Pso Based Intrusion Detection: A Pre-Implementation Discussion

Intrusion Detection Systems (IDS) are commonly used in almost every network. They are facing however challenges regarding, autonomy, scalability and efficiency. The pressing question is how to make the IDS capable of extracting the accurate network behavior while being intelligent enough to detect new intrusions. It is therefore, important to explore new possibilities that could lead to a further improvement in that respect. Our research is an inspiration of two fields, Data mining as a field that provide powerful techniques for learning and knowledge extraction and swarm intelligence a field that include collaborative and stigmergic heuristics for search and detection. As we tried to combine these two methodologies for intrusion detection problem, using Particle Swarm Optimization as a detection heuristic we did come up against many nuances that we tried to explain through this paper. We noticed that the literature has not provided yet guidelines to consider before any implementation. We believe that this work is a step in that direction.