TrustNet: A Novel Machine Learning-Driven Dynamic Trust Establishment for Intrusion Detection in Wireless Ad Hoc Networks

This research pioneers a unique approach to wireless ad hoc network security by introducing a novel machine learning-driven dynamic trust establishment mechanism called TrustNet. Traditional methods often rely on static security measures, rendering them vulnerable to emerging threats in dynamic ad hoc environments. TrustNet leverages ensemble learning and deep neural networks to continuously assess and adapt trust levels among network nodes based on realtime behaviour analysis. Through extensive simulations and experiments, TrustNet achieved an impressive 97.5% accuracy in intrusion detection and a 95% success rate in mitigating malicious activities, surpassing the limitations of conventional methods. This novel approach fortifies wireless ad hoc networks against evolving threats and sets a precedent for dynamic trust management, showcasing its potential to redefine security paradigms in decentralized communication systems.