Intelligent analytics algorithms in breach detection systems for securing VANETs and data for smart transportation management

Abstract

Due to the increasing prevalence of sensor-centric communication and computing devices installed within a vehicle for a wide range of functions, including vehicle monitoring, physical wiring minimization, and driving efficiency, in-vehicle communication has emerged as an integral part of the modern driving environment. Nevertheless, there are presently no certain remedies for in-vehicle cyber dangers provided by the relevant literature on cyber security for in-vehicle communication techniques. Existing solutions do not provide a complete security framework for in-vehicle communication since they depend heavily on protocol-specific security techniques. The purpose of this research is to provide a method for smart vehicle management in VANET breach detection that makes use of efficient data transfer and clever machine learning. In this research, we use a novel approach to intrusion detection based on an ensemble of adversarial Boltzmann CNNs. And then we utilize the secure short hop opportunistic local routing protocol to get the information where it needs to go. Experimentation analysis for a wide range of security-based datasets includes throughput, QoS, training accuracy, validation accuracy, and network security analysis. Throughput of 88%, Quality of Service of 77%, Training Accuracy of 93%, Validation Accuracy of 96%, Network Security Analyses of 63%, Scalability of 75% were all achieved using the suggested method.