Movable Antenna Array Design for Jamming Mitigation

The revolutionary technology relying on movable antenna (MA) array is gaining rising interest due to the ability of flexible antenna positioning, which enables the benefit of accurate beamforming over the conventional fixed-position antenna array. In this paper, we consider the scenario in which multiple legitimate users attempt to send useful signals to a base station equipped with an MA array under the attack of multiple malicious jammers. Specifically, considering the jamming channel uncertainty, we propose an MA array-based robust malicious jamming mitigation scheme. We formulate an average sum-rate maximization problem subject to the antenna position constraints. Due to the non-convexity and highly coupled variables of the formulated problem, an alternating optimization framework for the receive beamforming and the antenna position vector (APV) is developed to address the challenging issue. In particular, the optimal receive beamforming is derived in closed-form and the APV is updated using the stochastic successive convex approximation. In addition, we propose a model-driven deep learning scheme to accelerate the convergence. Numerical results demonstrate that the proposed scheme outperforms the conventional fixed-position antenna array in terms of jamming mitigation.