When reputation enforces evolutionary cooperation in unreliable MANETs

TL;DR

This paper introduces an evolutionary game theory-based framework to promote cooperation among self-interested nodes in unreliable MANETs, ensuring network functionality despite packet loss and noisy observations.

Contribution

It develops a novel indirect reciprocity model for cooperative packet forwarding in unreliable MANETs and analyzes the conditions for stable cooperation.

Findings

Cooperation is enforced when benefit-to-cost ratio exceeds a critical threshold.

The proposed strategy achieves network throughput close to full cooperation.

Numerical simulations validate the effectiveness of the evolutionary game approach.

Abstract

In self-organized mobile ad hoc networks (MANETs), network functions rely on cooperation of self-interested nodes, where a challenge is to enforce their mutual cooperation. In this paper, we study cooperative packet forwarding in a one-hop unreliable channel which results from loss of packets and noisy observation of transmissions. We propose an indirect reciprocity framework based on evolutionary game theory, and enforce cooperation of packet forwarding strategies in both structured and unstructured MANETs. Furthermore, we analyze the evolutionary dynamics of cooperative strategies, and derive the threshold of benefit-to-cost ratio to guarantee the convergence of cooperation. The numerical simulations verify that the proposed evolutionary game theoretic solution enforces cooperation when the benefit-to-cost ratio of the altruistic exceeds the critical condition. In addition, the network throughput performance of our proposed strategy in structured MANETs is measured, which is in close agreement with that of the full cooperative strategy.