深度学习辅助的编码超表面阵列智能设计OA
Deep learning-assisted intelligent design of coding metasurface arrays
针对卫星地面协同通信、低空经济、智慧城市等新兴复杂应用场景下的通信需求,提出了一种用于编码超表面快速波束赋形的卷积神经网络(convolutional neural network,CNN),以实现直接的编码超表面阵列逆向设计过程.该CNN主要基于视觉几何组(visual geometry group,VGG)网络结构,通过引入通道注意力机制以进一步提高网络预测精度.通过相位补偿和多种群遗传算法收集数据集用于训练,训练后的网络能够在几十毫秒内计算单波束和双波束的超表面编码矩阵,并且相比于现有基准网络,训练收敛速度提升 40%,预测精度提升1.31%.该网络极大提高了大规模超表面编码的设计效率,为快速实时的编码超表面控制提供了一种可行的解决方案.
To address the communication demands in complex scenarios such as satellite-terrestrial collaborative communication,low-altitude economy,and smart cities,this study proposes a convolutional neural network(CNN)for fast beamforming with coding metasurface arrays,enabling a direct inverse design process.The CNN is primarily based on the visual geometry group(VGG)network architecture,incorporating channel attention mechanisms to further enhance its prediction accuracy.Dataset is collected through phase compensation and multi-population genetic algorithm(MPGA).The trained network can generate coding matrices for both single-beam and dual-beam within milliseconds,achieving 40%faster training convergence,and 1.31%higher prediction accuracy compared to existing benchmark networks.This network significantly improves the design efficiency of large-scale coding metasurface arrays,offering a feasible solution for fast and real-time control of coding metasurface arrays.
张子奕;张嘉男;游检卫
东南大学信息科学与工程学院,南京 211189东南大学信息科学与工程学院,南京 211189东南大学信息科学与工程学院,南京 211189
信息技术与安全科学
编码超表面波束赋形深度学习卷积神经网络(CNN)通道注意力
coding metasurfacebeamformingdeep learningconvolutional neural network(CNN)channel attention mechanism
《电波科学学报》 2026 (1)
89-97,9
国家自然科学基金(62401132)国家重点研发计划(2023YFC2411100,2023YFB3813100)东南大学科研启动经费(RF1028623061)National Natural Science Foundation of China(62401132)National Key Research and Development Program of China(2023YFC2411100,2023YFB3813100)Southeast University Research Start-up Fund(RF1028623061)
评论