Speedy Signature Based Intrusion Detection System Using Finite State Machine and Hashing Techniques

This paper proposes a secure system designs for client-server based communication systems. In this system, security services are implemented on server, as generally data received on the servers contains malicious contents. The technique that we used is to perform speedy intrusive signature matching received inside a network with the known signatures from the training database. Probable intrusive signatures, which get filtered from hash value matching, are exposed to a finite state model that inspects those signatures against a finite automaton. Other systems like anomaly based detection may not detect all malicious activity signatures. Also, we have taken a note to reduce the false positive rate to nil while implementing the system which gets generated in other detection systems during the communication process. The proposed system works on a client-server based model.