Attack detection and mitigation using Intelligent attack graph model for Forensic in IoT Networks

This article focuses on the urgent cybersecurity concerns in the Internet of Things (IoT) environment, highlighting the crucial importance of protecting these networks in the face of increasing amounts of IoT data. The paper explores the intricacies of deploying security mechanisms for Internet of Things (IoT) devices, specifically those that are restricted by limited resources. This study examines the inherent weaknesses in IoT systems and analyses the strategies used by malicious individuals to gain control and privileges. In order to tackle these difficulties, the study suggests a sophisticated security system that combines artificial intelligence and an intelligent attack graph. An outstanding characteristic of the model incorporates a method devised to restrain virus spread and accelerate network restoration by introducing virtual nodes. The research showcases the results of the vulnerable attack path predictor (VAPP) module of the proposed model, emphasising its exceptional accuracy in distinguishing between black (0) and red (1) attack paths compared to alternative Machine Learning techniques. Moreover, a thorough evaluation of the module's performance is carried out, with a specific emphasis on security concerns and predictive capacities. Proverif is utilised to validate the security settings and evaluate the resilience of the secret keys. The findings demonstrate a detection rate of 98.48% and an authentication rate of 85%, outperforming the achievements of earlier studies. The contributions greatly enhance the ability of IoT networks to withstand challenges, and the use of cryptographic verification confirms its dependability in the ever-changing digital environment.