基于BILSTM与改进PSO算法的农业物联网入侵行为检测

Intrusion Detection Method of Agriculture IoT Based on Bidirectional LSTM and Improved PSO Algorithm

为了提升农业物联网的安全性,针对农业物联网的入侵行为特点,设计了基于BILSTM与改进粒子群优化算法的入侵检测方法。提出一种改进粒子群优化算法对BILSTM模型参数进行优化的算法,采用优化后的BILSTM模型对农业物联网网络入侵行为进行时间特征提取,使用Sofmax分类器进行分类。通过模拟实验,测试了农业物联网入侵行为检测方法的实际效果。实验表明,在入侵行为种类增加的情况下,该方法相比传统方法具有更高的检测率和更低的误检率,能够更加有效地针对农业物联网的入侵行为进行检测。

To improve the security of agriculture Internet of Things, an intrusion detection method based on bidirectional LSTM and improved particle swarm optimization algorithm is designed according to the intrusion behavior characteristics of agriculture Internet of things. An improved particle swarm optimization algorithm is proposed to optimize the parameters of the bidirectional LSTM model. The bidirectional long-term and short-term memory neural network is used to extract the temporal characteristics of the intrusion behavior, and the sofmax classifier is used for classification. Through the simulation intrusion experiment, the detection effect of agriculture Internet of things intrusion detection method is tested. The experiment shows that the method in this paper has higher detection rate and lower false detection rate than traditional methods when the types of intrusion behaviors increase, and it can detect the intrusion behaviors of the agricultural Internet of Things more effectively.