Triggering in intelligent networks is fundamental because the original purpose of th IN was to enable call control software to recognise pre-assigned events and conditions and suspend call processing whilst it obtained further call processing instructions from an external service control point. However, when IN was introduced, the software which handles this 'breaking out' process could not simply be added alongside existing call processing functions. It had to be built in as an integral part of the basic switch call control. In order to achieve progress in building the standard recommendations for enhancing telephone networks with IN features, it was necessary to introduce a hypothetical modelling device known as the 'service switching function' (SSF) to the telephone exchanges.

IN was once heralded as a network re-arrangement that would provide any telecommunications service need that any user could conceivably want. In practice, over the years, commercial priorities, financial constraints and inter-working limitations have inevitably limited the vision potential. Whilst in the early days the IN concept had high aspirations, reality had to intervene sooner or later to remind us what might be feasible and what might not be. For example, call routing decisions based on the time of day are perfectly feasible, whereas a routing decision based on the colour of a users shirt may be less feasible! As already discussed, AT&T and then Bellcore, in the USA, made the running on early exploratory work on IN specifications and the European operators watched the results carefully. In the early 1990s there was a lot of activity in the USA on IN specification production, although these standards being formulated by Bellcore were obviously designed to meet US needs and US switching system requirements, some of which were less relevant outside the US. In the early 1990s, the CCITT (now the ITU-T) set up a study group hierarchy for intelligent network standards, and the first result was the CCITT IN CS-1 (Capability Set 1) in 1993. CS-1 was planned to be the first of several steps towards the introduction of intelligent network concepts into telephony networks. It was seen as a relatively modest set of early IN capabilities for voice calls. It deals with single-ended, single point of control, IN service features.

Up to this point in the book the controlling service logic has been presented as being located (at least logically) at a single point and this has been represented by the IN SCP. The concept of remote data has been introduced with the IN SDP, but it has generally been assumed that the controlling logic that runs the service is hosted by the network operator on a single computing facility. Thus we have service-switching points, and service control points, such that an SSP might communicate with a 'home', or pre-assigned, SCP for its service requests. For many services, such as those described earlier, this model has proved to be sufficient. Early intelligent networking was to do with a set of discrete database nodes serving large numbers of telephone exchanges, with a reliable and intelligent signalling system in between. This ensured that the required message reached the intended destination database processor, and that the reply would be correctly returned to the originating network node.

This book has been primarily concerned with implementations based on the ETSI Core INAP standard because this has become a widely popular variant, and was in fact fed back into the ITU-T to be adopted as a refinement (CS-1R) to the early CS-1 ITU-T IN capability set. For this reason this book has concentrated primarily on the practicalities of the 'here and now' of IN as it is currently operating, rather than indulge in speculation of how evolving IN principles might be implemented in future mainstream networks.

Provides a list of the 26 INAP operations included in ITU-T CS-1, but excluded from ETSI Core INAP and ITU-T CS-1R