Lecture 6 - ISDN User Part in SS7: S38.3115 Signaling Protocols - Lecture Notes
Lecture 6 - ISDN User Part in SS7: S38.3115 Signaling Protocols - Lecture Notes
Lecture 6 - ISDN User Part in SS7: S38.3115 Signaling Protocols - Lecture Notes
lecture 6
Introduction
ISUP is a national and international telephony network signaling protocol for
call setup, supervision and release between exchanges in wire line and mobile
networks. It supports a rather wide range of supplementary services for ISDN
networks. Support of supplementary services achieved a sufficient level in
1992 and ISUP deployment in Finland started in 1994. ISUP mainly runs on
top of MTP of SS7.
The first telephony signaling protocol for SS7 was the Telephony User Part or
TUP. It was designed before DSS1. It achieved fast call setup because of the
use SS7 signaling infrastructure but supported only a limited set of
supplementary services. National variants of TUP exist for many countries due
to the need to take into account interworking requirements with different
lecture 6
national analogue signaling requirements. For example in the UK, the user part
corresponding to TUP is called the National User Part or NUP, in Finland it is
just called national TUP. The message coding style in TUP slightly differs
from DSS1 in favor of using a lesser amount of bits. TUP was difficult to
extend while preserving backward compatibility.
Operators and exchange vendors were looking for a further step towards more
global markets based on a wide support of supplementary services. Therefore,
instead of trying to extend TUP, a decision was made to specify a new user
part better aligned in terms of services with DSS1 and also taking into account
better support for backwards compatibility.
Compatibility of two versions or variants of a protocol means that two
instances conforming to the different versions of the specification can talk to
each other and can agree on the set of services support by both versions. This
eases the deployment of the new versions because deployment can progress
node by node and two ends of a signaling link can run using different versions
for a time. This is important because of the high availability requirements
placed on exchanges and the high cost of software upgrades. A major software
upgrade in an exchange may take a whole working day and often executed
during a night to reduce disturbance to live traffic. Besides the cost of the
software itself, one needs to consider also the operational costs incurred.
ISUP deployment was rather easy because the underlying infra of MTP was
already in place and had been in use with TUP.
Most recent developments in ISUP during this decade include the addition of
charging messages suitable for European countries although for example in
Finnish version charging messages were present even earlier. They are useful
when operators change their tariffing frequently using special campaign prices.
With their help other operators do not need to change their tariffing
configuration when another operator makes a tariff change.
Another development has been the specification of how to carry ISUP over IP
-networks. This is called the SIGTRAN architecture and we will discuss it on
the last lecture.
lecture 6
Operators and vendors widely believed that support for a broad set of standard
supplementary services was necessary for getting the ISDN phone market off
the ground which did not seem to happen although implementable DSS1
specifications were produced in 1988. A set of supplementary services
specifications called Euro-ISDN was produced in 1992. TUP naturally did not
support it but ISUP did. We will describe some of the Euro-ISDN services in a
later section.
In addition to bearer services implementable over 64 kbit/s timeslots such as
speech, audio at 3.1 kHz, audio at 7 kHz and unrestricted 64 kbit/s (data)
channel, ISUP also supports multi-slot services (2 64 kbit/s, 384 kbit/s, 1536
kbit/s and 1920 kbit/s all in the unrestricted mode). These services were
deemed necessary for video conferencing and data connections. In practice
their provision over PCM infra is however cumbersome because transmission
resources in practice need to dedicated to either 1 64 kbit/s services or to a
multi-slot service at a certain speed. When service penetration is low, this is
costly. The penetration staid low for example because telephone operators did
not know better than charge N times the cost of one slot for N time-slots.
Clearly, this was unattractive to users when Internet with flat rate charging
came along very quickly on the heels of ISUP deployment.
In terms of services support DSS1 and ISUP form the uniform services
offering of Narrow band ISDN. If a service is supported in DSS1 and needs
some signaling over wide area, that support is either present or can be added in
ISUP.
Identification in ISUP
Subscribers (A calling party, B called party and C if needed in a
supplementary service) are naturally identified with telephone numbers. ISUP
assumes that these numbers are routable in exchanges. If Number Portability is
supported in the network, directory numbers need to be translated into routing
numbers prior to ISUP signaling.
An exchange processes many simultaneous calls. Identification among
simultaneous calls between two exchanges is based on Circuit Identification
Code (CIC) and optionally on Call References. A CIC is a 12 bit field just
after the routing label that we showed when discussing MTP. The value of
CIC is an integer and each exchange needs to maintain a configuration table
that maps particular PCM lines and timeslots to CICs and vice versa. The
combination of routing information for ISUP message is depicted in Figure
6.1.
The use of mandatory CIC for call identification ties the ISUP state machine
for the call to the time-slot that is used for the call. This means that there can
be no signaling unless a voice circuit is reserved.
48
8n bits
ISUP
message
lecture 6
data
subservice
field
DCBA
service code
0101=ISUP
0100=TUP
CIC
4bits
12bits
SLS
OPC
4bits
14 bits
DPC
14 bits
lecture 6
Call Stages
set-up
ringing tone
IAM
SAM
ACM
CPG(ALERT)
ringing
Call Progress(ALERT)
call
speech
release
ANM
Answer Message
MPM
REL
Release Message
RLC
lecture 6
Message acknowledges that the next exchange has indeed received the REL
and that all resources that were reserved for the call can be released.
For the purpose of establishing the call tariff dynamically on a call by call
basis, ISUP supports charging messages. Figure 6.2 shows the charging
messages in the Finnish ISUP. Later in this lecture we will show the messages
in the ETSI ISUP. The idea is that each operator may set its tariffs
independently and that users should reap the benefits of lower campaign or
regular tariffs immediately without other operators having to alter their own
charging by a management command. If no charging message are exchanged
on a call by call basis, the result is that the originating exchange has calculate
the tariff based on its own configuration information that can not take into
account the tariff changes that may have been made by the transit or
terminating operator.
lecture 6
To help with upgrading procedures, from 1992 onwards ISUP has conformed
to forward compatibility rules and procedures. The idea is that two versions of
ISUP should be able to talk to each other and agree on the minimal level of
functionality supported by both.
To achieve the goal, a message in a new version may carry different indicators
referring to new information elements that were not present in the older
version. These indicators may tell the receiving ISUP to
Pass on the information element without processing it
Release the call if the element is unknown
Notify the sender if the element is unknown
Discard the message if the information element is unknown
etc
In addition the types of changes that are possible are limited. The following
should not be changed in ISUP:
Protocol procedures, messages, information elements, coding, except to
correct an error in the protocol.
Semantics of existing info elements.
Formatting and coding rules.
Adding new parameters into mandatory part of Messages, Optional
part can be extended.
Order of information items in an Information Element of variable
length, new items can be added to the end of the IE.
It is also agreed that
Information item value =all 0 == non-significant value.
lecture 6
lecture 6
Call Forwarding
Call Forwarding No-Reply (CNFR)
The idea is that when the callee does not answer within a timeout (1530s),
the call is forwarded by the terminating exchange into a new number (Cnumber). A subscriber has procedures for modifying the C-number.
The service is used for example to forward the incoming call to a voice mail in
case of no-answer.
Call Forwarding Unconditional
Immediately, when a new call arrives at a number, the call is forwarded to a
new number (C-number). A subscriber has procedures for modifying the Cnumber.
This service is useful during longer periods of absence.
Call Forwarding on Busy
If an arriving call hits a busy condition of the callee, the call is forwarded to a
new number (C-number).
Call Transfer
Call transfer is a service where the callee answers the call and then after the
call established (having probably talked to the caller) transfers the call to a
new number (C-number).
Featurism
It is a typical development in communications technology that systems tend to
grow over time. New and new requirements are assigned and implemented in
systems that have achieved some level of penetration or become popular.
When a system is sold and the buyer has to answer the question: is it better to
have this added feature or not to have it, the answer almost invariably is yes,
Raimo Kantola (2008)
lecture 6
of course it is better have it. This applies even when the person answering the
question does not know anything about the feature or has made any analysis of
the usefulness of the feature in terms of business. This person, although a
decision maker, typically has no idea about the cost of implementing these
features either.
The result is that systems become gradually more and more complex. We call
this phenomenon featurism.
A good example of featurism is the long lists of supplementary services that
have been specified and implemented in public network switching systems and
PABXs. There are hundreds of them. Only a few are really useful. In our
presentation we tried to pick our example from the most common and useful
ones.
Legacy method
For analogue signaling system interworking (e.g. analogue subscriber
signaling and R2) ITU-T (or actually its predecessor CCITT) used a method
Raimo Kantola (2008)
10
lecture 6
Layer method
In layer based interworking specifications the model is that an incoming
signaling system is connected to an outgoing signaling system through call
control. Incoming signals create primitives that go from the incoming
signaling system to call control. These are indications if the incoming message
was a new request or confirmations if the incoming message is seen as an
acknowledgement to an earlier message. Call control will issue Request
primitives towards the outgoing signaling system related to earlier events on
the incoming side. Based on an event that is seen as an acknowledgement in
the backward direction, call control will issue a response primitive towards the
incoming signaling system.
Using the layer method, interworking of for example DSS1 and ISUP has been
specified using signaling message flow charts between DSS1, Call Control and
ISUP for a significant number of signaling scenarios. In addition information
mapping tables are provided. The tables give details on how information from
one system is mapped into information elements in the other.
11
lecture 6
ISUP
Setup
Exchange
Exchange
Setup_ack
Info
Call_proceeding
Alerting
Connect
Connect_ack
Conversation or data transfer phase
Disconnect
Release
Release_comp
Figure 6.3: Interworking of DSS1 and ISUP for Successful call setup
The flow chart in Figure 6.3 shows the mapping of messages in DSS1 and
ISUP.
Such flow charts are needed for B-subscriber busy, no-answer, call forwarding
etc.
In a detailed interworking flow chart the interworking specification shows the
actions taken by call control such as: reservation of outgoing timeslot, forward
through-connection, backward through-connection, bi-directional throughconnection and tear-down of through-connection.
12
lecture 6
Finally, when all charging messages have been acknowledged, ACM will
complete call setup and charging determination.
Charging registration
and generation point
IAM
CHG(AoC)
CHGA
CHG(AoC)
CHGA
ACM
IAM
CHG
CHGA
ACM
13
lecture 6
- Communicating
state machine
- Statement
-Client/server
- Illustration:
use cases
Concept
Purpose
Between who?
Identification
-of communicating
parties
- of simultaneous
sessions
Structure
Collaboration
Modeling
Protocol
-versions: compatibility
- interworking with other
- makes use of
- is made use by
Messages
Encoding
-user to user
- user to LE
- between network nodes
- - vertical or horizontal
-message types
- mandatory info
- options?
- fixed structure vs
variable str.
-analogue
-digital: binary
-digital: text based
14