QEGT-Based Adaptive Routing for Energy-Efficient and Reliable Communication in UAV Swarm Networks

Abstract

Unmanned Aerial Vehicle (UAV) swarm networks play a vital role as real-time communication relays. It provides communication services in applications such as disaster management, intelligent transport, and environmental monitoring. The inherent characteristics of Unmanned Swarm Networks (USNs), such as flexible deployment and high operational adaptability, make them valuable in environments with limited infrastructure. However, the highly dynamic and decentralized nature of USNs presents major challenges in designing efficient and reliable routing protocols. Traditional approaches often struggle with frequent topology changes and high mobility of USNs. To address these challenges, this study proposes an intelligent Q-learning-enhanced Evolutionary Game Theory (QEGT) routing mechanism for USNs. The proposed scheme leverages game-theoretic incentives and Q-learning to adaptively select strategies. The optimal next-hop UAV is selected based on residual energy, queue length, and proximity to the destination. Extensive simulations in NS-3 demonstrate the effectiveness of the proposed QEGT scheme. Simulation results show that the proposed QEGT scheme outperforms the state-of-the-art approaches in terms of network survival time, the number of successfully delivered packets, average hop count, and average delay.