基于大数据技术的光纤通信安全性防御方法研究

Research on security defense method of optical fiber communication based on big data technology

针对当前光纤通信攻击信号检测误报率较高,光纤通信安全性防御成功率下降的问题,提出基于大数据技术的光纤通信安全性防御方法。首先应用大数据技术中的贝叶斯理论,优化光纤通信异常信号采集与识别环节。接着使用支持向量机,对采集到的原始异常信号进行分类,提取攻击信号并确定攻击类型。最后构建防御时间序列,对得到的攻击信息进行防御,实现通信安全性防御,完成光纤通信安全性防御方法设计。实验结果表明:此方法的误报率均在1.0%以内;光纤通信网络敏感区域标识准确性高,且标识效果较为稳定,用时较短在3 s左右;在安全防御中的应用效果较为稳定,整体波动较小且成功率均在98%以上。可见此方法能有效降低光纤通信攻击信号检测误报率及漏报率,提升光纤通信安全性防御成功率。

In view of the current optical fiber communication security defense methods,the classification accuracy of interference signals is low,which leads to the high false alarm rate of optical fiber communication attack signal de-tection,resulting in the decline of the success rate of optical fiber communication security defense.Bayesian theory in big data technology is applied to optimize the acquisition and identification of abnormal signals in optical fiber commu-nication.Support vector machine is used to classify the collected original abnormal signals,extract attack signals and determine the type of attack.In this study,a defense time series is constructed to defend the obtained attack informa-tion and achieve communication security defense.So far,the design of optical fiber communication security defense method based on big data technology has been completed.The experimental results show that the false positive rate of this method is less than 1.0%.The identification accuracy of sensitive areas of optical fiber communication network is high,and the identification effect is relatively stable and the time is short,about 3 minutes.The application effect in security defense is relatively stable,the overall fluctuation is small,and the success rate is more than 98%.It can be seen that this method can effectively reduce the false positive rate and false negative rate of optical fiber communication attack signal detection,and improve the success rate of optical fiber communication security defense.