网络安全态势预测方法的仿真研究

摘要

研究网络安全问题,网络安全是一种复杂的非线性系统,具有显著的时变性特点,传统预测算法不能够准确刻画网络安全复杂变化特点,预测精度低.为提高网络安全态势预测精度,提出一种遗传算法和支持向量机相结合的网络安全态势预测模型.首先将网络安全态势数据重构成多维时间序列,然后输入到支持向量机进行训练,并通过遗传算法解决训练模型参数优化难题,从而建立网络安全态势预测模型,最后采用预测模型对未来时刻网络安全态势进行预测.仿真结果表明,遗传算法和SVM结合的模型可以很好地反映网络的整体安全运行状况,提高了网络安全态势预测精度,可以指导管理员对网络安全进行控制.%Study the problem of network security. Network security is of the characteristics of nonlinear, time-va-rying , thus the traditional prediction algorithm is unable to accurately predict the network security situation. In order to improve the prediction accuracy of network security situation, we put forward a network security situation forecast model based on the genetic algorithm combined with support vector machine. The network security situation data con-stituted the multi-dimensional time series, and then were input to a support vector machine for training. The genetic algorithm was used to solve optimization problems of model parameters. Finally, the prediction model was used for network security situation forecast of future time. The simulation results show that, the algorithm can well reflect the overall network safe operation condition, improve the network security situation prediction accuracy, and guide the administrator of the network security control.