Game Theory Analysis and Modeling of Sophisticated Multi-Collusion Attack in MANETs

Abstract

Mobile Adhoc Network (MANET) has been a core topic of research since the last decade. Currently, this form of networking paradigm is increasingly being construed as an integral part of upcoming urban applications of Internet-of-Things (IoT), consisting of massive connectivity of diverse types of nodes. There is a significant barrier to the applicability of existing routing approaches in conventional MANETs when integrated with IoT. This routing mismatch can lead to security risks for the MANET-based application tied with the IoT platform. This paper examines a pragmatic scenario as a test case wherein the mobile nodes must exchange multimedia signals for supporting real-time streaming applications. There exist two essential security requirements viz. i) securing the data packet and ii) understanding the unpredictable behavior of the attacker. The current study considers sophistication on the part of attacker nodes. They are aware of each other's identity and thereby collude to conduct lethal attacks, which is rarely reflected in existing security modeling statistics. This research harnesses the potential modeling aspect of game theory to model the multiple-collusion attacker scenario. It contributes towards i) modeling strategies of regular/malicious nodes and ii) applying optimization principle using novel auxiliary information to formulate the optimal strategies. The model advances each regular node's capability to carry out precise computation about the opponent player's strategy prediction, i.e., malicious node. The simulation outcome of the proposed mathematical model in MATLAB ascertains that it outperforms the game theory's baseline approach.

Exploring Manet Security Aspects: Analysis of Attacks and Node Misbehaviour Issues

“…Reputation-based technique [77], [78], [79] SAODV detection rates range from 90 to 100 per cent, while AODV detection rate is about 70 per cent.…”

“…The fundamental goal of a reputation-based MANET solution is to identify problematic nodes through the monitoring process [77], [78], [79]. Every node in a network observes and recognises the behaviour of its neighbours in real-time, a technique known as direct or first-hand experience of reputation information.…”

SGM: Strategic Game Model for Resisting Node Misbehaviour in IoT-Cloud Ecosystem

“…The main purpose of the proposed scheme is to develop a framework capable of resisting the participation of misbehaved nodes, as well as strategies that can identify any form of nodes with an unauthorised presence in the data-forwarding process other than regular nodes. The work is an extension of our previous research [55][56][57][58] in which game theory on MANET is incorporated in addressing security vulnerabilities due to the presence of misbehaving nodes in a much larger scenario involving mobile IoT systems and cloud environments. The framework architecture for this purpose is shown in Figure 1.…”

MSA-SFO-based Secure and Optimal Energy Routing Protocol for MANET

“…Malicious nodes may now be accurately predicted by each normal node, thanks to the new model's enhanced capacity to do exact computations. Game theory's baseline method is outperformed by the suggested mathematical model in MATLAB simulations [7].…”