网络信息加密漏洞实时检测仿真研究

Simulation Research on Real Time Detection of Network Information Encryption Vulnerability

对网络信息加密过程中漏洞的检测,能够进一步实现网络信息安全保护。对信息加密过程漏洞的实时检测,需要寻找最佳特征子集以及KNN参数,利用人工鱼方法对漏洞位置精度不断寻优,完成漏洞实时检测。传统方法组建了状态转移矩阵,构建初始概率分布隐马尔科夫状态模型,但忽略了对漏洞位置精度的评估,导致所检测到的漏洞位置并不准确。提出基于KNN和粒子群的漏洞实时检测方法。建立网络漏洞检测模型,利用粒子间的信息交流,及相互协作,寻找最佳特征子集和KNN参数,完成最优网络漏洞实时检测模型的建立,以此提高检测准确性,减少漏检情况;通过人工鱼方法对漏洞位置进一步定位,并不断对当前的最优解进行更新,直到找到网络漏洞的准确位置,完成信息加密中漏洞的精准定位。实验表明,上述方法可有效提高网络漏洞检测和定位的准确性,降低了漏检率。

For real-time detection of vulnerabilities in information encryption, it is necessary to find best feature subset and KNN parameters. The traditional method ignores the accuracy evaluation of vulnerability location. There- fore, the detection of the vulnerability location is not accurate. A real-time detection method for vulnerability based on KNN and particle swarm is proposed. Firstly, the model of network vulnerability detection is built. Then, informa- tion exchange between particles and mutual cooperation is used to find the best feature subset and KNN parameters, so that the establishment of optimal real-time detection model of network vulnerability is completed, thus to improve the detection accuracy and reduce missed detection. Through the artificial fish method, the vulnerability position is determined further, the current optimal solution is updated continuously until the accurate position of network vulnerability is found. Thus, the precise positioning of vulnerability in information encryption is completed. Simulation results prove that this method can effectively improve the accuracy of detection and localization of network vulnerability and reduce the missing rate.