Fast Reinforcement Learning for Anti-jamming Communications

Abstract

This letter presents a fast reinforcement learning algorithm for anti-jamming communications which chooses previous action with probability and applies -greedy with probability (1-). A dynamic threshold based on the average value of previous several actions is designed and probability is formulated as a Gaussian-like function to guide the wireless devices. As a concrete example, the proposed algorithm is implemented in a wireless communication system against multiple jammers. Experimental results demonstrate that the proposed algorithm exceeds Q-learing, deep Q-networks (DQN), double DQN (DDQN), and prioritized experience reply based DDQN (PDDQN), in terms of signal-to-interference-plus-noise ratio and convergence rate.

Tasks

Reproductions