Optimized Weighted Trust Evaluation Based Intrusion Detection System in WSN

Abstract

The individual nodes of a wireless sensor network (WSN) implemented in a hostile atmosphere might be readily penetrated by an adversary owing to restrictions such as short-term battery lifespan, storage capacity, and processing capabilities. It is vital to recognise and separate infected nodes with the goal to prevent being misled by the opponent's fake information provided through hacked nodes. However, due of their low flexibility and high communications cost, flat topological networks are difficult to protect effectively. In this research, we suggested an optimised based on weighted-trust assessment to identify malicious networks on top of hierarchy WSN architecture. In this study, we suggested an optimised weighted trust assessment-based intrusion detection system in WSNs, which makes use of a highly adaptable hierarchical trust administration method for clustering wireless sensor networks. To begin, a reliable assessment framework for trust perceptions is offered, which can calculate the node's trusted value based on its activity in order to successfully detect and isolate harmful nodes. Secondly, the trust evaluation model is introduced into an optimized path selection to increase security measures for data forwarding. The simulations of the outputs indicate that the suggested method greatly improves efficiency with respect to of packet loss percentage, end-to-end latency, efficiency, and use of energy, and also that it is resistant to black hole attacks.