Dynamic Spectrum Access for UAV-Aided Cognitive Radio Network under Jamming Attacks

Unmanned Aerial Vehicle (UAV)-aided cognitive radio networks combine the high flexibility and controllability of UAV s, effectively alleviating the spectrum scarcity issue in wireless communications. However, the wireless transmission is vulnerable to malicious attacks due to its inherent character. This paper investigates the dynamic spectrum access strategy in UAV-aided cognitive radio scenarios under jamming attacks. By jointly optimizing the spectrum access decisions and mobility trajectories of UAV, we aim to maximize the communication rate between the UAV and the Secondary Users (SUs) under the jamming attacks, while considering the UAV's energy consumption and avoiding interference with the communication of the Primary Users (PUs). We formulate this optimization problem as a Markov Decision Process (MDP) and propose a reinforcement learning algorithm based on Double Deep Q Network (DDQN) to solve it. Additionally, the proposed algorithm is verified in various of simulation experiments.