基于改进的BP神经网络入侵检测方法研究

Research on intrusion detection method based on improved BP neural network

针对遗传算法局部搜索能力弱和收敛速度慢,在选择操作之后加上了禁忌搜索算法,并对交叉操作进行改进,最后用禁忌搜索作为变异操作,从而加快算法的收敛速度,并用此改进的遗传算法来优化BP神经网络的权值。实验证明,采用该方法优化BP神经网络权值,能克服BP神经网络收敛速度慢、局部极小问题。

For the problems that genetic algorithms local search ability is weak and convergence speed is slow , the paper adds tabu search algorithm after the selection operation, and the crossover operation is improved. Finally, it takes tabu search as the mutation operation to speed up the convergence rate. It uses this improved genetic algorithm to optimize BP neural network weights. Experiments show that optimizing BP neural network weights with the method proposed in this paper can overcome the shortages of slow convergence and local minima.