Flow Based Intrusion Detection and Prevention By Adaptive Network Learning

With the emergence of global connectivity with expansion of computer networks during the past decade, security threats in network have become a crucial issue for computer systems. Nowadays, it is very important to retain a high level security to ensure safe and trusted communication for information exchange across the network. Different softcomputing based methods and tools have been proposed in recent years for the development of intrusion detection systems on host based and host independent. There are various approaches being utilized in intrusion detections but unfortunately any of the systems so far is not completely flawless .This paper presents a Flow-based anomaly detector for intrusion detection in network in host independent by self learning process. This will handle the network flow and attack on network traffic in a fully automatic and unsupervised fashion. This is host independent and is conditioned on the flow of network rather than payload length. In this approach, the flow of data through the network is analyze instead of the contents of each individual packet. This model provides a classification of attacks and defense mechanism techniques to avoid intrusion.