Radio Waveforms Classification via Deep Q Learning Network
Siqi Lai, Mingliang Tao, Xiang Zhang, Ling Wang · 2021
Radio waveforms classification plays a foundation role in cognitive radio, which promises a broad prospect in spectrum monitoring and management. In this paper, a radio waveforms classification via deep Q learning is proposed, in which a deep reinforcement learning agent is trained to classify signal modulation type. Differ from the widely applied deep learning strategy, the proposed method has strong self-learning decision-making ability, which can find the optimal strategy by trial and error. The simulation results show that it can realize classification of radio signal modulation type with high accuracy.