Blockchain based ML Model for MANET Trust Calculation

Abstract

Internet security is crucial in the globally connected digital world, given the pervasive increase of the Internet of Things (IoT) over the past two decades. Among the rising IoT inventions, wireless and mobile ad hoc networks (MANET) are significant as they are utilized in a variety of fields. Low security and reliability risk are significant issues in IoT networks. Among the existing IoT network architectures, mobile ad hoc networks are more vulnerable in the aspects of safety and dependability due to the mobility of the nodes, and lack of computational power in nodes together with the decentralized architecture. Multiple studies have been conducted in recent years to enhance mobile ad-hoc network security. Yet, the research gap is still to be filled due to the complexity of the improved network architectures and the increased number of IoT mobile devices. Cyber computational trust can be utilized in improving the security and reliability of a mobile ad hoc network by propagating the trust among the network nodes to establish secured connection links. Adapting the Blockchain mechanism to propagate trust among network nodes is a trending research experiment due to the blockchain’s decentralized architecture and secure encrypted data transmission. The proposed approach in this journal paper explains a novel mechanism for computing the trust values of the nodes in a mobile ad-hoc network based on the blockchain mechanism with a consensus algorithm. The calculated trust values are used to train a supervised learning classifier to classify the trustworthy and malicious nodes to mitigate the risk of transmitting data in a mobile ad hoc network.