计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (10): 115-120.DOI: 10.3778/j.issn.1002-8331.1802-0218

• 网络、通信与安全 • 上一篇    下一篇

基于IQPSO-IDE算法的网络入侵检测方法

马占飞1,杨  晋2,金  溢2,边  琦3   

  1. 1.内蒙古科技大学 包头师范学院,内蒙古 包头 014030
    2.内蒙古科技大学 信息工程学院,内蒙古 包头 014010
    3.内蒙古师范大学 传媒学院,呼和浩特 010022
  • 出版日期:2019-05-15 发布日期:2019-05-13

Network Intrusion Detection Method Based on IQPSO-IDE Algorithm

MA Zhanfei1, YANG Jin2, JIN Yi2, BIAN Qi3   

  1. 1.Baotou Teachers College, Inner Mongolia University of Science and Technology, Baotou, Inner Mongolia 014030, China
    2.School of Information Engineering, Inner Mongolia University of Science and Technology, Baotou, Inner Mongolia 014010, China
    3.Vocational Skills Training Department, Inner Mongolia Normal University, Huhhot 010022, China
  • Online:2019-05-15 Published:2019-05-13

摘要: 为了提高网络入侵检测的准确性与检测效率,弥补由单一优化算法带来的计算精度低、易陷入局部极值等不足,将差分算法的思想引入量子粒子群算法中,提出了一种改进量子粒子群算法(Improved Quantum Particle Swarm Optimization algorithm,IQPSO)和改进差分算法(Improved Difference Evolution,IDE)相融合的IQPSO-IDE算法,并将IQPSO-IDE算法对支持向量机(Support Vector Machine,SVM)的参数进行优化。以此为基础,设计了一种基于IQPSO-IDE算法的网络入侵检测方法。实验结果表明,IQPSO-IDE算法与传统的QPSO、GA-DE、QPSO-DE算法相比,不仅在效率上有了明显的改善,而且在网络入侵检测的正确率上分别提高了5.12%、3.05%、2.26%,在误报率上分别降低了3.31%、1.54%、0.93%,在漏报率上分别降低了1.26%、0.73%、0.52%。

关键词: 网络安全, 入侵检测, 量子粒子群算法, 差分算法, 支持向量机

Abstract: In order to improve the testing efficiency and accuracy of network intrusion detection, and make up for the disadvantages of low computing precision and easy to get into local extremum caused by a single optimization algorithm, this paper introduces the idea of the difference algorithm into the quantum particle swarm algorithm, and proposes an IQPSO-IDE algorithm based on the Improved Quantum Particle Swarm Optimization algorithm(IQPSO) and the Improved Difference Evolution algorithm(IDE). In addition, the IQPSO-IDE algorithm also optimizes the parameters of support vector machines. Based on this, this paper designs a network intrusion detection method based on IQPSO-IDE algorithm. The experimental results show that the efficiency of the IQPSO-IDE algorithm is better than traditional QPSO algorithm, GA-DE algorithm, QPSO-DE algorithm, and the accuracy of network intrusion detection of this algorithm is increased by 5.12%, 3.05% and 2.26%, the rate of false positives is reduced by 3.31%, 1.54% and 0.93%, the non-response rate is decreased by 1.26%, 0.73% and 0.52%.

Key words: network security, intrusion detection, quantum particle swarm optimization, differential evolution, support vector machine