An algorithmic approach for feature interaction detection

State-of-the-art telecommunication switching systems provide an enormous variety of subscriber and network features. This feature set is dramatically increased with every software release as new features are delivered to the highly competitive market of deregulated public telecom networks. This causes the number of features in a switching system to increase enormously. There are hundreds of different features implemented in the existing switching systems and a lot of these features can influence each other. Therefore it is a critical issue for the switching software to control multiple features interacting within one call over the PSTN/ISDN in a robust and flexible way. The paper presents an algorithmic approach for a systematic feature interaction design in a telecommunication switching system. This approach can be used for interaction detection of features already implemented in the PSTN/ISDN as well as for investigation how new features will interact with those already existing. The algorithm can also be used for interaction detection of IN services and their service features with features provided by the PSTN/ISDN. Moreover, it can give some assistance to system designers in the cases where the system behaviour must be defined when new complex features (e.g. Call Party Handling (CPH) or Large Conference (CONF)) are being implemented. The presented approach is based on a structured feature description based on the "Feature Points in Call (FPIC)" concept. This description method represents a feature as an information (parameter) flow through functional units of a network based call model. Consequently, the interaction of features is understood as an interaction of different information flows. This approach is the base for the development of a software tool for detection of feature interaction in telecommunication systems.