Learning-Based Transmit Resource Management Scheme for Multiple Target Tracking With Active Jamming Mitigation

In this article, a learning-based transmit resource management scheme with active jamming mitigation (LRMS-AJM) is developed for multifunctional radar systems. The key idea of the method is to pick the joint transmit resource scheme w.r.t frequency, time, and power in order to achieve a low probability of intercept, further combat jamming and improve multiple targets tracking performance in electronic countermeasures environments. To combat the unknown risks of being intercepted and jammed by the target, we first present an active jamming mitigation partially observable Markov decision process model. Consider the elements of the learned jamming mitigation model and the limited system resources of the radar, a constrained optimization model is built to maximize the cumulative multiple target tracking performance for LRMS-AJM. Finally, the policy rollout-based online decision method is developed to solve the resulting optimization problem, where the base policy is designed as a greedy model, which is solved by the proposed sorting search algorithm, and it is demonstrated to accelerate the decision process. Numerical results also indicate that the proposed method can significantly reduce the impact of the jamming and improve the target tracking performance with a given resource budget.