Topology-Aware Resilient Routing Protocol for FANETs: An Adaptive Q-Learning Approach

Flying ad hoc networks (FANETs) play a crucial role in numerous military and\ncivil applications since it shortens mission duration and enhances coverage\nsignificantly compared with a single unmanned aerial vehicle (UAV). Whereas,\ndesigning an energy-efficient FANET routing protocol with a high packet\ndelivery rate (PDR) and low delay is challenging owing to the dynamic topology\nchanges. In this article, we propose a topology-aware resilient routing\nstrategy based on adaptive Q-learning (TARRAQ) to accurately capture topology\nchanges with low overhead and make routing decisions in a distributed and\nautonomous way. First, we analyze the dynamic behavior of UAV nodes via the\nqueuing theory, and then the closed-form solutions of neighbors' change rate\n(NCR) and neighbors' change interarrival time (NCIT) distribution are derived.\nBased on the real-time NCR and NCIT, a resilient sensing interval (SI) is\ndetermined by defining the expected sensing delay of network events. Besides,\nwe also present an adaptive Q-learning approach that enables UAVs to make\ndistributed, autonomous, and adaptive routing decisions, where the above SI\nensures that the action space can be updated in time at a low cost. The\nsimulation results verify the accuracy of the topology dynamic analysis model\nand also prove that our TARRAQ outperforms the Q-learning-based topology-aware\nrouting (QTAR), mobility prediction-based virtual routing (MPVR), and greedy\nperimeter stateless routing based on energy-efficient hello (EE-Hello) in terms\nof 25.23%, 20.24%, and 13.73% lower overhead, 9.41%, 14.77%, and 16.70% higher\nPDR, and 5.12%, 15.65%, and 11.31% lower energy consumption, respectively.\n