Network Intrusion Detection Systems Based On Neural Networks

Nowdays, internet and computer has become more and more extensive used, Corporations, governments and other organizations have also depend on computer network system increasingly. Therefore, security problem become more out standing. As an active security-defence technique, intrusion-detection offers realtime protect against interior or exterior attack, and mistaken operation ,as it can hold up an respond intrusion before the network is endangered. However, with the diversity of intrusion technology, tradition IDS can not fit current network environment ,therefor intelligent technologies must be used.This paper open with some elemental conceptions and theories of IDS. The paper analyzes intrusion-detection technique in existing IDS models and IDS products, discovers they are limited and hard to meet IDS 's needs which occupies real-time character, adaptability, accuracy and the ability of self-learning.Then study upon on neural network, the paper finds it is very suitable for the IDS in concept. An intrusion-detection system based on neural network will play a much role in the theory and practical if it can be designed and implemented. And the paper gives a detailed describing to the deducing of BP algorithm and its betterment arithmetic.This paper introduces the neural network technology in IDS model, and put forward a detailed design scheme of intrusion-detection model based on neural network. Great emphasis was put in key modules. Lastly according experimental through training and intrusion procedure, we get a fairly analysis, which indicates the neural network has a very great advantage in intrusion detection.