Enhancing for IoT based Network Security on Wormhole Attack Detection Techniques using Wireless Sensor Network

Abstract

The security of Internet of Things (IoT)-enabled wireless sensor networks is increasingly challenged by sophisticated attacks that exploit their unique characteristics, including resource constraints, heterogeneity, and decentralised architectures. Understanding and protecting IoT networks against wormhole attacks requires a multifaceted approach encompassing the attack mechanics, their impact on IoT, and appropriate detection and mitigation strategies adapted to IoT’s specific constraints. The primary goal of Proposed Enhanced Network Security with Wormhole Attacks (ENS-WHA) is to cover the several cutting-edge strategies that have been successfully employed over the years to identify WHA in wireless networks. The current WHA detection methods are inefficient due to the requirement for more technology, longer latency, and increased energy usage. Due to the lack of extra hardware requirements, round-trip time (RTT)-related detection approaches are becoming more successful. The existence of two machine learning models, a Support Vector Machine (SVM) and a Deep Neural Network (DNN), to classify the traffic data and identify malicious nodes in the network. Results from performance evaluations indicate that the ENS-WHA's method achieves better efficiency than SVM and DNN methods. The enhancement in packet delivery ratio, delay time, energy consumption, and throughput analysis confirms that it can reduce the latency and energy wastage and improve the data delivery success rates, positioning it as a more robust and effective network strategy.