information Resources
• Articles
– Dave Johnson, et al.: “The Evolution of a
Reliable Transport Network”
• IEEE Communications Magazine, August
1999, pp.52-57.

© 2001, Cisco Systems, Inc. Network Architecture-175

 Summary

After completing this chapter, you should

be able to perform the following tasks:
• Identify the main issues in topology design
• Define the main topologies
• Identify the main protection switching
concepts
• Describe the operations of typical topology
configurations

© 2001, Cisco Systems, Inc. Network Architecture-176

 Chapter 4

Time Synchronization
 Objectives

Upon completion of this chapter, you will

be able to perform the following tasks:
• Identify the main requirements for time
synchronization
• Describe the main synchronization modes
• Describe the principles of network
synchronization
• Describe the concept and operation of time
synchronization protection

© 2001, Cisco Systems, Inc. Network Architecture-178

 Agenda

4.1 - Time Synchronization Basics

4.2 - Synchronization Networks
4.3 - Synchronization Protection
Summary, Information Resources

© 2001, Cisco Systems, Inc. Network Architecture-179

 Section 4.1
Time Synchronization
Basics
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Describe the historical evolution of time
synchronization

• List main synchronization requirements

• Describe the network element synchronization

modes

© 2001, Cisco Systems, Inc. Network Architecture-181

 History - I.
• Time synchronization might be needed for all
digital voice communication networks
– On point-to-point links transmit and receive
frequencies should be the same
– In synchronous multiplexing transmit and
receive frequencies should be synchronized
everywhere in the network, otherwise
information might be lost
– In asynchronous multiplexing the
multiplexers are independent from the timing
of the multiplexed signals (PDH concept)

© 2001, Cisco Systems, Inc. Network Architecture-182

 History - II.
• SDH is quite different from PDH, since it uses
synchronous multiplexing
• Beginning of 1990s, SDH is used mainly as
point-to-point island, no synchronization with
E1, direct replacement for asynchronous
transport
• Middle of 1990s, SDH becomes time
synchronized, using complex topologies,
making pointer adjustments

© 2001, Cisco Systems, Inc. Network Architecture-183

 Synchronization in Classical
Voice Networks - I.

Network Clock 0.000001 ppm

(Stratum 1)

CB M14 LT LT M14 M14 LT LT M14 CB

DS0
Switch

VF E1 E4 prop. E4 E1 E1 E4 prop. E4 E1

Transport Switching Transport

Network Network Network

• Synchronous DS0/E1 switching network (Stratum 1)

• Asynchronous transport network
© 2001, Cisco Systems, Inc. Network Architecture-184
 Synchronization in Classical
Voice Networks - II.

Network Clock 0.000001 ppm

(Stratum 1)

CB M14 LT LT M14 M14 LT LT M14 CB

DS0
Switch

E1 E4 prop. E4 E1 E1 E4 prop. E4 E1

f1 f2 f3 f4 f5 f6
20ppm 20ppm 20ppm 20ppm 20ppm 20ppm

• Asynchronous transport network uses pulse stuffing and is

transparent to E1 timing

© 2001, Cisco Systems, Inc. Network Architecture-185

 Synchronization Distribution

Network Clock 0.000001 ppm

(Stratum 1)

Dedicated Timing
E1

CB M14 LT LT M14 M14 LT LT M13 CB

DS0
Switch
E1 E4 prop. E4 E1 E1 E4 prop. E4 E1

20ppm 20ppm

• Timing distribution is done using embedded E1 facility

• Asynchronous transport network is transparent to E1 timing

© 2001, Cisco Systems, Inc. Network Architecture-186

 Initial SDH Deployments

Network Clock 0.000001 ppm

(Stratum 1)

SDH SDH
M14 M14 M14 LT LT M14 CB
CB NE NE

DS0
Switch
E1 E4 STM-16 E4 E1 E1 E4 prop. E4 E1
f1 f2 f3 f4 f5 f6
20ppm 20ppm 20ppm 20ppm 20ppm 20ppm

• SDH used in point-point configuration

• Direct replacement for async transport
• SDH terminals free-run at 20ppm. Not network synchronized. No pointer
adjustments so no issues with E1/E4 mapping jitter !
© 2001, Cisco Systems, Inc. Network Architecture-187
 Current SDH Deployments - I.

Network Clock
0.000001 ppm
(Stratum 1)

???

CB STM-1 STM-1 STM-1 CB

STM-1

DS0
Switch
E1 E4 E1 E1 E1
STM-4
STM-16

Questions:
• How do I time the SDH network ?
• Can I still just free run all my SDH NEs at 20ppm ?
• What is the impact of pointer adjustments ?
• How do I distribute timing to the CBs and DS0 switches ?
© 2001, Cisco Systems, Inc. Network Architecture-188
 Current SDH Deployments - II.

Network Clock
0.000001 ppm
(Stratum 1)

BITS
CB STM-1 STM-1 CB
STM-1 STM-1

DS0
Switch

E1 E1 E1
STM-4
STM-16

• All STM-N interfaces traceable to PRS to avoid excessive pointers

• Excessive pointers cause jitter/wander in embedded E1/E4 payloads
• Timing distributed to CB and DS0 switches directly via STM-N lines

© 2001, Cisco Systems, Inc. Network Architecture-189

 Synchronization
Requirements
• Frequency variation of bits transmitted should be
inside the limits determined by the next hop’s ability to
transmit these bits further
– Stuffing allows for some limited tolerance
• Frequencies should be synchronized all over the
network to guarantee a low level of BER
• Synchronization is done by recovering the embedded
clock signal from the input signal
• Synchronization source should have a very precise
clock (reference clock)
• Reference clock might be reached only by multiple
hops
– Number of hops should be minimized

© 2001, Cisco Systems, Inc. Network Architecture-190

 Synchronization and Timing
Loops
Network Clock
0.000001 ppm
(Stratum 1)

BITS
CB STM-1 STM-1 CB
STM-1 STM-1

DS0
Switch
E1 E1 E1
Fiber STM-16
STM-4
Cut

Timing Loop
• Timing loops can be caused by either careless planning or fault conditions
• Timing loops cause unpredictable sync performance
• Elimination of timing loops was the driver for sync status messaging (SSM)

© 2001, Cisco Systems, Inc. Network Architecture-191

 Synchronization Modes for
Network Elements
• Each network element has to be configured for
time synchronization
• Time reference distribution should minimize
delay
• Various timing alternatives:
– External
– Line
– Loop
– Through

© 2001, Cisco Systems, Inc. Network Architecture-192

 External Timing
• All signals transmitted from a node are synchronized
to an external source received by that node; i.e. BITS
timing source

BITS

Network Element

W E
E A
S S
T T

© 2001, Cisco Systems, Inc. Network Architecture-193

 Line Timing
• All transmitted signals from a node are synchronized
to one received signal

Network Element

W E
E A
S S
T T

© 2001, Cisco Systems, Inc. Network Architecture-194

 Loop Timing
• The transmit signal in a optical link, east or west, is
synchronized to the received signal from the same
optical link

Network Element

W E
E A
S S
T T

© 2001, Cisco Systems, Inc. Network Architecture-195

 Through Timing
• The transmit signal in one direction of transmission
around the ring is synchronized to the received signal
from that same direction of transmission

Network Element

W E
E A
S S
T T

© 2001, Cisco Systems, Inc. Network Architecture-196

 Summary
• Describe the historical evolution of time synchronization

• List main synchronization requirements

• Describe the network element synchronization modes

© 2001, Cisco Systems, Inc. Network Architecture-197

 Review Questions
• In general, why is synchronization needed?

• How is synchronization achieved in PDH

networks?

• How is synchronization achieved in SDH

networks?

© 2001, Cisco Systems, Inc. Network Architecture-198

 Section 4.2
Synchronization
Networks
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify the reference clock concept
• Describe clock distribution methods
• Describe the concept of using alternative clock
sources
• Provide the basic design considerations for
synchronization network design

© 2001, Cisco Systems, Inc. Network Architecture-200

 Reference Clocks - I.
• Precision of internal clock is classified into so called
“Stratum” levels
– Accuracy is defined as the ratio of bit slip happening
(causing a bit error)
• Stratum 1 => 1 x 10-11 (synchronization to atomic
clock)
• Stratum 2 => 1.6 x 10-9
• Stratum 3E => 1 x 10-6
• Stratum 3 => 4.6 x 10-6
• Stratum 4 => 32 x 10-6 (typical for IP routers)
• Accuracy level might decrease at each hop in clock
distribution
© 2001, Cisco Systems, Inc. Network Architecture-201
 Reference Clocks - II.
• Originally providing Stratum 1 clocks for each network
element was far from being economical, even providing
this service at multiple locations was too much
demanding
– So clock distribution methods were developed to
minimize the number of high accuracy clocks needed
in the network
• Global Positioning System (GPS) includes Stratum 1
atomic clocks on the satellites
• Recently cheap GPS receivers make it possible to have
a Stratum 1 time source at almost any place
– Less need for time synchronization network (might
even go away in the future…)

© 2001, Cisco Systems, Inc. Network Architecture-202

 Clock Distribution Methods - I.
• External clock input might be used in case when all
equipment is at the same location
– BITS = Building Integrated Timing Signal
• Uses an empty T1 or E1 framing to embed clock
signal
• Root of the clock distribution tree
• Might be provided as a dedicated bus reaching
into each rack in a CO environment
– BITS should be generated from a Stratum 1 clock
• Typically with a hot spare alternative source for
fail-over

© 2001, Cisco Systems, Inc. Network Architecture-203

 Clock Distribution Methods -
II.
• Network elements not close to a BITS source should
recover clock from the line
• Clock distribution should not have loops, so a tree
distribution topology should be configured
• Typical carrier network element has Stratum 3 accuracy
when running free
– By synchronization to the reference clock, this clock
is running at the same rate as the reference clock
(that is Stratum 1)
• Minimum requirement for any network element is 20
ppm (that is between Stratum 3 and Stratum 4)

© 2001, Cisco Systems, Inc. Network Architecture-204

 Alternative Clock Sources - I.
• If the trail to the reference clock source is lost, the
network element still continues normal operation
– However, alarm might be generated
• After some time the clock might drift away so much,
that bit errors would occur
• Some time is left for switching over to an alternative
clock source
– The network element gets into a holdover state
– Requirement is to have less than 255 errors in 24
hours

© 2001, Cisco Systems, Inc. Network Architecture-205

 Alternative Clock Sources - II.
• A hierarchy of potential clock sources should be
configured at each network element to achieve a high-
availability operation
– Typically a maximum 3 alternative time reference
sources might be configured
– Meaningful only if there are different paths to the
alternative time reference sources
• If only one natural path exists to a single time reference
source, then the path must be protected by automatic
protection switching
– Requires some extra signaling to do it properly
– Called SPS = Synchronization Protection Switching

© 2001, Cisco Systems, Inc. Network Architecture-206

 Summary
• Identify the reference clock concept

• Describe clock distribution methods

• Describe the concept of using alternative clock sources

• Provide the basic design considerations for synchronization network design

© 2001, Cisco Systems, Inc. Network Architecture-207

 Review Questions
• What is the maximum allowed shift in frequency
in a SDH network?

• What are the different methods of providing the

master clock?

• What is the alternative to embedded clock

distribution?

© 2001, Cisco Systems, Inc. Network Architecture-208

 Section 4.3
Synchronization
Protection
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify the basic concepts of synchronization
protection

• Describe the operational steps in

synchronization protection

© 2001, Cisco Systems, Inc. Network Architecture-210

 Synchronization Protection
Basics
• Normal synchronization
around a ring:
– Nodes B-F are line timed BITS
– Node A is timed to an
external reference A
• In case of failure a new
time source should be F B
selected in a reasonable
amount of time
• BER is increasing through E C
time if synchronization is
not restored
D

© 2001, Cisco Systems, Inc. Network Architecture-211

 SPS Timing Loops
• SPS = Synchronization
Protection Switching
• During a ring failure, BITS
simple reference
switching would result in
A
timing loops
F B

Fiber Cut

E C

D
Timing Loop

© 2001, Cisco Systems, Inc. Network Architecture-212

 Operations – Normal Flow
• Synchronization
messaging in normal
operation BITS
– S1 = Stratum 1
Traceable Synch Msg = S1
A
– DU = Don’t Use S1
S1 DU
F B S1
– HO = Holdover
DU
S1 DU
E DU C
DU
D S1
S1

© 2001, Cisco Systems, Inc. Network Architecture-213

 Operations – Fiber Cut - I.
• Node C goes into short
term holdover
BITS

S1
A
HO
S1 DU
F B S1

DU Fiber Cut
HO HO
E C Node C in
DU Holdover
DU
D
HO HO

© 2001, Cisco Systems, Inc. Network Architecture-214

 Operations – Fiber Cut - II.
• Node F switches to
timing from Node A
BITS

S1
A
DU
S1 DU
F B S1
S1
Fiber Cut
HO HO
E C Node C in
DU Holdover
DU
D
HO HO

© 2001, Cisco Systems, Inc. Network Architecture-215

 Operations – Fiber Cut - III.
• Ring is reconfigured and
all nodes are again
BITS
synchronized to BITS
S1
A
DU
S1 DU
F B S1
S1
Fiber Cut
S1
DU Node C
E S1 C
comes out
S1 of Holdover
D DU
DU

© 2001, Cisco Systems, Inc. Network Architecture-216

 Summary

• Identify the basic concepts of

synchronization protection

• Describe the operational steps in

synchronization protection

© 2001, Cisco Systems, Inc. Network Architecture-217

 Review Questions
• What kind of signaling is used to prevent timing
loops?

• Does SPS work properly in a complex meshed

network?

• How long is the Hold-over period in SPS?

© 2001, Cisco Systems, Inc. Network Architecture-218

 Summary
Information Resources

© 2001, Cisco Systems, Inc.

 Remember...
• External timing: all signals transmitted from a node
are synchronized to an external source received by
that node; i.e. BITS timing source
• Line timing: all signals transmitted from a node are
synchronized to one receive signal
• Loop timing: the transmit signal in a optical link, east
or west, is synchronized to the received signal from
the same optical link
• Through timing: the transmit signal in one direction of
transmission around the ring is synchronized to the
received signal from that same direction of
transmission

© 2001, Cisco Systems, Inc. Network Architecture-220

 Remember...
• All network elements should be able to trace back clock
synchronization to a single reference clock
• Synchronization messaging: messaging procedure that
avoids timing loops when protection switching the
synchronization source on a ring
• Messaging is based on bits in the S1 byte in the MS
Overhead

© 2001, Cisco Systems, Inc. Network Architecture-221

 Remember...
• Synchronization protection switching is controlled by
synchronization messages in bits 5-8 of the S1 byte in
the SDH MS overhead, and is considered a MS
switching function
• Synchronization protection switching is required for
both BSHR/MS rings and USHR/P rings
• Short term holdover is defined as holdover during
synchronization reconfiguration on the ring

© 2001, Cisco Systems, Inc. Network Architecture-222

 Questions

© 2001, Cisco Systems, Inc.

? Network Architecture-223
 Summary

After completing this chapter, you should

be able to perform the following tasks:
• Identify the main requirements for time
synchronization
• Describe the main synchronization modes
• Describe the principles of network
synchronization
• Describe the concept and operation of time
synchronization protection

© 2001, Cisco Systems, Inc. Network Architecture-224

SONET versus SDH
Chapter 5
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Relate SONET and SDH concepts to each other
• Translate between SONET and SDH
terminology
• Compare SONET and SDH terminology
• Describe the internetworking principles
between SONET and SDH

© 2001, Cisco Systems, Inc. Network Architecture-226

 Standardization - I.
• First: many proprietary solutions
• In 1984 ECSA (Exchange Carriers
Standards Association) started on SONET
• SONET became an ANSI standard
– Tuned to carry US PDH payloads
• Later CCITT created SDH as a superset
– Tuned to carry European and
international PDH payloads including E4
(140 Mbit/s)
© 2001, Cisco Systems, Inc. Network Architecture-227
 Standardization - II.
Divestiture CCITT Expresses
Interest in SONET
Exchange Carriers British and Japanese SONET/SDH
Standards Associate (ECSA) Participation in T1X1 Standards
T1 Committee Formed Approved
ANSI T1X1 CCITT XVIII CEPT Proposes
Bellcore Proposed Approves Begins Study
SONET Principles Merged ANSI/CCITT
Project Group Standard
To ANSI T1X1

1984 1985 1986 1987 1988

SONET Concept Developed by Bellcore

• More than 400 technical proposals US T1X1 Accepts

Modifications
• Rate discussions AT&T vs. Bellcore
• International changes for byte/bit ANSI Approves
interleaving, frames, data rates SYNTRAN

• Phase I, II, III, separate APS etc.

© 2001, Cisco Systems, Inc. Network Architecture-228
 Synchronous TDM Hierarchy
• SONET Synchronous Transport Signal
– STS-<n> electrical, OC-<n> optical
• SDH Synchronous Transport Module
– STM-<n>E electrical, STM-<n>O optical
SONET SDH Abbrev.
Bit Rate Signal Channels Signal Channels speed
(Mbit/s) DS1 DS3 E1 E4 (Gbit/s)
51.84 STS-1 28 1 STM-0 21 0
155.52 STS-3 84 3 STM-1 63 1
622.08 STS-12 336 12 STM-4 252 4
2488.32 STS-48 1344 48 STM-16 1008 16 2.5
9953.28 STS-192 5376 192 STM-64 4032 64 10
39813.1 STS-768 21504 768 STM-256 16128 256 40
© 2001, Cisco Systems, Inc. Network Architecture-229
 Comparison of Technology
• Both SONET and SDH use the same
technology components
• All the differences might be implemented in
software
• Still US based companies tend to be late in
SDH implementations
– Mostly because of other customer
environment related standards
• Privately SONET or SDH might be used
• Interfacing to public networks requires
configuring the proper selection
© 2001, Cisco Systems, Inc. Network Architecture-230
 Comparison of Hierarchy
• Hierarchies aligned at 155 Mbit/s

2 8 34 140 Mbit/s

ETSI PDH

1.5 6 45 Mbit/s

US PDH T3

VT1.5 VT6 OC-1 OC-3 OC-12 OC-48 OC-192

SONET
52 155 622 2488 9953
TU-12 TUG-2 AU-3 AU-4 Mbit/s

SDH STM-1 STM-4 STM-16 STM-64

© 2001, Cisco Systems, Inc. Network Architecture-231

 Comparison of Framing
• Same framing concept of using 9 rows
• STM-1 frame can be subdivided into 3 virtual
STM-0 frames
– STM-0 frame compatible with STS-1 frame
• Multiplexing of STM and STS is the same
• Overhead byte interpretation is slightly
different
– Based on different needs for PDH
multiplexing and protection

© 2001, Cisco Systems, Inc. Network Architecture-232

 Comparison of Payloads
• Similar concepts, but significant differences in
details of payload multiplexing
SONET OC-3
VT 1.5 VT 6 7 STS-1 STS-3
(1) 3

DS1
28 DS1s 84 DS1s
(4)
1.5 Mb/s

SDH
VC-12 TUG 7
VC-3 3 AUG STM-1
(1)

E1
(3) 21 E1s 63 E1s
2 Mb/s
© 2001, Cisco Systems, Inc. Network Architecture-233
 Comparison of Network
Architectures
• Similar layered architecture
• Slightly different terminology
– Regenerator section (SDH) = section
(SONET)
– Multiplex section (SDH) = line (SONET)
• SDH defines high- and low-order paths, too

© 2001, Cisco Systems, Inc. Network Architecture-234

 Comparison of Protection
• SONET APS schemes are almost the same as
SDH MPS and SNCP schemes
• Operations are practically the same
– APS protocol is equivalent
• Terminology is different
– UPSR (SONET) = USHR/P (SDH)
– BLSR (SONET) = BSHR/L (SDH)

© 2001, Cisco Systems, Inc. Network Architecture-235

 Internetworking
• Voice internetworking still requires conversion
between μ-law and a-law
• Voice trunks can be accessed with a single step of
demultiplexing
• SONET and SDH might carry each others PDH load,
so the voice conversion point might be located
flexibly anywhere inside the SONET or the SDH
network
• Repacking PDH between SONET and SDH might be
done in a single step by a single device
• Data internetworking is easy at STM-1 (STS-3) or
higher since the payloads are the same size and
structure

© 2001, Cisco Systems, Inc. Network Architecture-236

 Summary

• Relate SONET and SDH concepts to

each other
• Translate between SONET and SDH
terminology
• Compare SONET and SDH terminology
• Describe the internetworking principles
between SONET and SDH

© 2001, Cisco Systems, Inc. Network Architecture-237

 Review Questions
• What is the major difference between SONET
and SDH?
• Why does SONET start at a signaling rate of
51,84 Mbit/sec?
• Why does SDH start at a signaling rate of 155,52
Mbit/sec?
• What is the purpose of an STM-0 frame?

© 2001, Cisco Systems, Inc. Network Architecture-238

 Questions

© 2001, Cisco Systems, Inc.

? Network Architecture-239
 Chapter 6

Network Management
 Objectives

Upon completion of this chapter, you will

be able to perform the following tasks:
• Describe the basic network management
functions needed in TDM networks
• Describe the architecture and main
components of implementing network
management solutions

© 2001, Cisco Systems, Inc. Network Architecture-242

 Agenda

5.1 - Network Management Basics

5.2 - Network Management Internals
Summary, Information Resources

© 2001, Cisco Systems, Inc. Network Architecture-243

 Section 5.1
Network Management
Basics
 Objectives

Upon completion of this section, you will

be able to perform the following tasks:
• Identify the main operational tasks

• Describe OAM functions and layers

• Describe the in-band network management

channels

© 2001, Cisco Systems, Inc. Network Architecture-245

 Operational Tasks - I.
• Basic operational tasks:
– Protection
• Circuit recovery in milliseconds (so failure should
not be detected by voice customers)
– Restoration
• Circuit recovery in seconds or minutes (done by
manual configuration)
– Provisioning
• Allocation of capacity to preferred routes
(according to certain time schedules)
• Configuration time is separated from activation
time
© 2001, Cisco Systems, Inc. Network Architecture-246
 Operational Tasks - II.
– Consolidation
• Moving traffic from unfilled bearers onto fewer
bearers to reduce waste trunk capacity
– Grooming
• Sorting of different traffic types from mixed
payloads into separate destinations for each type
of traffic

© 2001, Cisco Systems, Inc. Network Architecture-247

 OAM Functions and Layers
• Level 3 - Path
– Assembly and disassembly, cell delineation
control
• Level 2 - Multiplex Section
– Loss of frame synchronization, degraded
error performance
• Level 1 - Regenerator Section
– Loss of synchronization, signal quality
degradation

© 2001, Cisco Systems, Inc. Network Architecture-248

 Data Communication Channel
(DCC)
• DCC is a 192 kb/s in-band channel to facilitate
communication between all Network Elements (NE) in a
network
– Remote login, alarms reporting, software download,
provisioning
Management
Client

OSS
Alarm and
Event
Forwarding DCN
TDM
GNE
GNE
SDH ADM
GNE
GNE GNE DCC
GNE

Management Management SDH

TDM
Clients Server SDH DCC
DCC
ADM
Network Operations Center TDM TDM

TDM
ADM ADM
© 2001, Cisco Systems, Inc. Network Architecture-249
 Summary

• Identify the main operational tasks

• Describe OAM functions and layers

• Describe the in-band network

management channels

© 2001, Cisco Systems, Inc. Network Architecture-250

 Review Questions
• What are the operational tasks of a Network
Management System?

• What are the OAM functions needed for?

• What is the DCC channel used for?

© 2001, Cisco Systems, Inc. Network Architecture-251

 Section 5.2

Network Management
Internals
 Objectives
Upon completion of this section, you will
be able to perform the following tasks:
• Identify main requirements for network
element management support

• Describe the typical management interfaces

used to access network elements

• List management functions and features

typically supported by network elements

© 2001, Cisco Systems, Inc. Network Architecture-253

 Network Element
Requirements
• All network elements should support various
management interfaces
– Local (craft terminal using TL1)
– Remote (TMN DCN model)
• All network elements should support certain
management function areas
– FCAPS (fault, configuration, accounting,
performance, security)
• Since original payload is voice, a separate
management network is needed for remote
management operations
– Data Communications Network (DCN)
© 2001, Cisco Systems, Inc. Network Architecture-254
 Management Interfaces - I.

TMN Model as of M.3010

TMN OS
Reference point

Q3/X/F
X
DCN
F Q3 Q3
WS NE QA
© 2001, Cisco Systems, Inc. Network Architecture-255
 Management Interfaces - II.
• CMIP over OSI
– TMN Manager/Agent communication standard

Agent
Element Manager
Layer Mgr.

CMIP/OSI

Agent
Network Element
Layer

© 2001, Cisco Systems, Inc. Network Architecture-256

 Configuration Management
Support
• Installation
– Setting up basic parameters (identification,
management access and authorization, etc.)
– Activating and testing hardware
• Provisioning
– Implementing add-drop commands
• Status and control
– APS switching messages for manual control

© 2001, Cisco Systems, Inc. Network Architecture-257

 Performance Monitoring
Support
• Separate handling of section, line, path termination
• Performance data collection
– Error counts: B1, B2, B3
– Historical data presentations
• Threshold settings
• Threshold crossing alerts (TCA)
– For B1, B2, B3
• Performance data reporting
• Accuracy and resolution should be considered
– Typically 15-minute interval is the basis
• Monitoring should be done even in trouble conditions

© 2001, Cisco Systems, Inc. Network Architecture-258

 Fault Management Support
• Alarms surveillance
– Reactive mechanisms
Alarm Hierarchy
– Autonomous and
LOS
requested alarms |
• Testing LOF
|
• Alarm configuration LAIS => LRDI
| |
PAIS => PRDI
|
LOP

© 2001, Cisco Systems, Inc. Network Architecture-259

 Accounting Support
• In voice networks accounting support is implemented
in toll switches
• In data networks accounting is supported by data link
layer or network layer traffic data collection
• In general, neither voice networks, nor data networks
transport on top of SDH require accounting support
• Transport network billing is not based on traffic, since
bandwidth is allocated in fixed amounts
– Billing records might be generated by the OSS
controlling the provisioning of paths
• Typically no requirement for accounting support in
SDH network elements
© 2001, Cisco Systems, Inc. Network Architecture-260
 Summary
• Identify main requirements for network element management support

• Describe the typical management interfaces used to access network

elements

• List management functions and features typically supported by

network elements

© 2001, Cisco Systems, Inc. Network Architecture-261

 Review Questions
• What network management protocol is mainly
used in telecommunication?

• What three functions does Fault Management

support?

© 2001, Cisco Systems, Inc. Network Architecture-262

 Summary
Information Resources
 Questions

© 2001, Cisco Systems, Inc.

? Network Architecture-264
 Summary

After completing this chapter, you should

be able to perform the following tasks:
• Describe the basic network management
functions needed in TDM networks
• Describe the architecture and main
components of implementing network
management solutions

© 2001, Cisco Systems, Inc. Network Architecture-265

© 2001, Cisco Systems, Inc. Network Management-266