The Research On IDS Immune Mode And Gene Algorithm

Computer security becomes more and more important with the increasing frequency of computer-related security problems. IDS (Intrusion Detection System) can identify an intrusion before it causes damage. IDS is the key for a P~2DR (Policy, Protection, Detection, Response) security model. This innovative IDS immune model aims to Improve computer security.Current research on the mechanics of biological immune systems is incomplete. Yet compared with our knowledge of biological immune response, research regarding computer immunology is in its earliest phases. Although there are many differences between living organisms and computers, the similarities are compelling, and may potentially point the way to improvements in computer security.By considering the biological immune system, the author of this paper presents an innovation in IDS modeling.In addition, the author of this paper considers the generation and evolution of a gene bank through the use of a genetic algorithm, augmented by a simulated annealing algorithm.The topic of this paper is a problem of great scientific concern, because currently IDS is all based on pattern recognition, and thus cannot perform anomaly detection. The development of an IDS immune model would enable IDS to not only find previously identified intrusion but also those that have not yet been found. In the current IDS model there exist several problems, including multi-layered protection, flexibility, validity, etc. This new IDS model would correct these characteristics. The new model comprises four different gene banks, and several agencies.The four gene banks are: User Genus Bank; System Genus Bank, Course Genus Bank and Network Genus Bank. Every gene bank could generate and evolve its own gene bank. Each gene bank independently generates an agency that has liquidity and adaptability. Each agency is comprised of detectors. At first, immature agency is generated by a set of detectors combined at random within each gene bank.An immature agency becomes mature by applying the negative-selection algorithm.This model uses the genetic algorithm augmented by the simulated annealing algorithm to generate the initial gene bank, and evolves the gene bank with the return of the highly adaptive agency.The paper under discussion provides the algorithm in full detail. This model sets several adjustable parameters to control false alarms.To achieve flexibility, the four gene...