Research on Bayesian Optimization of Operating Parameters of the Deceptive Infrared Jammer

Abstract The working parameters of deceptive infrared jammers are generally set manually by combat personnel based on their accumulated combat experience, which is not conducive to adapting to the various combat conditions on the battlefield. To solve the problem, in this paper the working parameters of the deceptive infrared jammer are optimized using the Bayesian method from machine learning to achieve the best jamming effect. It have been studied that the selection of optimized parameters, the establishment of the objective function, and the choice of the probability surrogate function and acquisition function. Experimental results show that the parameters optimized by Bayesian optimization enable the infrared jammer to perform deceptive jamming more quickly and effectively, with the seeker offset reaching 25.1°.