Research Of Intrusion Detection Based On Dynamic Clonal Selection

With the fast development of network technology, computer and network has been widely used and become the most important basic instrument, and network security has become an important issue. The research of IDS inspired by biological immune system is one hot point after neural network and evolving calculation.The shortcoming of current intrusion detection system is unable to detect new types of attacks quickly and exactly. After deeply research on the theory and all kinds of information processing mechanisms in biological immune system, this paper puts forward an intrusion detection model based on dynamic clonal selection algorithm. The module of capturing packet and intrusion detection are accomplished.First, because the rate of losing packets is high in the traditional method, an improved method is put forward in the design of capturing packets. The new method improves the rate of capturing packets by simplifying the data processing mode in the user buffer. Testing result shows that, the new method is very effective at reducing the rate of the losing packet.Secondly, by studying and analyzing dynamic clonal selection algorithm, a large shortcoming of this algorithm, that is, to obtain satisfactory detection probability, a large number of co-stimulation is needed and the performance is cut down. So, an extended algorithm is proposed and is applied to the module of intrusion detection. In the new algorithm, unknown attacks are detected by memory detectors and mature detectors. The memory detectors are optimized according to the fitness to adapt to the current intrusion and keep high detection performance.At last, an examination is test to verify the validity of the model. It is...