Reputation-Based Opportunistic Routing Protocol Using Q-Learning For Manet Attacked By Malicious Nodes: A Survey

Abstract

Mobile Ad-hoc Networks (MANETs) have emerged as a viable paradigm for creating dynamic communication between mobile nodes in the absence of a permanent infrastructure. However, since MANETs are inherently decentralized and self-organizing, they are vulnerable to a variety of security concerns, especially when hostile nodes seek to interrupt communication and undermine network operations. As a consequence, developing efficient and robust routing protocols is critical to ensuring dependable and secure data transmission in MANETs. This review study gives an in-depth look into Reputation-Based Opportunistic Routing Protocols (RORPs), which use Q-Learning methods to improve MANET resistance against malicious node assaults. We provide a thorough study of current RORPs, delving into their core design ideas, underlying processes, and comparative performance evaluations. We begin with a comprehensive introduction to the fundamental concepts, issues, and security risks associated with MANETs. Following that, we explore the concepts of reputation-based routing, emphasizing the need of reputation management in discriminating between malicious and genuine nodes. The review then goes into further detail on the incorporation of Q-Learning methods into RORPs, emphasizing on how reinforcement learning mechanisms may adaptively change routing choices depending on the developing network state and the dynamic reputations of participating nodes. This article reviews 38 research papers of opportunistic routing protocol and explores the potential of computer-assisted methods for MANET.  We next divide the studied RORPs into categories depending on their approach to reputation computation, learning algorithms, and cooperative enforcement methods. Furthermore, we analyze their strengths and weaknesses in terms of resilience to different sorts of attacks, as well as their influence on network performance measures including packet delivery ratio, end-to-end latency, and throughput.