STUDY ON MULTI-SENSOR DATA FUSIION AND APPLY IT TO THE IMPLEMENTATION OF AN INTRUSION DETECTION (ID) SYSTEM PROTOTYPE
BY
This project addresses the lack of efficiency of current ID system to detect sophisticated user attacks and to introduce multi-sensor data fusion as a functional framework for building next generation ID system. This project investigates and classifies the common cyber attackers and the result will be the base of developing intrusion detection template and intrusion pattern. The project also investigates data mining and heuristic techniques by which new intrusion pattern can be discovered and many intrusion attributes can be inferred from. The project comes with an ID system prototype which is capable of revealing a relative broad range of cyber intrusions and has low false alarm rate
Full text source:
December 2003