The Research And Applications Of Intrusion Detection System Based On The Immune Mechanism

The issue comes from the Shanghai software company Sybase’s e-banking project,which is a web-based inter-bank program include transfer, bill pay, wire and so on. Itrelates to the large amount of sensitive data so network security has become animportant issue. With the rapid growth of network, the growing ranks of Internet usersand the network intrusion incidents are also increasing. Firewalls and other defensivetechnology have been difficult to stop the illegal invasion and can not effectively solvethe problem of network security. Therefore, the network security technologies and newmethods of research have drawn increasing attention.Intrusion detection system can detect intrusions by collecting and anglicizingintrusions and normal behavior of the difference. It found great majority of intrusionsby appropriate security measures, such as alarm, record, cut or blocked, etc. Therebyenhancing the resilience of the network security is more necessary. Intrusion detectionsystem as an information security safeguards, so that many scholars and researchers athome and abroad to start the research on intrusion detection systems and made a lot ofresearch results.The paper has e-banking system as project background, to carry out research onintrusion detection technology. The existing intrusion detection systems at home andabroad have a brief comparative analysis of the existing intrusion detection systemdeficiencies. Then analyze the principle of biological immune system and itscomponents. The immune system and intrusion detection process which has a similarlink between the two is presented based on the mechanism of immune-based intrusiondetection model proposed by establish mechanisms for intrusion detection, the modelestablished in the snort rule base and inherent identifier, research and application of theimmune system to generate systems for intrusion detection.In this paper, the research is focused on the following two aspects: First, based onthe immune biological characteristics of multi-layer defense, design the intrusiondetection model comprised of inherent detection and adaptive composition. Invasion isdetecting violation of security policy and signs of attack by analyzing the informationfrom the critical points in the computer network or computer system. On the base ofin-depth analysis of Invasive nature, the paper studies matching algorithm and element detection algorithm, simplifies and optimizes Snort rules, improves the widely usednegative selection algorithm, and designs an advanced element detection method.Second, in the study of detection sub-algorithm, the paper analyzes the advantages anddisadvantages of the traditional detection sub-algorithm, applies random generationalgorithm of detection element and two times tolerance and mutation mechanism withthe random generation algorithm, and then proposes a new random generation algorithmof detection element. The improved algorithm is applied to intrusion detection systemunder study, implemented in the Windows environment using the Java programminglanguage, and tested in a large number of data to prove that this random generationalgorithm is effective and viable in intrusion detection system.