CLUSTER HEAD ROTATION AND HYBRID CNN-SVM FOR SECURE ROUTING IN WSNs

Abstract

Trust management system in Wireless Sensor Network (WSN) is rudimentary to identify malicious, selfish and compromised nodes. Many proposed trust management systems provide security to WSN from various internal attacks. However, like other security systems, trust management schemes are also vulnerable to attacks. However existing work failed to focus on energy efficiency of the nodes in the network and it may lead to the network failure. To overcome those issues this work first introduced a Improved fuzzy  c means clustering (IFCM) model for network node clustering and then cluster heads will be selected based on adaptive fuzzy min-max neural network (AFMN). Then in this work, a cluster-head rotating election routing protocol is proposed to alleviate the problem. We had positive effects on reducing the energy consumption of cluster head election and intra cluster communication. And additionally define a new attack detection system which suitable to the properties of WSNs. The proposed work is implemented in the Guard against Trust Management with Hybrid convolutional neural network and support vector machine classifier (GATE-HCNNSVM) and weighted Artificial Bee Colony Optimization (WABC) for optimal selection of window length. The consideration of positive integer based trust domain and dynamic timing window demands fewer system resources. Then, the intruder is recognized efficiently in the group by checking the network flow information. Direct trust establishment scheme uses both reward and penalty policies for each good and malicious behavior of a node, respectively, to make the trust computation more realistic. Theory and simulation show that the proposed GATE method has a high detection rate of the malicious nodes and it demands fewer resources than some of the recently proposed trust management schemes.