摘要
基于RIS的通感一体技术可以通过DOA和波束成形有效提升系统通信和感知的整体性能。针对RIS编码优化中计算复杂度大和设计自由度受限等问题,提出基于神经网络的RIS通感一体编码优化方法。该方法利用基于异步时空编码超表面的神经网络,将-12 dB的低信噪比条件下的DOA估计误差降至0.22°;并利用以自由形式设计指标为导向的串联神经网络,实现波束成形高精度优化,误差仅为0.025。该方法为RIS通感一体编码优化提供了低复杂度和高实时性的解决方案。
The integrated sensing and communication (ISAC) technology based on reconfigurable intelligent surface (RIS) caneffectively enhance the overall system communication and sensing capabilities through direction of arrival (DOA) andbeamforming techniques. To address the challenges of high computational complexity and constrained design flexibility inRIS encoding optimization, this paper proposes a neural network-based approach for coding optimization in RIS-enabledISAC systems. This approach employs a neural network based on an asynchronous space-time coding metasurface to reducethe DOA estimation error to 0.22° under low signal-to-noise ratio (SNR) conditions of -12 dB. Furthermore, by utilizing acascaded neural network driven by free-form design indicators, it accomplishes high-precision beamforming optimizationwith an error of merely 0.025. This method offers a solution with low complexity and high real-time performance for thecoding optimization in RIS-enabled ISAC systems.
作者
关东方
卞小贝
谷紫洋
陈冠潮
安康
GUAN Dongfang;BIAN Xiaobei;GU Ziyang;CHEN Guanchao;AN Kang(College of Electronic Science and Technology,National University of Defense Technology,Changsha 410073,China;Sixty-third Research Institute,National University of Defense Technology,Nanjing 210007,China)
出处
《移动通信》
2024年第4期54-60,共7页
Mobile Communications
基金
国家自然科学基金面上项目“信息超材料宽角散射增强及雷达无源干扰技术”(62271493)。
关键词
智能超表面
通感一体化
深度学习
循环神经网络
DOA估计
编码优化
Reconfigurable intelligent surface
integrated sensing and communications
deep learning
RNN
DOA estimation
coding optimization
