Misuse Detection System

The target system contributes to the reduction of the time required for the reaction to misuse and for the detection of new misuses, not only by employing the sophisticated AI techniques, but also by automation of related problem handling processes.

The following areas of network management are supported by the prototype developed in the project:

• Fault management
• Fraud management
• Performance management
• Business procedures

The developed prototype will exploit the cutting-edge AI-based techniques, mainly based on the nature-based concepts, such as neural networks, genetic evolution, immune systems principles and other technologies like fuzzy logic, Case Based Reasoning etc.

The prototype will be further integrated in the commercial products produced by the project partners.

Complexity of rules and difficult maintenance of rule sets caused the development of the AI-based environment for the misuse detection in mobile telecommunication network. AI-empowered modules will be deployed in the areas of fault and performance management and fraud detection.

Two very important factors that come into play must be considered: the speed with which the correlation occurs and the accuracy of data returned by the correlation. Obviously, even an infinitely fast system that returns incorrect information is just as bad as a system which produces a perfect diagnosis every time, but only does with such delay from the original fact, that it is of no use in diagnosing and fixing the problem. There is a need for a system which provides on-line or near on-line misuse analysis with the high accuracy.