Cluster-Based Data Traffic Monitoring and Trusted Neighbors Prediction for Secure Data Transmission in Wireless Sensor Network
TL;DR
This research develops an Archer Fish Hopfield Forecasting System (AFHFS) for cluster-based data traffic monitoring and prediction of trusted neighbors in WSNs and shows that the suggested framework provides improved network reliability, energy efficiency, and secure communication in WSN.
Abstract
Wireless Sensor Networks (WSNs) are now vital components in environmental monitoring systems, healthcare, industrial automation, and military missions. Unfortunately, the resource-limited features of sensor nodes, along with a dynamic network environment and malicious attacks, lead to a number of difficulties associated with secure and energy-efficient data transmission. Most of the trust-based routing schemes have low efficiency in terms of ineffective cluster formation, high energy utilization, increased communication delay, and inaccurate malicious node detection. This, in turn, leads to poor network performance. Therefore, the present research develops an Archer Fish Hopfield Forecasting System (AFHFS) for cluster-based data traffic monitoring and prediction of trusted neighbors in WSNs. First, the Archer Fish Optimization (AHO) approach is utilized to find out optimal clusters and cluster formation is enabled. After that, network traffic is constantly monitored to analyze node behavior and determine trustworthiness based on a number of communication factors. Finally, the obtained trust information is fed to the Hopfield Neural Network (HNN) for the classification of neighboring nodes in terms of trustworthiness. Experimental analysis has revealed that the suggested AFHFS performs better. The results have shown that the suggested framework provides improved network reliability, energy efficiency, and secure communication in WSN.