入侵检测系统是当前信息安全领域的研究热点,在保障信息安全方面起着重要的作用。在对BP神经网络优化算法进行对比研究的基础上,利用Levenberg-Marquardt算法对传统BP算法进行改进,成功地将LMBP算法运用到入侵检测中去。实验结果表明,运用Levenberg-Marquardt算法优化BP神经网络进行入侵检测,可以较好地提高学习速率,缩短收敛时间。 Intrusion detection system is the hotspot in the field of information security and plays an important role in ensuring information security. Based on the comparison of BP neural network optimization algorithms, the Levenberg-Marquardt algorithm is used to improve the traditional BP algorithm, and the LMBP algorithm is successfully applied to intrusion detection. Experimental results show that using Levenberg-Marquardt algorithm to optimize BP neural network for intrusion detection can improve the learning rate and shorten the convergence time.