To address the challenge of anti-jamming transmission in wireless communication systems under complex interference environments, this paper proposes a movable antennas systems anti-jamming transmission scheme based on deep unfolding networks. Although existing research has enhanced the diversity of anti-jamming strategies through mobile antenna technology, its optimization algorithms are complex, system overhead is large, and traditional deep learning schemes face problems such as strong data dependence and poor interpretability. Therefore, this paper innovatively introduces deep unfolding networks into the design of movable antennas systems, achieving efficient optimization through the fusion of algorithm unfolding and neural networks. Specifically, for the multi-jammer confrontation scenario, the non-convex joint optimization problem is decoupled into an alternating optimization problem of the receiving beamforming vector and antenna position, where the beamforming is solved in closed form through the generalized Rayleigh quotient. Further, by mapping the iterative optimization steps to the hidden layers of the neural network, a lightweight deep unfolding framework is constructed, significantly reducing the computational complexity of traditional iterative algorithms. Simulation results verify the effectiveness and reliability of the proposed scheme in real-time anti-jamming transmission under jamming environments.
原文链接
分享
引用
