Advanced Intelligent Networks - A SEMINAR
Introduction
The
telecommunications industry is going through a fundamental change, brought
about by competition and the need to differentiate services. ‘Intelligent
networking’ is one of the key
enablers for service differentiation which will revolutionise
telecommunications over the next decade.
This paper
concentrates on the technical concepts and systems which make up an intelligent
network. However, it should be noted that the main motivation for this
technology has been its ability to add new revenues and to provide competitive
services faster than other competing networks.
Fig.1. The intelligent network concept
HISTORICAL
DEVELOPMENT OF THE ADVANCED INTELLIGENT NETWORK
¨
1967 The Freephone service is
introduced in the
¨
1970s Stored-programme-control switching systems and
common-channel-signalling systems are introduced into the AT&T network.
¨
1980 AT&T introduce a proprietary distributed call control system
with network databases using common channel signalling to control credit card
verification and Freephone (800) services.
CCITT publish the Yellow Book containing
CCIIT No. 7 Signalling,
for circuit-related call control.
CCITT Red Book work starts on the signaling connection control part
(SCCP).
¨
1984 Work starts on CCllT No. 7 signalling
for noncircuit-related transactions (TCAP). Also the Intelligent Network 1 (IN/l) standardization starts in
¨
1987 Intelligent Network 2 (IN/2) concepts published by
Bellcore.
¨
1988/9 Intelligent Network 1 deployed in
¨
1990 Bellcore publish initial
AIN Requirements.
¨
1991 / 92 First AIN trials in the
¨
1992 CCITT publish Q.1200 series
of intelligent network standards.
DEVELOPMENT
OF THE INTELLIGENT NETWORK CONCEPT
The expression
‘intelligent network emerged in 1983/4 as part of the work done in the
USA by Bellcore and others to define the first nonproprietary network design
capable of supporting freephone, credit card verification and virtual private
network services.
First step: 1980-84
The first step in
the development of advanced intelligent networks was to separate the logic and
data needed to implement a special service away from the switching system and
to locate it on remote databases. Interaction between the switching systems and
these databases was defined only for the application software of the specific services which the databases
supported, e.g. Freephone.
Second step: 1984-87
The second step was
to make the interface between the switching system and the database a standard
interface based on CCI’IT Signalling System No. 7. This open interface enabled
intenvorking between equipment from different suppliers.
The opening-up of
the switch-to-database interface, as provided for in the Intelligent Network 1 (IN/l)
concept published by the American Bellcore company, was a significant step
towards a more distributed call-processing environment.
Third step: 1987-93
The third step
essentially defined what we now call ‘Advanced Intelligent Network (AIN). This
has become the latest term to apply to the intelligent-network architecture and
system functionality which is being trialled and tested in
The release plan for
AIN has undergone some revisions since 1989 with redefinition of the AIN
near-term and longterm architectures. These changes characterise the problems
in trying to define an open call-processing environment for flexible,
service-independent systems.
IN
ARCHITECTURE AND SYSTEM FUNCTIONALITY
The advanced
intelligent-network architecture builds on the following key concepts:
¨
The separation
of service control logic from service switching the logical separation of
applications from underlying processing platforms
¨
The control of
service switching by discrete call control instructions called ‘functional
components’ (FCs)
¨
The support of
multiple service applications on a standardised service logic execution
environment (SLEE)
¨
The use of
service creation environments to create, simulate and test new services prior
to delivering them to target service control platforms
¨
The
standardisation of the call control model.
Fig. 2 show the architectures produced by Bellcore.
Physical elements
The intelligent
network architecture consists of network environments supported by the
following physical elements, which are implemented within a telecommunications
network.
The service switching point
(SSP), also known as AIN switch
capabilities (SC), (Figs. 1, 3, 4) is
an enhanced switching system. This node has extra call-processing software to
enable the switching system to recognise when call-processing at the external
service control point (SCP) is required. When the conditions for external
processing are detected a query/begin transaction message is launched to the
service control point, where the application service logic will decide how to
process the call.
The special call
control conditions are called triggers. In intelligent networking, a switching
system can be equipped with originating triggers, which operate at the originating
switching system during call set-up, mid-call triggers, which work during the
speech phase of a call, and
terminating triggers, which operate during the set-up of the call at the
terminating switching system.
To be able to communicate with the service control point the
service switching point must have at least the Red Book version of CCITT
signalling System No. 7 (see Section 4),
to allow long (272 octet) MTP (message transfer part) signalling
messages, SCCP (signalling connection control part) and TCAF’ (transaction
capabilities application part) to carry the IN call-processing instructions.
Section 4 provides more details
on CCITT No. 7. The service control point (SCP) is a network database which is
one of the IN physical elements which also contain service control logic.
Typically the platform is a switching system or a commercial computer. The SCP
basically consists of two layers of software: the service control point’s
service logic execution environment (SLEE) and the service logic programs
(SLPs) themselves.
The service logic
execution environment supports multiple service logic programs, which represent
multiple services. The SLPs are the software application programs created by a
service creation environment to run particular services. Within the AIN
architecture the service logic execution environment can exist on the SCP nodes
and on ‘adjunct’ processors (see below). It provides the interfaces to the
operational support and management systems to enable service provision and
administration. The management function allows the addition, modification or removal
of service logic programs as well as modification of service data and functional
components. Within the Bellcore AIN and CCITT standards an extensive set of functional
components is defined. These functional
components represent the
application functions carried in the protocol between the service switching
point and the other entities in the intelligent network.
An adjunct processor (AP) also
provides the service control function but, rather than being physically
separate at a service control point, it is located with the service switching
point, to which it is linked by a high-speed interface.
Intelligent peripherals (IPS) handle
specialized interactions between the user and the intelligent network. The IP
has resources such as tones and announcements, voice recognition and voice
synthesis. As a result it has
both signalling and voice circuits to the service switching point.
If the SCP instructs
the SSP to route a call to an intelligent peripheral, the latter is instructed
to perform a particular function, for example ‘play announcements and collect
digits’ (the caller is requested to provide further information needed to route
the call by dialling extra digits). On completion of the announcement any DTMF (dual-tone
multifrequency) digits dialled by the caller are passed to the service control
point. The intelligent peripheral will then have the additional information required
to set up the call.
The service node
(SN) is capable of controlling IN services in its own right as well as engaging
in flexible interactions with its users. It is essentially a combined service
control point and intelligent peripheral and so has voice circuits as well as
signalling to the service switching point, i.e. ISDN links instead of CCITT No.
7. If required a service node can enter into a dialogue with a service
switching point.
Functional entities and logical environments
The IN standards
define the functions, referred to as functional entities (
AIN groups these
functional entities into local environments and then maps them onto physical
network elements (Fig. 4). This
mapping is as follows. The AIN switching environment maps to the service
control point, the service node execution environment maps onto the service
node and the resource control executive environment maps onto the intelligent
peripheral.
CCITT SIGNALING SYSTEM NO.7 FOR
INTELLIGENT NETWORKS
The CCTT No.7 signalling system is the means by which calls
are controlled in digital telecommunication networks and is a prerequisite of
an intelligent network. CCITT No. 7 is designed for a signalling network made up of
- Signalling
points addressable by point codes
- Signalling
links across which messages are
- Transmitted
securely and
- Combinations
of links into linksets and routesets
The transaction capabilities
application part (TCAP) protocol consists of two elements which support the flexible
interchange of intelligent-network call control instructions. The first allows
a dialogue, or ‘transaction’, to be established between two applications. To
initiate a transaction, an application sends a ‘begin’ message to the other
application. If the other application is able to undertake the transaction it
unpacks the message, processes the components and, either 0
- Responds
with a ‘continue’ package type, to continue with a more complex
transaction, or
- Responds
with an ‘end’ package type to terminate the transaction.
Within each TCAP message is a
collection of message components. TCAP can bundle up a number of these components
in a flexible way to process call-processing instructions.
TCAP components enable actions to be
started and completed in a simple, yet very flexible way. The components are:
- ‘Invoke’ to start an
operation
- ‘Return results’ to
successfully complete an operation
- ‘Reject’
to inform the originating application that a lowlevel protocol syntax
error has been identified, i.e. that the component included parameters
which the receiver did not understand or could not decode
- Return
error’ to indicate that an application (semantic) error has occurred, i.e.
that the message has been understood by the protocol but that the
parameter value is not supported by the application.
An Application of Advanced intelligent Network
Technology to Personal communication services:
Personal communication services (PCS) are necessary for
users to free communicate with anyone at any where according to their
demand. In order to realize the above
mentioned requirements, a wireless access system must be introduced to the
telecommunication networks, in addition to the wide access system.
This paper proposes a network architecture which is based
on an advanced intelligent network in
addition to wired access network, and which makes possible terminal and personal mobility, service
personalization, and service portability, so providing universal personal
communication.
PCSN
Based on Advanced IN
This shows a network configuration for the personal
communication service network based on the advanced IN structured PSTN and
ISDN. The SDF in the SCP retains the
location information, the authentication information, and the parameter for the
service customization to support personal and terminal mobility. These data, in other words, are managed by
the location register. Two types of
location register, T-LR and P-LR are needed to support terminal mobility and
personal mobility independently. The SLP
is defined separately as SLP(c) for customization and SLP (N) for network
control, and is down loaded to the SCP.
In addition to the BCSM for the CCF / SSF supporting the call related
interaction, the BCUSM for the CUSF supporting call unrelated interaction is to
be implemented. The SRF supports DTMF
based in channel interaction.
CCITTNo. 7 is
designed for a signalling network made up OE
¨
signalling
points addressable by point codes
¨
signalling
links across which messages are transmitted securely and
¨
combinations of
links into linksets and routesets
The transaction
capabilities application part (TCAP) protocol consists of two elements which
support the flexible interchange of intelligent-network call control instructions.
The first allows a dialogue, or ‘transaction’, to be established between two
applications. To initiate a transaction, an application sends a ‘begin’ message
to the other application. If the other application is able to undertake the
transaction it unpacks the message, processes the components and, either 0 responds with a ‘continue’ package type, to continue with
a more complex transaction, or e responds with an ‘end’ package type to
terminate the transaction.
Within each TCAP
message is a collection of message components. TCAP can bundle up a number of
these components in a flexible way to process call-processing instructions.
TCAP components
enable actions to be started and completed in a simple, yet very flexible way.
The components are:
¨
‘invoke’ to
start an operation
¨
‘return
results’ to successfully complete an operation
¨
‘reject’
to inform the originating application that a lowlevel protocol syntax error has
been identified, i.e. that the component included parameters which the receiver
did not understand or could not decode ‘return error’ to indicate that an
application (semantic) error has occurred, i.e. that the message has been understood
by the protocol but that the
¨
parameter value
is not supported by the application.
Services
With the service
switching functions installed in the switching systems, and the CCITT No. 7
signalling in place, the next item to cover is the services. Advanced
intelligent networks are designed to have a service-independent architecture,
i.e. they are not designed or optimised for a particular service or feature. The
intent is to create the flexibility to add new services without changing the
network infrastructure. However, Bellcore’s AIN Release 1 and CCI’IT capability
set 1 are validated against a set of services. CCI’IT’s service list includes
25 services, for example abbreviated dialling and account card calling.
Services are realised by:
a. Creating a set of reusable service elements enabling
bundling and unbundling of service features. In the CCI’IT IN recommendations
these are called SIBs - service independent building blocks. These SIBS represent
service functions in the network that could be liked together in many different
ways.
b. Using a ‘service creation’ system, to link together selected
SIBS to create new services. The SIBs are reusable for creation of other services. All service features
are described by one SIB or chain of SIBs.
c. installing the so created services onto the service control
point, service node or adjunct platform as service logic programs (SLPs). Once
loaded the network management systems will provide service data for service
subscribers. When the service logic program is activated the SCP/service
node/adjunct will send out call-processing instructions (functional components)
in accordance with the IN application protocol ONAP).
Standards
Since 1987 there has
been a set of advanced-intelligent network attributes which have been present
throughout the standardisation efforts; these are:
¨
Distribution of intelligence to allow centralised or
network-wide service control
¨
Service-independent
protocols and systems
¨
Faster
deployment of new services
¨
Service
creation by switch suppliers, third parties or network operators.
Conclusions
IN standards are
still maturing. CCITT has a significant standardisation task ahead and the goal
of using intelligent networking as a generic means for flexible service delivery,
in the assumed network infrastructure for UFT and third generation mobile
systems (TGMS), will be challenging.
Advanced intelligent
networking can be viewed as a key telecommunications technology for the 1990s
as product and market differentiation becomes a stronger competitive issue for
suppliers and network operators alike. The move towards multimedia broadband communications
and the need to separate out call control from the underlying bearer service is
a fundamental need.
The roll-out of
advanced intelligent networks is dependent on a number of other technology
building blocks being in place. Key to the underlying network infrastructure is
the use of digital exchanges and Blue Book CCITT No. 7 Signalling with TCAP and
SCCP. It should be noted that the future IN services will probably need the new
features of SCCP segmentation and TCAP context negotiation.
Early ambitious IN/2
and IN1+ plans have moderated expectations on roll-out. Practical
implementation options have been progressively defined, making the transition
to live network deployment of the infrastructure slower than originally
envisioned.
The early views of
AIN as representing a convergence of telecommunications and commercial computer
AIN systems have not been fully realised, though trends in intelligent-networking
partnerships between computer companies and switch suppliers continue. The development
of ‘proprietary AI” switching and service control systems needs extensive
network and service planning and development. The introduction of integrated switching/service
control ‘feature nodes’ appears to represent the quickest way to introduce
service independent platforms capable of offering rapid service delivery with
the minimum of change.
AIN standards
continue to be developed, but important IN standards work packages need to be
completed over the next few years to be an effective way of reducing development
costs. The scope and complexity of the IN standards have grown significantly,
and the benefits to both switch suppliers and network operators of aligning to standards
have meant a more pragmatic, albeit slower,
standardisation effort.
The advanced
intelligent network is becoming a reality for those network operators involved
in AIN trials in the
With the publication
of an extensive set of CCITT standards the future use of AIN as a core telecommunications
call-processing technology looks very positive. Finally, it is worth noting
that the experiences gained in using the latest AIN systems mean that the
solution to the original problem of software development lead-times now appears
to be in sight. Service-creation environment tools are available which produce
new senrice logic programs for network operators in a few days or weeks. This
is the first evidence of the software productivity problems being cured, and
moves the new focus of service productivity and roll-out to management systems
and processes.
REFERENCES
- C.D. Sharp and K. Clegg
- IEEE IN’96 Workshop, 21st – 24th
April, 1996,
Comments
Post a Comment