Distributed reinforcement learning based framework for energy-efficient UAV relay against jamming

Unmanned aerial vehicle (UAV) network is vulnerable to jamming attacks, which may cause severe damage like communication outages. Due the energy constraint, source UAV cannot blindly enlarge transmit power, along with complex topology high mobility, makes destination unable evade jammer by flying at will. To maintain a limited battery capacity in networks presence of greedy jammer, this paper, we propose distributed reinforcement learning (RL) based energy-efficient framework for constrained under attacks improve quality while minimizing total consumption network. This enables each relay independently select its power on historical state-related information without knowing moving trajectory other UAVs as well jammer. The location and level need not be shared UAVs. We also deep RL anti-jamming approach portable computation equipment Raspberry Pi achieve higher faster performance. study Nash equilibrium (NE) performance bounds formulated control game. Simulation results show that proposed schemes can reduce bit error rate (BER) compared benchmark method.