What is an Intelligent Network?
Overview of Intelligent Network Architecture
Over the last thirty years, one of the major changes in the implementation of Public Switched Telephone Networks (PSTNs) has been the migration from analog to digital switches. Coupled with this change has been the growth of intelligence in the switching nodes. From a customer’s and network provider’s point of view, this has meant that new features could be offered and used.
Since the feature handling functionality was resident in the switches, the way in which new features were introduced into the network was by introducing changes in all the switches. This was time-consuming and fraught with the risk of malfunction because of proprietary feature handling in the individual switches.
To overcome these constraints the Intelligent Network architecture was evolved both as a network and service architecture.
In the (Intelligent Network) IN architecture, the service logic and service control functions are taken out of the individual switches and centralized in a special purpose computer. The interface between the switches and the central computer is standardized. The switches utilize the services of the specialized computer whenever a call involving a service feature is to be handled. The call is switched according to the advice received by the requesting switch from the computer. For normal call handling, the switches do not have to communicate with the central computer.
1 Objectives of the Intelligent Network
The main objectives of the IN (Intelligent Network) are the introduction and modification of new services in a manner that leads to a substantial reduction in lead times and hence development costs, and to introduce more complex network functions.
An objective of IN (Intelligent Network) is also to allow the inclusion of the additional capabilities and flexibility to facilitate the provisioning of services independent of the underlying network’s details. Service independence allows the service providers to define their own services independent of the basic call handling implementation of the network owner.
The key needs that are driving the implementation of IN are :
- Rapid Service Deployment
Most businesses today require a faster responses from their suppliers, including telecommunication operators. By separating the service logic from the underlying switch call processing software, IN enables the operator to provide new services much more rapidly.
- Reduced Deployment Risk
Prior to IN (Intelligent Network), the risk associated with the deployment of new services was substantial. Major investments had to be made in developing the software for the services and then deploying them in all of the switches.
With the service creation environment available, the IN services can be prototyped, tested and accessed by multiple switches simultaneously. The validated services can then be rolled out to other networks as well.
- Cost Reduction
Because the IN (Intelligent Network) services are designed from the beginning to be reusable, many new services can be implemented by building on or modifying an existing service. Reusability reduces the overall cost of developing services. Also, IN is an architecture-independent concept, i.e. it allows a network operator to choose suitable development hardware without having to redevelop a service in the event that the network configuration changes.
Prior to IN, due to the complexity of switch-based feature handling software, the considerable time frame required for service development prevented the provider from easily going back to redefine the service after the customer started to use it. With IN, the process of modifying the service or customization of service for a specific customer is much less expensive and time-consuming.
The customization of services is further facilitated by the integration of advanced peripherals in the IN (Intelligent Network) through standard interfaces. Facilities such as voice response systems, customized announcements, and text to speech converters lead to better call completion rates and user-friendliness of the services.
Building upon the discussion in the previous section, one can envisage that an IN would consist of the following nodes :
- A specialized computer system for – holding service logic, feature control, service creation, customer data, and service management.
- Switching nodes for basic call handling.
- Specialized resources node.
The physical realization of the various nodes and the functions inherent in them is flexible. This accrues from the “open” nature of IN interfaces.
Let us now look at the nodes that are actually to be found in an IN implementation.
The service logic is concentrated in a central node called the Service Control Point (SCP0.
The switch with basic call handling capability and modified call processing model for querying the SCP is referred to as the Service Switching Point (SSP).
Intelligent Peripheral (IP) is also a central node and contains specialized resources required for IN (Intelligent Network) service call handling. It connects the requested resource towards an SSP upon the advice of the SCP.
Service Management Point (SMP0 is the management node that manages services logic, customer data and traffic, and billing data. The concept of SMP was introduced in order to prevent possible SCP malfunction due to on-the-fly service logic or customer data modification. These are first validated at the SMP and then updated at the SCP during lean traffic hours. The user interface to the SCP is thus via the SMP.
All the nodes communicate via standard interfaces at which protocols have been defined by international standardization bodies. The distributed functional architecture, which is evident from the above discussion, and the underlying physical entities are best described in terms of layers or planes. The following sections are dedicated to the discussion of the physical and functional planes.
3 Physical Plane
Service Switching Point (SSP)
The SSP serves as an access point for IN (Intelligent Network) services. All IN services calls must first be routed through the PSTN to the “nearest” SSP. The SSP identifies the incoming call as an IN service call by analyzing the initial digits (comprising the “Service Key”) dialed by the calling subscriber and launches a Transaction Capabilities Application Part (TCAP) query to the SCP after suspending further call processing. When a TCAP response is obtained from the SCP containing advice for further call processing, SSP resumes call processing.
The interface between the SCP and the SSP is G.703 digital trunk. The MTR, SCCP, TCAP and INAP protocols of the CCS7 protocol stack are defined in this interface.
Service Control Point (SCP)
The SCP is a fault-tolerant online computer system. It communicates with the SSPs and the IP for providing guidelines on handling IN service calls. The physical interface to the SSPs is G.703 digital trunk. It communicates with the IP via the requesting SSP for connecting specialized resources.
SCP stores large amounts of data concerning the network, service logic, and the IN customers. For this, secondary storage and I/O devices are supported. For more details refer to the chapter on the “SCP Architecture”.
As has been commented before, the service programs and the data at the SCP are updated from the SMP.
Service Management Point (SMP)
The SMP, which is a computer system, is the front-end to the SCP and provides the user interface. It is sometimes referred to as the Service Management System (SMS). It updates the SCP with new data and programs (service logic) and collects statistics from it. The SMP also enables the service subscriber to control his own service parameters via a remote terminal connected through dial-up connection or X.25 PSPDN. This modification is filtered or validated by the network operator before replicating it on the SCP.
The SMP may contain the service creation environment as well. In that case, the new services are created and validated first on the SMP before downloading to the SCP.
One SMP may be used to manage more than one SCPs.
Intelligent Peripheral (IP)
The IP provides enhanced services to all the SSPs in an IN under the control of the SCP. It is centralized since it is more economical for several users to share the specialized resources available in the IP which may be too expensive to replicate in all the SSPs. The following are examples of resources that may be provided by an IP:
- Voice response system
- Voice mailboxes
- Speech recognition system
- Text-to-speech converters
- The IP is switch-based or is a specialized computer. It interfaces to the SSPs via ISDN Primary Rate Interface or G.703 interface at which ISUP, INAP, TCAP, SCCP and MTP protocols of the CCS7 protocol stack are defined.
The IN architecture is depicted in Fig.1.
4 Distributed Functional Plane
The functional model of IN contains nine functional entities (FE’s) which are distributed over various physical entities (PE’s) described in the previous section. A functional entity is a set of unique functions. Brief description of the FE’s is given below :
Call Control Agent Function, gives users access to the network.
Call Control Function provides the basic facility for connecting the transport (e.g. speech). It involves the basic switching function and trigger function for handling the criteria relating to the use of IN (Intelligent Network).
Service Switching Function is used to switch calls based on the advice of the SCF at the SCP. This function provides a service independent interface.
It contains the service logic components and advises the SSF at SSP on further call handling.
Service Data Function contains the user-related data and data internal to the network.
Specialized Resources Function covers all types of specialized resources other than the connection resources that are in the exchange (e.g. recorded announcements, tones, conference bridges, etc.).
Service Creation Environment Function specifies, develops, tests and deploys the services on the network.
Service Management Access Function provides an interface between service management function and the service manager who may be an operator.
Service Management Function enables a service to be deployed and used on IN. Fig. 2 depicts the distribution and interconnection of the various functional entities.
The distribution of functional entities over the physical entities and their inter-connection is summarized in Tables 1 and 2 below. It may be noted that all the physical entities may not be present in all INs as the choice of functional entities to be provisioned is entirely up to the service provider.
Table 1 Distribution of FE’s over PE’s
|Physical Entity||Possible Functional Entities|
|SSP||CCF, SSF, CCAF|
|SMP||SCEF, SMF, SMAF|
Table 2 FE-FE Relationship to PE-PE Relationship
|SSF-SCF||SSP-SCP||INAP, TCAP, SCCP, and MTP|
|SCF-SDF||SCP-SDP||X.25 or Proprietary|
|INAP, TCAP, SCCP and MTP
ISUP, INAP, TCAP, SCCP and MTP
|SRF-SSF||SSP-IP||ISUP and MTP|
5 IN Services
The IN services proposed to be introduced in the Indian network have been derived from ITU-T recommendations. Q.1211 (April ’92). This document briefly gives the description of 25 services mentioned in Capability set no. 1 (CS1) of above mentioned ITU-T recommendations. CS1 basically deals with single-ended services (which ITU-T calls as Type-A services). Single needed services apply to only one party in the call.
1) ABD – Abbreviated dialing
The subscriber can register a short dialing code and use the same for access to any PSTN Number.
2) ACC – Account Card Calling
- A special telephone instrument is required.
- The user dials an access code and gets an acceptance tone.
- Then he dials a PIN (personal identification no.) code and dials the called no. The Exchange reads the account number from the card.
- The Billing is debited to an account number (Telephone no.) as defined by the card.
- In another variation of the service, the account number can be given through a DTMF telephone instrument.
- The follow-on feature facilitates the subscriber to dial another number without disconnecting the call and without need to dial PIN and account number again.
3) AAB – Automatic Alternative Billing
- The call can be initiated by any user and any instrument.
- The call charges are billed in the user’s account and that account need not be a calling or a called party.
- The user first dials access code.
- Receives an announcement to dial the account code and PIN (which is given by management).
- The account code and PIN are validated to check its correctness and expired credit limit.
- On getting acceptance tone the user dials the called number.
- In another variation of the service, the called party may be billed based on his concurrence.
4) CD – Call Distribution
- This service allows subscribers to have I/C calls routed to different destinations according to allocation law specified by management (The Subscriber has multiple installations).
- Three types of laws exist :
- Uniform load distribution
- % Load distribution
- Priority list distribution
- In case of congestion or fault, the alternative overflow is specified.
5) CFU – Call Forwarding Unconditional
The subscriber can forward all incoming calls to a specified destination number. Optionally an alerting ring/reminder ring can be given to the forwarding subscriber whenever there is an incoming call.
6) CRD – Call Rerouting Distribution
- Calls are rerouted as per conditions encountered, e.g. busy or no reply (time specified) or overload or call limiter.
- Then as per selected condition, the call is rerouted to predefined choice, e.g. paper, vocal box, announcement or queue.
7) Completion of calls to busy subscriber
The service cannot be fully implemented with CSI capability since the status of called party need to be known.
- The calls are completed when subscriber who is busy becomes free.
- On getting a busy tone – the user dials a code.
- The user disconnects.
- On called party becoming free, a call is made by the exchange first to originating then to terminating subscriber (without any call attempt by the user).
8) CON – Conference Calling
The service cannot be fully implemented with CSI capability. In adding or dropping the parties concerned it is not possible to check the authenticity of the parties. This service requires a special transmission bridge to allow conversation among multiple subscribers.
- The user reserves the CON resources in advance indicating the date, time of conference and duration.
- Controlled by the user.
- In the active phase of the conference, parties can be added, deleted, isolated again reattached or split the group of parties.
- CON-Meet-ME – Conference calling meet me
- Users reserve the resource the same as 8A.
- Each participant dials a special number at a specific time (specified at the time of booking of the conference) and reaches the conference bridge.
9) CCC – Credit Card Calling
- The Credit Card Calling service allows subscribers to place calls from any normal access interface to any destination number and have the cost of these calls charged to account specified by the CCC number.
- A special instrument is not required. The caller has to dial card number and PIN using the DTMF instrument.
- A follow-on feature may be provided optionally.
10) DCR – Destination Call Routing
The call is routed to the destination pertaining to the following conditions :
- Time of day, day of week
- Area of call originating
- Calling identity of the customer
- Services attributes (nonpayment charges against subscriber)
- Charge rates applicable for destination
- Proportional routing of traffic
- Optionally the subscribers can be provided with traffic details
11) FMD – Follow me Diversion
- A subscriber can remotely control the call forwarding capabilities.
- It can be done from any point in the network using a password.
- It is required if the subscriber moves from place to place in a day.
- The service subscriber will pay for a diverted portion of the call.
12) FPH – Free Phone
- The called subscriber is charged for an active phase of a call.
- For the calling user, no charging is done.
- The called subscriber can have multiple destinations and have a DCR facility.
13) MCI – Malicious Call Indication
- The subscriber requests the Administration to register his number for MCI.
- The administration registers the subscriber for MCI.
- The called subscriber (who has registered this service) invokes the service during the active phase of the call if he feels that the call is malicious.
- The call is logged in the network with calling and called party number and Date and time of invoking the service.
- Optionally, the network can log unanswered calls also.
- Optionally, the facility to HOLD the connection may be provided.
14) MAS – Mass Calling
- It involves a high volume of traffic.
- Calls can be routed to one or multiple destinations depending on geographical location or time of day.
- Mainly used in Televoting.
- The network operator allots a service number.
- The user dials this number to register his vote.
- The user is played an announcement and asked to give his choice.
- At the end of the service, the network operator provides the call details and the count on various preferences.
- After the service, the same number can be reallocated to another subscriber.
- Calls made to this MAS number may be charged differently.
15) OCS – Originating Call Screening
- This helps the subscriber to screen outgoing calls as per day and time.
- The screening list may be managed by the subscriber.
- The restriction of the screening list may be overridden by PIN or password. Three call cases are possible :
- Call screened and allowed
- Call screened and rejected
- Call passed by using override option
16) PRM – Premium Rate
- The local call is charged at a higher (premium) rate.
- This service is used by service providers for value added information services, e.g. jobs, fortune, forecast, etc.
- The revenue is shared between the network operators and service providers.
- The network operator allots a specific number to the service providers, which can be reached from any point in the network.
- The provision exists for multiple site providers, in order to achieve minimum expenditure on the actual call.
17) SEC – Security Screening
- This capability allows security screening to be performed in the network before an end user gains access to subscriber’s network, systems or applications.
- It detects the invalid access attempts: how many, over what time period, by whom and from where.
- It provides an added layer of security.
18) SCF – Selected Call Forwarding (Busy/Don’t answer)
- This facility is used for a group of 5 to 10 subscribers.
- A list of SCF is prepared by a subscriber.
- The list contains the choices as per conditions and calling subscribers of the group.
- A call from outside the group is forwarded to the default telephone number.
- The variation in the SCF list can be done as per the time of the day.
19) SPL – Split Charging
- It allows service subscribers to share the call charges with calling parties on per call basis.
20) VOT – Televoting
- It is used to survey public opinion by different agencies.
- The network operator allocates a single telephone number to the surveyor.
- Each time the user makes a call he can get access to televoting.
- An announcement asks him to input further choice digits as per preference.
- As the user presses the digits the choice counter is incremented.
- After voting is ceased the service subscriber is supplied with the results.
21) TCS – Terminating Call Screening
- The incoming calls are screened as per the screening list.
- Calls are allowed as per the list and time of the day.
22) UAN – Universal Access Number
- The national number is published by the subscriber.
- The subscriber may specify the incoming calls to be routed to a number of different destinations based on geographical locations of the caller.
23) UPT – Universal Personal Telecommunications
- A universal number is defined.
- Whenever a subscriber changes the destination, he inputs that number from the telephone.
- When a call comes, UPT number is translated to an actual number.
- This number can be accessed across various multiple networks, e.g. mobile and fixed.
- It can be accessed from any user network access.
24) UDR – User Defined Routing Intelligent Network
- The user is allowed to define the routing of outgoing calls through a different network such as a private, public, virtual or mixed network.
- As per the time of the day, for example, the call is routed to either public or private network whichever is cheaper.
- For example, outstation calls can have different routes at different times of the day.
25) VPN – Virtual Private Network
- A private network is built using public network resources.
- A virtual PABX is created using different switches.
- A PNP (private numbering plan) can be incorporated on those numbers.
- Facilities such as CT, CH, dialed restrictions and other supplementary services can be provided within the network.
- Each line or user is assigned a class of service and specific rights in the network.
- To access the VPN from outside by one of VPN user, he is required to dial a password.
- Screening feature can be used to put a restriction on outgoing and incoming calls.
- Call charges are assigned to the VPN service subscriber.
- Additional Account Codes are assigned to service subscribers to analyze the cost line wise.
The IN services can be broadly divided into three categories for charging purposes :
– No charging for calling user
– Charging of calling user as per the local call
– Charging of calling the user at higher rates
No charging for calling user: FPH, VCC, and VPN services fall under this category. Level ‘160’ is free at present and is proposed to be allotted to such services. Local exchanges need to analyze only ‘160’ and route the call to SSP. This level has to be created as charge-free. New services of this type can be introduced in the future without any requirement of further modification in local exchanges
Charging of calling user as per local call: UN (local) falls under this category. Level ‘190’ is free at present and is proposed to be allotted to such services. Local exchanges need to analyze only ‘190’ and route the call to SSP. This level has to be created as a local charge. New services of this type can be introduced in the future without any requirement of further modification in local exchanges.
Charging of calling the user at higher rates: PRM and UN (long-distance) falls under this category. Since the charging is at a higher rate it is proposed that prefix ‘0’ may be used to have a barring facility. Level ‘090’ may be used for such purpose. The local exchange will analyze ‘090’ and route the call to SSP. This level has to be created as ‘charge on junction pulses’. New services of this type can be introduced in the future without any requirement of further modification in local exchanges.
The access code of various IN services as proposed is as follows :
No charging for calling user :
Password change for VCC 1602
Charging of calling user as per local call :
UN (local) 1901
Charging of calling the user at higher rates :
UN (Long distance) 0901