Intelligent Differential Evolution based Feature Selection with Deep Neural Network for Intrusion Detection in Wireless Sensor Networks

Abstract

Wireless sensor network (WSN) is mainly utilized for data gathering and surveillance applications. As WSN is majorly deployed in harsh and hostile environments, security remains a critical issue which needs to be resolved. An intrusion detection system (IDS) is one of the proficient ways used to determine the presence of abnormal behaviors (i.e. intrusions) in the network. Earlier studies have focused on the design of machine learning (ML) and deep learning (ML) models to design IDS. With this motivation, this paper presents an intelligent differential evolution based feature selection with deep neural network (IDEFS-DNN) for intrusion detection in WSN. The proposed IDEFS-DNN model aims to select optimum set of features and classify the intrusions in the network. In addition, the IDEFS-DNN technique involves the design of IDEFS technique to choose a subset of optimum features. Moreover, the chosen features are fed into the DNN technique for classification purposes. The usage of IDEFS technique helps to reduce the complexity and increase the classifier outcome. In order to portray the improved performance of the IDEFS-DNN technique, wide ranging experiments take place on benchmark datasets and the results are inspected under varying aspects. The simulation results ensured the enhanced intrusion detection performance of the IDEFS-DNN technique over the other IDS models.