CAVIaR Crayfish Algorithm Enabled Deep Kronecker Network for Wormhole Attack Mitigation on Network Control System

Wireless Sensor Network (WSN) is expansively employed in the Network Control System (NCS). Wormhole (WH) attack detection is a burdensome process, specifically over WSN. A distributed passive state of WH attack makes the network extremely challenging to identify. Here, the CAVIaR crayfish algorithm enabled Deep Kronecker Network (CCA-DKN) is introduced for mitigation of WH attack on NCS. Firstly, the WSN model is simulated and routing is executed by Low Energy Adaptive Clustering Hierarchy (LEACH) protocol. Then, three stages namely in-band (IB)- WH detection, out-of-band (OB)-WH detection, and Neighbour Ratio Threshold (NRT) are accomplished for WH detection. After the accomplishment of the NRT stage, the transmission range stage is performed in OB- WH detection. Thereafter, the Packet Delivery Ratio (PDR) and Round Trip Time (RTT) are done on IB- WH detection. Next, the detection of WH attack is performed by ResNeXt-DSAE. Furthermore, ResNeXt-DSAE is designed by combining ResNeXt and Deep Stacked Autoencoder (DSAE). Lastly, attack mitigation is executed, whereupon a percentage of data reduction is carried out by exploiting the Deep Kronecker Network (DKN) and it is trained by the CAVIaR crayfish algorithm (CCA). Furthermore, CCA is an assimilation of Conditional Autoregressive Value at Risk (CAViaR) and Crayfish Optimization Algorithm (COA). Additionally, CCA-DKN achieved high PDR and throughput of 0.769 and 0.533Mbps as well as low delay of 0.712sec.