Intrusion Detection System in Wireless IoT Network using feature selection and Bi-LSTM

Currently, the Internet of Things has grown widely and has provided new possibilities for applications in various fields. However, because of its dynamic architecture and constrained device resources, the Internet of Things presents numerous security challenges. Messages can be transported between wireless sensor network nodes at the network layer thanks to the Routing Protocol for Low Power Network (RPL). Various methods have been presented to identify RPL network assaults in recent years due to the sensitivity of the RPL routing protocol. In this research, an approach to detect IoT attacks based on the RPL protocol with the help of machine learning methods, principal component analysis (PCA) and neighborhood component analysis (NCA) for feature selection and classification based on Bi-LSTM neural network is presented. The results of the simulation and performance evaluation of the proposed method in the form of Matlab programming language show that the proposed approach has been able to provide a much better performance than the compared method.