基于遗传神经网络的通信网络安全威胁智能评估方法

Intelligent Evaluation of Communication Network Security Threat Based on Genetic Neural Network

摘要:

针对当前的通信网络安全威胁评估方法面临着数据量大、威胁类型多样、环境动态变化等挑战，基于规则和简单统计分析的传统评估方法难以满足实时性、准确性需求的局限性，提出了一种基于遗传神经网络的通信网络安全威胁智能评估方法。通过构建包含通信网络受攻击程度、受攻击后的通信质量和通信容量等方面的通信网络安全评估指标体系，并采用非数值型指标量化、正向化处理、无量纲标准化对评估指标进行规范化处理，设计了基于遗传算法优化的神经网络评估模型，实现对通信网络安全威胁的准确、智能评估。通过TOPSIS方法生成的数据集对所提出的评估方法进行了实验验证，结果显示评估准确率达到了92%，证明了该评估方法的有效性。

Abstract:

Facing the challenges of large data volumes, diverse threat types, and dynamic environmental changes in current communication network security threat assessment methods, traditional assessment methods based on rules and simple statistical analysis struggle to meet the demands for real-time and accuracy. To address these limitations, this paper proposes an intelligent assessment method for communication network security threats based on genetic neural networks. By constructing a communication network security assessment index system that includes aspects such as the degree of network attacks, post-attack communication quality, and communication capacity, and employing non-numerical indicator quantification, positive processing, and dimensionless standardization for normalization of assessment indicators, a neural network assessment model optimized by genetic algorithms is designed to achieve accurate and intelligent assessment of communication network security threats. The proposed assessment method was experimentally validated using a dataset generated by the TOPSIS method, and the results showed an assessment accuracy rate of 92%, proving the effectiveness of the assessment method.