基于改进LSTM的网络入侵检测方法

Improve the network intrusion detection method of LSTM

摘要:

随着网络数据的增加，以及黑客技术的不断发展，网络入侵检测技术的精度以及效率需要进一步提升。针对此问题，研究提出一种基于逃避网络数据和改进长短时记忆网络的网络入侵检测模型。该模型将黑客入侵过程中为躲避检测出现的逃避行为数据作为训练集和测试集。之后利用麻雀优化算法改进长短时记忆网络模型。研究将改进后的长短时记忆网络模型和卷积神经网络结合，并通过强化学习进一步提升模型的检测精度。实验结果表明，研究提出的模型的检测准确率达到了98.51%，且响应时间仅为0.84s，漏报率和误报率分别为1.23%，误报率为0.36%。研究提出的模型能够实现高效的网络入侵检测，实时保障网络安全，实现网络入侵防御，为网络安全提供可靠的技术支持。研究设计方法在网络攻防领域具有积极意义，为相关领域研究提供了新的思路。

Abstract:

With the increase of network data and the continuous development of hacker technology, the accuracy and efficiency of network intrusion detection technology need to be further improved. To solve this problem, a network intrusion detection model based on network data evasion and improved long term memory network is proposed. The model takes the evasion behavior data in the process of hacking to avoid detection as a training set and a test set. Then the Sparrow optimization algorithm is used to improve the short-time memory network model. The improved long term memory network model is combined with convolutional neural network, and the detection accuracy of the model is further improved by reinforcement learning. The experimental results show that the detection accuracy of the proposed model is 98.51%, the response time is only 0.84s, and the false positive rate is 1.23% and false positive rate is 0.36%, respectively. The proposed model can realize efficient network intrusion detection, guarantee network security in real time, realize network intrusion prevention, and provide reliable technical support for network security. The research design method has positive significance in the field of network attack and defense, and provides a new idea for the research of related fields.