An Intrusion Prediction Technique Based on Co-evolutionary Immune System for Network Security (CoCo-IDP)

Mohammad Reza Ahmadi

《International Journal of Network Security》 9卷3期 (2009/11) Pp. 290-300

https://doi.org/10.6633/IJNS.200911.9(3).11

並列摘要

Forecasting the unknown and detecting the known threats1 and targeted attacks^2 are the most concern of network security especially in large scale environment. We have presented an intrusion^3 detection and prediction system using cooperative co-evolutionary immune system for distributed data networks. This is an intelligent technique based on genetic algorithm and co-evolutionary immune system where the detectors can discriminate the existing incidents^4 and predicting the new incidents in a distributed environment. We have prepared a prototype of CoCo-IDP^5 in a Jini platform running grid computing^6 in distributed systems. Evaluation results show that, the CoCo-IDP can adaptively converge for the best answer and can detect or predict the incidents in a selected boundary. Moreover, the system generates the flexible detectors with diversity in a variable threshold. In comparison with pure Immune System (IS), the obtained results show that the proposed system has simpler rules, more powerful detection and prediction capabilities with high accuracy metric. We have compared the probability of detection and false accuracy rate in KDD^7 database with several well known methods for proof and validation of our results.

並列關鍵字

Cooperative co-evolutionary algorithm ； grid computing ； intrusion prediction system