Intrusion Detection System Research Based On Neural Network

With the development of the Internet, the rich resources on Internet bring users so much convenience as well as problems of security the users must face with at the same time. Nowadays, there are more and more attacking means and the intrusion means are continually updating. The losses resulting from network attacking are inestimable,A lot of secure policies, strategies and tools are analyzed and applied to solve these problems, including IDS (Intrusion Dection System). As an active security-defence technique, intrusion-detection offers realtime protect against interior or exterior attack, and mistaken operation, as it can hold up a respond intrusion before the network is endangered. However, with the diversity of intrusion technology, tradition IDS can't 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, the result of examination is satisfactory. The false positive rate is not great, and the detection rate of new kind intrusion is not too low. So the BP net has wide space of development and foreground of ues in IDS. An analysis of the insufficiency of this system and sums up my study were given in the end of this thesis.The author brought forward the method to improve on this system and pointed out the next task of this system.