In the context of dynamic countermeasures between radar and active jammer, the working frequency of radar and adversarial jammer were modeled as the combat action space based on the multi-arm bandit model in online learning theory. By exploring the uncertainty of the jamming environment state through multiple-round pulse transmission, a frequency channel jamming recognizer based on a convolutional neural network was constructed to obtain the posterior probability estimation of the belief state of each frequency channel. Then the Thompson sampling algorithm efficiently solved the built multi-arm bandit model, achieving a balance between exploration and exploitation. Simulation results show that compared with random frequency agility and deep reinforcement learning algorithms, the method had higher convergence performance and was more adaptable to dynamic fast-changing jamming environments, which can give full potential to the antagonism advantage of radar active waveform transmission.