Fast Adaptive Anti-Jamming Channel Access via Deep Q Learning and Coarse-Grained Spectrum Prediction
Jianshu Zhang, Xiaofu Wu, Junquan Hu
TL;DR
The paper tackles anti-jamming channel access under dynamic and unknown jamming strategies by marrying deep Q-learning with a supervised coarse-grained spectrum predictor. This auxiliary task accelerates learning and yields a Q-function superior to standard DRL methods, achieving faster convergence and higher throughput than Nash-equilibrium-based baselines. The authors formulate the interaction as a Markov Game and demonstrate, through extensive simulations, that their approach reduces training episodes by up to 70% and improves throughput by about 10% over NE strategies, while remaining robust to non-synchronous framing and varying jammer update rates. A joint inference architecture with a shared feature extractor leverages both spectrum predictions and value estimates to make more informed channel-access decisions. The work offers a practical path toward adaptive anti-jamming in realistic, non-stationary wireless environments and highlights the trade-offs between computational complexity and convergence speed.
Abstract
This paper investigates the anti-jamming channel access problem in complex and unknown jamming environments, where the jammer could dynamically adjust its strategies to target different channels. Traditional channel hopping anti-jamming approaches using fixed patterns are ineffective against such dynamic jamming attacks. Although the emerging deep reinforcement learning (DRL) based dynamic channel access approach could achieve the Nash equilibrium (NE) under fast-changing jamming attacks, it requires extensive training episodes. To address this issue, we propose a fast adaptive anti-jamming channel access approach guided by the intuition of ``learning faster than the jammer", where a synchronously updated coarse-grained spectrum prediction serves as an auxiliary task for the deep Q network (DQN) based anti-jamming model. This helps the model identify a superior Q-function compared to standard DRL while significantly reducing the number of training episodes. Numerical results indicate that the proposed approach significantly accelerates the rate of convergence in model training, reducing the required training episodes by up to 70\% compared to standard DRL. Additionally, it also achieves a 10\% improvement in throughput over NE strategies, owing to the effective use of coarse-grained spectrum prediction.