An intrusion detection system for wireless sensor networks using deep neural network

Wireless sensor network comprises of a large number of sensor nodes to acquire and transmit data to the central location. However, due to resource constrained nodes, deployment strategies and communication channel introduce numerous security challenges to the wireless sensor networks. So, it is essential to detect unauthorized access to improve the security features of wireless sensor networks. Network intrusion detection systems provide such services to the network and it becomes inevitable for any communication network. Machine learning (ML) techniques are widely used in intrusion detection systems; however, the performance of ML techniques is not satisfactory while handling imbalanced attacks. To solve this and to improve the performance, this research work proposed an intrusion detection system based on deep neural network (DNN). Cross-correlation process is used to select the optimal features from the dataset and the selected parameters are used as building blocks for deep neural network structure to find intrusions. The experimental results confirmed that the proposed DNN performs better than conventional machine learning models such as support vector machine, decision tree, and random forest and efficiently identifies the attacks.