光纤通信网络入侵干扰信号定位检测仿真

针对当前检测方法存在干扰信号定位效果较差、检测时间过长,提出了基于粒子群算法的光纤通信网络入侵干扰信号定位检测方法。对网络入侵信号进行分析,并以此构建网络入侵干扰信号采样和信号传输结构模型,结合时间序列分析法,对序列入侵干扰信号进行FIR滤波进行抗干扰滤波处理,根据模型给出的幅值参数,提取入侵干扰信号特征。将提取的入侵干扰信号特征与LSSVM参数编制为二进制粒子,利用网络入侵检测的正确率和特征子集维数权值构造粒子群目标函数,利用粒子群寻找到最优特征子集以及LSSVM参数,并且引入混沌机制保证粒子群的多样性,并在此基础上,构建最优网络入侵检测模型,将K-means算法引入到入侵检测模型中,将网络入侵干扰信号进行定位,实现光纤通信网络入侵干扰信号定位检测。实验结果表明,所提方法有效减少了检测时间,并且减少了干扰信号定位误差,提高定位精度。

Optical Fiber Communication Network Intrusion Interference Signal Location Detection Simulation

In current detection methods, the effect of interference signal location is poor and the detection time is too long. Therefore, a method for detecting and locating the intrusion interference signals in optical fiber communication network was proposed. At first, the network intrusion signal was analyzed and the model of network intrusion interference signal sampling and signal transmission structure was constructed. Combined with the time series analysis method, FIR filter and anti-interference filter were applied to the sequence intrusion interference signal. According to the model, the amplitude parameters were given to extract the characteristic of intrusion interference signal. The extracted intrusion interference signal features and LSSVM parameters were compiled into binary particles. Moreover, the correct rate of network intrusion detection and feature subset dimension weight were used to construct the particle swarm objective function, and then the particle swarm was used to find the optimal feature subset and LSSVM parameters. Furthermore, the chaotic mechanism was introduced to ensure the variousness of particle swarm. On this basis, the optimal network intrusion detection model was built. In addition, the K-means algorithm was introduced into the intrusion detection model to locate the network intrusion interference signal. Finally, the location and detection for intrusion interference signal in optical fiber communication network was achieved. Simulation results prove that the proposed method effectively reduces the detection time and the interference signal positioning error. Meanwhile, this method improves the location accuracy.