Enhancing Cognitive Radio Networks Performance with Game-Theoretic Anti-Jamming Strategies and Trial-Error Learning

This study employs a Stackelberg game-theoretic framework to investigate anti-jamming strategies in cognitive radio networks. The study assesses the efficiency of primary and secondary users' throughput as well as the channels within the cognitive radio system utilizing a trial-and-error hierarchical learning mechanism. The effectiveness of the algorithm is measured by its convergence properties and overall system throughput under different scenarios. Simulation results indicate that the algorithm achieves convergence, stabilizing channel selection probabilities by the 54th iteration, with primary throughput decreasing from 4.3 to 1.1 as the number of channels increases from 4 to 50. Comparative analysis reveals that the Best Response Algorithm maintains throughput stability at 1.8 after 20 channels, while other algorithms experience more pronounced declines. In the presence of jamming, throughput experiences a notable decrease, starting at 3.3 and dropping to 0.8 at 50 channels. The pattern observed in secondary throughput underscores the importance of robust anti-jamming strategies. The Stackelberg game strategy outperforms the Nash equilibrium, showcasing greater utility for the leader due to its strategic positioning. These results emphasize the potential of the Stackelberg game-theoretic approach in improving network performance in the face of interference in dynamic spectrum access environments.