ATM

Loan Pham

12/08/21 1
 Fast Relay Networks
 Relaying traffic as quickly as possible
Fast Relay

Frame Relay Cell Relay

(Variable size (Fixed size
PDUs – frames) PDUs – cells)

PVC SVC ATM based 802.6 based

(LAPD) (Q.931) (For B-ISDN) (For SMDS)

PVC SVC
(Q.2931)

12/08/21 2
 Transfer Mode
 Transfer mode
 Specific way of transmitting and switching through the network
 Synchronous Mode
 Synchronous systems negotiate the connection at the data-link
level before communication begins. Basic synchronous systems will
synchronize two clocks before transmission.
 Asynchronous mode
 Asynchronous systems do not send separate information to indicate
the encoding or clocking information. The receiver must decide the
clocking of the signal on it's own. This means that the receiver
must decide where to look in the signal stream to find ones and
zeroes, and decide for itself where each individual bit stops and
starts.

12/08/21 3
 Switching technologies
 Circuit Switching
 circuit is established for the complete duration
 based on TDM (Time Division Multiplexing)
 also referred to as STM (Synchronous Transfer Mode)
 very inflexible, since once the duration of time slot is determined, the relative
 bit rate is fixed
 Packet Switching
 user information is encapsulated in packets
 packets contain additional information used inside the network for routing,
 error correction, flow control, etc
 packets have variable length and hence require complex buffer management
 inside of the network
 Frame relaying
 eliminates as much as possible of the overhead of packet switched networks
 no hop-by-hop flow control and error control
 can be viewed as streamlined version of X.25

12/08/21 4
 Switching technologies (cont.)
Circuit Message Packet Frame Relay Cell Relay
switching switching switching (switching) (switching)
Technique
Relay

Direct Store & Hold & Hold & Hold &

connection forward forward forward forward

Copper, Copper, Copper,

Media

Copper, Copper, wireless, wireless, wireless,

wireless wireless optical optical optical

Variable, Variable, Variable, Fixed, very

Size of
PDU

No such large to large to large to small

thing small small small
Delay

Very fast Slow Fast Faster Very fast

12/08/21 5
 ATM
 ATM Introduction
 What is an ATM (Asynchronous Transfer Mode)?
 ATM Overview
 ATM Topology
 Standard
 Basic principle
 ATM Network Interfaces
 ATM Cells
 ATM virtual connections
 ATM Reference Model
 Physical Layer
 ATM Layer
 ATM Adapted Layer
 ATM Services

12/08/21 6
 ATM
 ATM Introduction
 What is an ATM (Asynchronous Transfer Mode)?
 ATM Overview
 ATM Topology
 Standard
 Basic principle
 ATM Network Interfaces
 ATM Cells
 ATM virtual connections
 ATM Reference Model
 Physical Layer
 ATM Layer
 ATM Adapted Layer
 ATM Services

12/08/21 7
 ATM Introduction

 ATM (Asynchronous Transfer Mode) is is an

International Telecommunication Union-
Telecommunication Standardization Sector (ITU-T)
standard for cell relay wherein information for multiple
service types, such as voice, video, or data, is
conveyed in small, fixed-size cells
 ATM is a cell-based, connection-oriented, switching
and multiplexing technology designed to be a fast,
general purpose transfer mode for multiple services.

12/08/21 8
 ATM Overview
 Proposed telecommunication standard for Broadband ISDN
 Circuit-switched, connection-oriented networking technology
 Provides dedicated, high-speed connections to a virtually unlimited
number of users
 Dedicated media connections running in parallel allow ATM to support
simultaneously multiple transmissions
 Adding more users does not necessarily decrease the average
bandwidth available to connections on the network
 Integrates voice, video, and data
 Uses short, fixed length packets called cells
 Best effort delivery system
 Bandwidth on demand
 Potential to remove performance bottlenecks in today’s LANs and WANs

12/08/21 9
ATM – Advantages and Disadvantages
 Advantages of ATM
 Flexible bandwidth allocation
 Simple routing due to connection-oriented technology
 High bandwidth utilization due to statistical multiplexing
 Potential QoS guarantees
 Disadvantages of ATM
 Overhead of cell header (5 bytes per cell)
 Complex mechanisms for achieving QoS
 Congestion may cause cell losses

12/08/21 10
 ATM Network Topology
 Physically, an ATM network is composed of a
collection or a mesh of ATM switches and stations
which are interconnected via ATM media.
 ATM switches are simply devices which participate in
the creation and in the tear-down of ATM virtual
circuits, and route ATM cells according to the
appropriate virtual circuit.
 Because ATM cells are uniform in structure, switching
of cells can be done very rapidly and with extremely
low latency. Cell switching is also accomplished in
parallel over the various connections in a switch.

12/08/21 11
 ATM
 ATM Introduction
 What is an ATM (Asynchronous Transfer Mode)?
 ATM Overview
 ATM Topology
 Standard
 Basic principle
 ATM Network Interfaces
 ATM Cells
 ATM virtual connections
 ATM Reference Model
 Physical Layer
 ATM Layer
 ATM Adapted Layer
 ATM Services

12/08/21 12
 Basic Principle of ATMs
 Specific packet oriented transfer mode based on fixed cell length
 Each packet consist of an information field and a header (used to
determine the virtual channel and to perform appropriate routing)
 Cell sequence integrity is preserved per virtual channel
 Connection-oriented (header values are assigned to each section
of a connection for the complete duration of the connection)
 Signalling and user information are carried on separate virtual
channels
 Information field of ATM cells is carried transparently through the
network, no processing like error control is performed inside the
network
 Connectionless services are provided by adaptation functions to fit
information into ATM cells or to provide service specific functions

12/08/21 13
 Basic Principles of ATM (cont.)
 Connection-oriented
 connections are established for the duration of a call
 Establishment includes
 allocation of a Virtual Channel Identifier (VCI)
 allocation of a Virtual Path Identifier (VPI)
 allocation of the required recourses (usual in terms of throughput)
 Establishment uses
 negotiation between user and network, with respect to the
recourses
 separate signalling channel (for signalling messages)
 To establish and to release a signalling virtual channel a special
Metasignaling Channel is used (predefined VCI/VPI defined on
UNI)

12/08/21 14
 ATM Network Interfaces
 An ATM network consists of a set of ATM switches
interconnected by point-to-point ATM links or interfaces.
 ATM switches support two types of interfaces:
 UNI (User to Network Interface): ATM endpoint-ATM switch
 Public UNI, Private UNI
 NNI (Network to Node Interface): ATM switch-ATM switch
 Private NNI (P-NNI)
 Public NNI – Inter-Switching System interface (ISSI)
 The Broadband Interexchange Carrier Interconnect (B-
ICI)
 Connects two public switches from different service providers
 Data Exchange Interface (DXI)
 Between packet routers and ATM Digital Service Units (DSU)

12/08/21 15
 ATM Network Interfaces (cont.)
Regional Carriers

Private Private Public Public Public Public

Computer
Switch UNI Switch NNI Switch
UNI
Private
NNI B-ICI
Private Private
Computer Long Distance
Switch Public
UNI
Carrier Switch

B-ICI

DXI Digital Private Public

Computer Router Service
UNI Switch
Unit (DSU)
12/08/21 16
 ATM Network Interfaces (cont.)

12/08/21 17
 ATM Virtual Connections
 Virtual connection
 Limited functionality in ATM headers
 source address, destination address, sequence number are not required
 error control is only present if required by the service
 no flow control mechanism
 Remaining function is the identification of the virtual connection
(performed by 2 sub-fields):
 Virtual Channel Identifier (VCI)
 Virtual Path Identifier (VPI)
 Virtual channel
 Optical transmission links are capable of transporting hundreds of MBit/s
 Virtual channels may fill only KBit/s
 A large number of simultaneous channels have to be supported on an ATM
transmission link
 VCI is assigned during call set-up
 VCI is only significant on the link between ATM nodes
 When the connection is released, the VCI values on the involved links will
be released and can be used by other connections
 Enable multi-component services
12/08/21 18
 ATM Virtual Connections (cont.)
 Virtual Path
 Enable semi-permanent
connections between
endpoints
VC1
 These connections have to
VP1 VC2
transport a large number of
VC3
simultaneous connections
 Concept is also known as VC1
virtual network (resources of Link VP2 VC2
the network are semi- VC3
permanently allocated) VC1
 Allows efficient and simple VP3 VC2
management VC3

A VP is a bundle of VCs

12/08/21 19
 VPI/VCI used in an ATM network
VPI = 7
VCI = 1,2,3

ATM VPI = 5
VPI = 9
Node 1 VCI = 1,2,3
VCI = 3,4
ATM
Node 2
VPI = 7
VCI = 3,4

ATM
Node 3

VPI = 3
VCI = 3,4

12/08/21 20
 ATM Virtual Connections (cont.)
 VPI/VCI=0/0 used for Idle cells; 0/n used for Signalling
 Each cell contains a 24/28-bit connection identifier
 First 8/12 bits: Virtual Path, Last 16 bits: Virtual Channel
 The use of VPIs for static connections on the ATM
network
 Site-to-site connectivity
 Between service provider and customer
 VPIs have local significance
 VCIs are for dynamic connections on ATM network
 They are for the actual connections between applications on
ATM network

12/08/21 21
 ATM Virtual Connections (cont.)
 VP connections (VPCs) = Series of VP Links
 VC connections (VCCs) = Series of VC Links to make an end-to-
end link
 Call = Multiple connections

12/08/21 22
 ATM connection types

 Point-to point
 For direct connectivity, using for connections
between adjacent network nodes
 Point-to-multipoint
 For multicast or broadcast-type services
 Multipoint-to-multipoint
 For conference arrangements, ATM network node
is responsible for sending out multiple copies of a
single cell received on a single in port

12/08/21 23
 ATM
 ATM Introduction
 What is an ATM (Asynchronous Transfer Mode)?
 ATM Overview
 ATM Topology
 Standard
 Basic principle
 ATM Network Interfaces
 ATM Cells
 ATM virtual connections
 ATM Reference Model
 Physical Layer
 ATM Layer
 ATM Adapted Layer
 ATM Services

12/08/21 24
 ATM Reference Model

12/08/21 25
 ATM Reference Model (cont.)
 ATM reference model contains three Planes: User Plane, Control
Plane, and Management Plane as shown in Figure 1.
 User Plane – we have already discussed extensively for data
transmission
 Control Plane – Performs Set-up of the connection (SVC, PVC),
Manage the end-to-end connection, and release of the connection
 Management plane – deals with two items:
 Layer management which deals with managing the network resources
 Plane management which deals with co-ordination of other planes.

User Information User Plane User Information

AAL AAL

ATM ATM ATM ATM

PHY PHY PHY PHY

End system Network End system

12/08/21 26
 ATM Reference Model (cont.)

 Management Plane is divided into: Layer management and

Plane management.
 Layer management function is further split into Control
plane management and User plane management.
 Layers in the control plane management are the functions needed
to performs setting up the connection , monitoring and
disconnection. We have discussed this above.
 The above functions are only needed in the switched virtual
connection and is not required in permanent virtual connection.
 Layers in user plane comprise the functions required for the
transmission of user information.

12/08/21 27
 ATM Reference Model (cont.)
 Plane management functions supervise the operations of the whole network
and has no layered structure. This includes operation and maintenance (OAM)
 Performance monitoring
 Long term system evaluation
 Short term service quality control or preventive actions
 Defect and failure detection
 Enables failure detection → localization
 System protection
 Failed entity will be excluded from operation
 Minimizing the effect of the failure
 Failure or performance information
 Informs other management entities (system protection)
 Fault localization
 Internal or external test systems will localize the failed entity
 Fault-management OAM cells have the leading 4 bits of the cell payload set to 0001.
 The function type (FT) field, indicate the type of function performed by the cell:
alarm indication signal (AIS), signaled by FT = 0000; far end receive failure (FERF),
signaled by FT = 0001; and loopback cell, signaled by FT = 1000.

12/08/21 28
 E n d -to -e n d

S egm ent

T e rm in a l P r iv a te P u b lic P u b lic P r iv a te T e rm in a l
o r r o u te r ATM A T M sw itc h A T M sw itc h ATM o r r o u te r
s w itc h s w itc h

S a m e a s u s e r ,s c e l l s
V PC : F4 3: Segm ent
4 : E n d -to -e n d 0 0 0 1 = F a u lt m a n a g e m e n t

F u n c tio n -
G FC V PI VCI PT CLP HEC OAM FT C R C -1 0
s p e c if ic
c e ll ty p e
fie ld

100 = segm ent F u n c tio n ty p e : 0 0 0 0 = A IS

V CC : F5 1 0 1 = E n d -to -e n d 0001 = FE R F
S a m e a s u s e r ,s c e l l s 1000 = L oopback
12/08/21 29
 ATM Layered Model (cont.)
 ATM Adaptation layer (AAL)
 How to break application messages to cells
 ATM layer
 Transmission/Switching/Reception
 Congestion control/Buffer management
 Cell header generation/ Remove at Destination/Source
 Reset the connection identifiers for the next hop
 Cell address translation
 Sequential delivery
 Physical layer (PHY)
 Transmission the information through physical media

12/08/21 30
 ATM and OSI Model

12/08/21 31
 ATM Physical Layer
 ATM physical layer transmits the ATM cells as a
sequence of bits link by link through the ATM network
 Two sub-layers
 Physical Medium Sublayer
 Responsible for the correct transmission and reception of bits on
the appropriate physical medium
 Lowest level is media dependent (optical, electrical, ...)
 Upper level guarantees a proper bit timing reconstruction at the
receiver
 Transmitting peer entity inserts required bit timing information and
line coding

12/08/21 32
 ATM Physical Layer (cont.)
 Transport Convergence (TC)
 Cell rate decoupling – Insertion and extraction of idle cells.
 Header error control (HEC) generation and verification –
HEC is generated on the ATM cell header fields by the sender and
verified by the receiver. That is, the HEC is generated and
compared with the received value. If the cell header errors can be
corrected and the cell processed. If not, the entire cell is discarded
 Cell delineation – In the receiver, detection of cell boundaries
 Transmission frame adaptation – Adapts cell flow according to
the payload of the Physical level frame being used, e.g. for SDH
 Transmission frame generation and recovery – Transmission
frame usage will involve more than just packing ATM cells bit by bit
into a transmission frame and sending it out.

12/08/21 33
 ATM Layer

 This layer is independent of the physical medium over which

transmission is to take place.
 ATM layer consists of a stream of cells (OAM cells, data cells,
signaling cells)
 Generic flow control (GFC) function. This can be used to
alleviate short term overload conditions above the ATM layer, as it is
accessible by the user.
 Cell header generation and extraction. At the transmitter, adds
header information to a cell, and at the receiver removes it.
 Cell multiplex and demultiplex. At the transmitter, multiplex cells
into one continuous stream, and at the header demultiplex the cells
according to VPI and VCI values.
 Switching. Eventually a translation at ATM switches and
crossconnects is required (performed on the VCI and VPI separately
or on both simultaneously)

12/08/21 34
 ATM Cells
 Cells used by the ATM Layer are called ATM cells
 Contain 53 byte (5 byte header, 48 byte payload)
 Information field of an ATM cell carries ATM user
data
 “normal” user data
 Signalling data
 Metasignaling data
 Management data related to the ATM Layer
 ATM Cell Header Bits
 Priorities are provided via a Cell Loss Priority (CLP) bit in the
header
 Payload Type (PT) field, Idle Cells, Unassigned Cells,
Physical Layer, OAM Cells, Signalling Cells

12/08/21 35
 ATM cells (cont.)
 ATM transfers information in fixed-size units called
cells. Each cell consists of 53 octets, or bytes
GFC

12/08/21 36
ATM Cell-Header Fields
 General Flow Control (GFC):
Slow down senders in heavily used portions of the
network
Used in UNI, not in NNI
 Virtual Path Identifier (VPI):
conjunction with the VCI, identifies the next
destination of a cell
 Virtual Channel Identifier (VCI):
conjunction with the VPI, identifies the next
destination of a cell

12/08/21 37
ATM Cell-Header Fields (Cont.)
 Payload Type (PT)
Indicates whether the cell contains user data or control data
 user data

 signalling data

 VP-OAM data (Virtual Path - Operation and Maintenance)

 VC-OAM data (Virtual Cannel - Operation and Maintenance)

 Congestion Loss Priority (CLP)

Indicates whether the cell should be discarded if it encounters
extreme congestion
 Header Error Control (HEC)
Calculates checksum only on the header itself.

12/08/21 38
 QoS
 Process of determining and delivering the required QoS parameters for
each connection on an ATM network is referred as the traffic contract
specification
 Traffic contract:
 constrain data bursts, limit peak data rate, cell-loss rate …
 Traffic shaping:
 forcing your traffic to conform to a certain specified behavior(adhering to the
contract)
 allocate resources inside the network such that guarantees about availability of
bandwidth and maximum delays can be given
 Traffic policing:
 estimates the the parameters of the incoming traffic and takes some action if they
measure traffic exceeding agreed parameters
 QoS parameters
 CBR (Constant Bit Rate)
 VBR-NRT (Variable bit rate – non-real time)
 VBR-RT (Variable bit rate – real time)
 ABR (Available bit rate)
 UBR (Unspecified bit rate)

12/08/21 39
 Traffic management
 The process of controlling traffic on a UNI in an ATM network
 ATM traffic management are to deliver quality-of-service (QoS)
guarantees and provide overall optimization of network resources
 The various functions of ATM traffic management can be
categorized into three distinct elements
 Nodal-level controls – implemented in hardware and include queues
supporting different loss and delay priorities, fairly weighted queue-
servicing algorithms, and rate controls that provide policing and traffic
shaping.
 Network-level controls – the heart of any traffic-management
system, it is implemented in software including connection admission
control (CAC) for new connections, network routing and rerouting
systems, and flow-control-rate adaptation schemes.
 Flow control – involves adjusting the cell rate of the source in
response to congestion conditions and requires the implementation of
closed loop congestion mechanisms.

12/08/21 40
 Traffic management

 Parameters
 Cell loss ratio (CLR)
 Cell misinsertion ratio (CMR)
 Severely errored cell block ratio
 Mean cell transfer delay (MCTD)
 Cell delay variation (CDV)

12/08/21 41
 ATM Switching
 ATM switching principle
 The cell is received across a link on a known VCI or VPI value
 The switch looks up the connection value in a local
translation table to determine the outgoing port (or ports) of
the connection and the new VPI/VCI value of the connection
on that link.
 The switch then retransmits the cell on that outgoing link
with the appropriate connection identifiers.

12/08/21 42
 ATM Switching (cont.)
Data 23
Video
23 56
3 56
2
Data Voice Switch 4 65
37 34
5
76
1
6

In Out
Port VPI/VCI Port VPI/VCI
1 0/37 3 1/23
1 0/34 4 0/56
2 0/23 5 0/65
2 0/56 6 4/76
12/08/21 43
 ATM Switching (cont.)

12/08/21 44
 ATM Adaptation Layer
 An application’s data needing to be sent across an ATM
network typically will have to be adapted to the ATM
network
 This layer is responsible for mapping the service offered
by ATM to the service expected by the higher layers
 Segmentation and reassembly (SAR)
 SAR is responsible for presenting the ATM service to the higher
layers
 SAR makes cells of higher-layer data units and remakes the data
units at the destination
 Convergence services (CS)
 CS makes sure that the cell stream set up and sent is capable of
providing the needed service to the application
 CS is responsible for, at the transmitter, splitting the higher level
PDU into 48 octet chunks, and at the receiving side, to reassemble
the 48 octet chunks back into the original PDU.

12/08/21 45
 ATM Adaptation Layer (cont.)
 The AAL may enhance the service provided by the
ATM Layer to the requirements of a specific service
(user, control, management).
 Acts on ATM Layer data streams (mapping for the next
higher layer)
 Different requirements of the protocols on top of the AAL
→ several AAL protocols are required
 AAL protocols are characterized by a common set of
functions
 required by several protocols to be run over an ATM network
 specific adaptation requirements of protocols (originally
designed for other network types)
 Provides several Layer Services with different layers in OSI-
RM
 User data and Signalling data typically require adaptation by
these Services
12/08/21 46
 ATM Adaptation Layer (cont.)
 Up to now, 4 (basic) AAL protocol stacks (AAL types) are defined
 AAL1, AAL2, AAL3/4, AAL5
 AAL Types support different user data stream requirements
 AAL functions contain SAR functions and some CS functions
 CS of these stacks has been further sub-structured:
 Upper Service Specific CS (SSCS)
 Lower Common Part of the CS (CPCS)

12/08/21 47
 ATM Adaptation Layer (cont.)
 AAL1 - supports connection-oriented services that require constant bit
rates and have specific timing and delay requirements
 AAL2 - supports connection-oriented services that do not require
constant bit rates (in other words, variable bit rate applications like
some video schemes)
 AAL3/4 - this AAL is intended for both connectionless and
connection oriented variable bit rate services (originally two distinct
adaptation layers)
 AAL3 and 4 have been merged into a single AAL which is named
AAL3/4 for historical reasons
 AAL5 - supports connection-less variable bit rate data services
 Provides a smaller bandwidth overhead (then AAL3/4), simpler
processing requirements, and reduced implementation complexity
 AAL5 has been proposed for use with both connection-oriented and
connectionless services

12/08/21 48
 AAL1
 The layer services provided by AAL type 1 to the AAL user
are:
 Transfer of Service Data Units (SDU) with a constant source bit rate
and the delivery of them with the same bit rate
 Blocking / deblocking
 Transfer of timing information between source and destination
 Transfer of structured information between source and destination
 Indication of lost or defective information which is not recovered by
AAL type 1 if needed
 Typical Application: constant bit rate audio (e.g. ISDN-
Telephony)

12/08/21 49
 AAL1 - SAR/CS-PDU format
 SN: Sequence Number (4 bits)
 SNP: Sequence Number Protection (4 bits)
 CRC: Cyclic Redundancy Check (3 bit)
 P: Pointer SAR-PDU payload = SAR SDU
Cell header SN SNP
(8 bits) (47 octets)
 CSI:
Sequence
CS Indication CSI
count
CRC parity
(1 bit)
CSI = 0
 Parity (1 bit) Non-P format AAL user data (47 octets)
CSI = 1
 Offset (7 bit) P format
 sequence count (3 bit) P AAL user data (46 octets)

parity offset

12/08/21 50
 AAL 2
 The layer services provided by AAL type 2 to the AAL
user may include:
 Transfer of service data units with a variable source bit rate
 Transfer of timing information between source and destination
 Indication of lost or defective information which is not recovered
by AAL type 2, if needed
 AAL2 CPS fits the problem of low bit rates
 Transmission of only partially filled cells is necessary
 It avoids partially filled cells by multiplexing several data streams
into the same cell
 Typical application: variable bit rate high quality audio
and video

12/08/21 51
 AAL 2 - SAR-PDU format
SAR-PDU (48 octets)
SAR-PDU header SAR-PDU trailer

Cell SAR-PDU payload

SN IT LI CRC
header (44 octets)

 IT (Information Type) – indicates beginning of message, continuation

of message, or end of message and also a component of the video or
audio signal
 SN (Sequence Number)
 LI (Length Indication) (6 bit) – Indicates the number of octets of the CS
(Convergence Sublayer) PDU (Protocol Data Unit) that is included in
the SAR (Segmentation and Reassembly) payload
 CRC (Cyclic Redundancy Check) (10 bits) – Used to detect errors up to
two correlated bit errors in the SAR PDU

12/08/21 52
 AAL 3/4
 Supports the non-assured transfer of user data frames
 An internal multiplexing function allows the establishment
of several concurrent AAL type 3/4 user connections on
one ATM connection
 On each of such connections, the integrity of the data
sequence is maintained and transmission errors are
detected
 Two new appendices describe the multiplexing AAL type
3/4 connections on an ATM connection using the
Multiplexing Identification (MID) field and one procedure
for dynamic MID allocation

12/08/21 53
 AAL3/4 - SAR-PDU format
SAR-PDU (48 octets)
SAR-PDU header SAR-PDU trailer

Cell SAR-PDU payload

ST SN MID LI CRC
header (44 octets)

 ST: Segment Type (2bits)

 ST coding interpretation:
 00 COM: Continuation of Message
 01 EOM: End of Message
 10 BOM: Begin of Message
 11 SSM: Single Segment Message
 MID: Multiplexing Identifier (10 bits)
 SN: Sequence Number LI Length Indication (6 bit)
 CRC: Cyclic Redundancy Check (10 bits)

12/08/21 54
 AAL3/4 - CPCS-PDU format
CPCS-PDU (up to 65544 octets)

CPCS-PDU header CPCS-PDU trailer

SAR-PDU payload
CPI Blag BA Size PAD AL Elag Length
(44 octets)

 CPI: Common Part Indicator (1 octet)

 Btag: Beginning Tag (1 octet)
 BA Size: Buffer Allocation Size (2 octets)
 PAD: PADding (0 to 3 octets)
 AL: Alignment (1 octet)
 Etag: End Tag (1 octet)
 Length: length of CPCS-PDU payload (2 octets)

12/08/21 55
 AAL 3/4 - Services
 Message Mode service
 CPCS-SDU is passed across the CPCS interface in one CPCS-IDU
 provides the transport of a single CPCS-SDU in one CPCS-PDU.
 Streaming Mode service
 CPCS-SDU passes in one or more CPCS-IDUs across the CPCS interface
 transfer across the CPCS interface may occur separated in time
 provides the transport of all the CPCS-IDUs belonging to a single CPCS-
SDU in one CPCS-PDU
 internal pipelining function in the CPCS may be applied which provides
the means by which the sending CPCS-entity initiates the transfer to the
receiving CPCS-entity before the complete CPCSSDU is available
 includes an abort service by which the discarding of a CPCS-SDU
partially transferred across the interface can be requested

12/08/21 56
 AAL5
 Designed to support packet_oriented AAL user data transfer with
minimum transmission overhead but with nevertheless
`machine_friendly' PDU formats.
 The application of AAL5 for signalling message transport, for the
support of the DL_Core Service as used to implement e. g. the Frame
Relaying Bearer Service is currently standardised.
 It is also highly probable that AAL5 will be used to support at least
short term solutions for Video on Demand applications based on quasi
constant bit rate transmission of MPEG Transport Streams.
 The Message Mode service, Streaming Mode service, and assured and
non-assured operations for AAL type 5 are identical to those defined
for AAL type 3/4

12/08/21 57
 AAL5 - Protocols

 AAL5 SAR protocol

 The AAL5 SAR protocol only provides for a simple CPCS_PDU
delineation mechanism using the AUU bit in the cell header
to differentiate between end and ‘non_end’ CPCS_PDU
segments.
 AAL5 CPCS protocol
 The AAL5 CPCS protocol maps CPCS_SDUs of variable size
to/from the 48_octet ATM_SDUs, supports the transparent
transfer of an additional, separate octet of CPCS user data
and includes bit error and cell loss and misinsertion
detection capabilities.

12/08/21 58
 AAL5 - CPCS-PDU format
CPCS-PDU (n * 48 octets)
CPCS-PDU trailer

SAR-PDU payload CPCS

PAD CPI Length CRC
(44 octets) -UU

 PAD: PADing (0 to 47 octets)

 CPCS-UU: CPCS-to-CPCS User indication (1 octet)
 CPI: Common Part Indicator (1 octet)
 Length: Length of CPCS-PDU payload (2 octets)
 CRC: Cyclic Redundancy Check (4 octets)

12/08/21 59
 ATM Signaling and Addressing
 ATM signaling is initiated by an ATM end-system that
desires to set up a connection through an ATM network
 Signaling packets are sent on a well known*5* virtual
channel, VPI=0, VCI=5.

12/08/21 60
 UNI Signaling
 As shown below in the figure the signaling message is
transferred across the UNI using the services of SAAL (Signaling
AAL layer) layer in the control plane.

UNI Signaling

12/08/21 61
 UNI Signaling (cont.)
 Signaling procedures specify the sequence of message
exchanges to establish and release connections as shown in the
bounce diagram below. Many error conditions are accounted
which for simplicity has been removed here. An extreme simple
case is only considered.
UNI UNI
Source Network Destination
Setup
Setup
Call proceeding
Call proceeding
Connection
Connection Connect ack
Connect ack

Release
Release
Release complete
Release complete
12/08/21 Signaling Example 62
 UNI Signaling (cont.)
 Sequence of messages are
1. Host A sends a SETUP message on VPI/VCI = 0/5 (Default value for AAL5)
identifying the destination (host B) and Other Parameters specifying details of
the requested connection.
2. The first switch analyzes the contents of the SETUP message to see whether it
can handle the requested connection. If the switch can handle the request, the
network returns a CALL ROCEEDING message to the host containing the
VPI/VCI (0/5)for the first link. It also forwards the SETUP message across the
network to the destination.
3. Upon arrival of the SETUP message, the destination sends a CALL PROCEEDING
message.
4. If the destination accepts the call, it sends a CONNECT message that is
forwarded across the network back to host A. The CONNECT messages trigger
CONNECT ACKNOWLEDGE messages from the network and eventually from the
source.
5. The connection is now established, and the source and destination can
exchange cells
6. Either party (caller or called) can subsequently initiate the termination of the
call by issuing a RELEASE message. This step will trigger RELEASE COMPLETE
messages from the network and from the other party.

12/08/21 63
 PNNI Signaling
 PNNI uses source routing where the first switch selects the route to the
destination. Here, the source host requests a connection to host B by sending
a SETUP message, using UNI signaling. The first switch carries out the
connection admission control (CAC) function and returns a CALL PROCEEDING
message if it can handle the connection request. The first switch maintains
and uses a topology database to calculate a route to the destination that can
meet the requirements of the connection contract. The route consists of a
vector of switches that are to be traversed.
 The SETUP message propagates across the network, using the source route.
Each switch along the path performs CAC and forwards the SETUP message
along the next hop if it can handle the connection request.
 It also issues a CALL PROCEEDING message to the preceding switch along the
route.
 If the destination accepts the call, a connect message is returned across the
network to the source. Connection release proceeds in similar fashion as
shown in Figure below

12/08/21 64
 ATM Addressing
 These existing protocols all have their own addressing schemes and
associated routing protocols. One proposal was to also use these
same addressing schemes within ATM networks
 Hence ATM endpoints would be identified by existing network layer
addresses (such as IP addresses), and ATM signaling requests would
carry such addresses

Peer Model

Overlay Model

12/08/21 65
 ATM
 ATM Introduction
 What is an ATM (Asynchronous Transfer Mode)?
 ATM Overview
 ATM Topology
 Standard
 Basic principle
 ATM Network Interfaces
 ATM Cells
 ATM virtual connections
 ATM Reference Model
 Physical Layer
 ATM Layer
 ATM Adapted Layer
 ATM Services

12/08/21 66
 ATM Services
 Frame-relay bearer services
 Connectionless services
 LAN Emulation (LANE) services
 ATM video and audio services – Video over ATM
 ATM circuit emulation services – Voice over ATM(CES)

12/08/21 67
 Frame-relay bearer services
 Connect frame-relay bearer service (FRBS) networks
over the ATM network or even to establish connections
that interconnect frame-relay network and ATM
networks.
 To support FRBS, the ATM network must provide the
interworking function (IWF) at the AAL (AAL 5).

FR ATM
FR
FRRouter
Router IWF IWF FR FR
FRRouter
Router
networ networ
k k

ATM Interworking Function

12/08/21 68
 Connectionless services
 LANs are connectionless networks, the network
linking them together should be connectionless also.
These will use AAL-3/4 for transporting their traffic
over an ATM network.
 Uses VPI= 0, VCI = 15 as the default connection for
connectionless services over a UNI
LAN A CLNS

User A Router

Workstation Router
To other
User B LAN B ATM Switch
Cells ATM
nodes

IP Datagram
12/08/21 69
 LAN Emulation Services (LES)
 Where the ATM network may connect distant LANs totally transparently to users
and application and even connect separate private ATM networks over an
intervening LAN
 LES actually emulates a LAN itself, making the two separate LANs think they are
actually all one big LAN, whether they are separated by two floors or 2000 kms.
1. Clients get
recipient's 3. Messages for
LES server ATM clients are
address from
LES and setup delivered directly.
a VC.

ATM client

ATM client
ATM Switch ATM Switch
2. Clients
sends Bridge
4. Messages for
messages on
non-ATM clients
VC.
are forwarded
through a bridge
12/08/21 Broadcast/unknown server 70
Non-ATM client
 LAN Emulation Services (cont.)

12/08/21 71
 LAN Emulation Services (cont.)
 LANE components
 LAN Emulation Client (LEC): A LEC is the entity in an end system that performs
data forwarding, address resolution, and other control functions for a single end-
system within a single ELAN. A LEC also provides a standard LAN service interface
to any higher layer entity that interfaces to the LEC.
 LAN Emulation Server (LES): The LES implements the control function for a
particular LAN. There is only one logical LES per LAN, and to belong to a particular
LAN means to have a control relationship with that LAN's particular LES. Each LES
is identified by a unique ATM address. The operation of the LES is described
below.
 Broadcast and Unknown Server (BUS): The BUS is a multicast server that is
used to flood unknown destination address traffic and forward multicast and
broadcast traffic to clients within a particular LAN. Each LEC is associated with
only a single BUS per LAN, but there may be multiple BUSs within a particular LAN
that communicate and coordinate in some vendor- specific manner; this action is
outside the scope of the Phase 1 LANE protocol. The BUS to which a LEC connects
is identified by a unique ATM address. In the LES, this is associated with the
broadcast MAC address ("all ones"), and this mapping is normally configured into
the LES.

12/08/21 72
 ATM video and audio services

 Sending video and the accompanying audio

soundtrack on ATM networks is quickly becoming as
complex an offering as data
 All video used to be constant-bit-rate video, so AAL-1 was a
nice fit
 With a video compression, it is now just a lowdelay,
variable-bit-rate service (provided by AAL 2) is a good fit as
well.

12/08/21 73
 ATM circuit emulation services
 Take existing leased private lines and run them over the
ATM network.
This is the easiest way to bring ATM services into an existing network
 Use the ALL-1 interface (Constant bit rate)

CBR CBR
PBX channels channels PBX

Telephone TDM IWF ATM IWF TDM Telephone

ATM Interworking Function

T-1 or E-1 links

12/08/21 74
 Q&A

12/08/21 75