一种基于粒子群的入侵检测研究

A study of Intrusion Detection Based on particle swarm optimization

当今攻击网络的手段是多种多样的,为保护网络的用户不受来自网络的攻击,网络在使用中需要安全设备和安全技术。入侵检测技术是一种安全检测技术,该技术能够来阻止网络攻击行为。但要阻止网络的攻击行为,必须检测到该行为。本文在简述了入侵检测技术,粒子群知识后,然后提出了粒子群在入侵检测技术上的应用。该技术在入侵检测上的应用将使得检测方法具有一定的智能性,将粒子群技术应用到入侵检测中属于是首次。本文提出的具有一定智能性检测算法可分为两个步骤:①首先通过函数?xfy)(判断链路中的数据流是否在正常范围内,还是属于异常。②然后如果某种数据流属于异常的流,则使用粒子群算法来对未知属性数据流的属性进行定性判断。本文提出的算法具有一定的智能性,能够作为现有的入侵检测算法的补充。

in today＇s attack means is varied, for the protection of network users do not suffer from network attacks, network safety equipment and technology in use. Intrusion detection technology is a security detection technology, this technology can prevent network attacks. But to prevent network attacks, we must detect the behavior. In this paper, the technology of intrusion detection, particle swarm optimization knowledge, and then proposes the application of particle swarm in intrusion detection technology. The application of technology in the intrusion detection will make detection method has certain intelligence, particle swarm techniques applied to intrusion detection belongs is the first. This paper presents an intelligent detection algorithms can be divided into two steps ： firstly, the function to judge the link in the data stream is in the normal range, belonging to the abnormal. Secondly if a data flow belongs to the abnormal flow, qualitative judgment attributes using particle swarm algorithm to unknown attribute data stream. The algorithm proposed in this paper has certain intelligence, could be used as a complement to existing intrusion detection algorithm.