改进PSO优化RBF的网络安全态势预测研究

Research on Network Security Situation Prediction Based on RBF Optimized by Improved PSO

针对网络安全态势预测,为了提高预测精度和预测算法的收敛速度,采用一种改进的粒子群算法(PSO)来优化径向基函数(RBF)神经网络。首先,PSO的惯性权重因子按一条开口向左的抛物线递减,在保证全局寻优的同时又增强了局部搜索能力;其次,通过权重因子的调节自动寻优,并将搜寻到的全局最优值解码成RBF的网络参数;最后,通过优化的RBF网络进行网络安全态势预测。仿真实验表明,改进后的算法能较准确地预测网络安全态势。与BP算法和RBF算法相比,本文算法在预测精度上有所提高,同时收敛速度加快,能达到更好的预测效果。

According to the network security situation prediction, an improved particle swarm optimization（PSO） is used to optimize radial basis function（ RBF） neural network to obtain higher forecasting precision and faster convergence speed. Firstly, the inertia weight factor of PSO method decreases in a parabola manner,whose opening is facing left, it is ensures the good global optimization and enhances the local search ability.Secondly, the inertia weight is automatically optimized by its own adjustment, and then the final global optimization is decoded into the network parameters of RBF. Finally, the optimized RBF network is used to predict the network security situation. The simulation experiments indicate that the improved method can accurately predict the network security situation. Compared with the BP method and RBF method, this method obtains better prediction results with higher prediction precision and faster convergence speed.