{"title":"用于 RIS 中联合相位和前导器优化的几何感知元学习神经网络","authors":"Dahlia Devapriya, Sheetal Kalyani","doi":"arxiv-2409.11270","DOIUrl":null,"url":null,"abstract":"In reconfigurable intelligent surface (RIS) aided systems, the joint\noptimization of the precoder matrix at the base station and the phase shifts of\nthe RIS elements involves significant complexity. In this paper, we propose a\ncomplex-valued, geometry aware meta-learning neural network that maximizes the\nweighted sum rate in a multi-user multiple input single output system. By\nleveraging the complex circle geometry for phase shifts and spherical geometry\nfor the precoder, the optimization occurs on Riemannian manifolds, leading to\nfaster convergence. We use a complex-valued neural network for phase shifts and\nan Euler inspired update for the precoder network. Our approach outperforms\nexisting neural network-based algorithms, offering higher weighted sum rates,\nlower power consumption, and significantly faster convergence. Specifically, it\nconverges faster by nearly 100 epochs, with a 0.7 bps improvement in weighted\nsum rate and a 1.8 dBm power gain when compared with existing work.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"54 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Geometry Aware Meta-Learning Neural Network for Joint Phase and Precoder Optimization in RIS\",\"authors\":\"Dahlia Devapriya, Sheetal Kalyani\",\"doi\":\"arxiv-2409.11270\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In reconfigurable intelligent surface (RIS) aided systems, the joint\\noptimization of the precoder matrix at the base station and the phase shifts of\\nthe RIS elements involves significant complexity. In this paper, we propose a\\ncomplex-valued, geometry aware meta-learning neural network that maximizes the\\nweighted sum rate in a multi-user multiple input single output system. By\\nleveraging the complex circle geometry for phase shifts and spherical geometry\\nfor the precoder, the optimization occurs on Riemannian manifolds, leading to\\nfaster convergence. We use a complex-valued neural network for phase shifts and\\nan Euler inspired update for the precoder network. Our approach outperforms\\nexisting neural network-based algorithms, offering higher weighted sum rates,\\nlower power consumption, and significantly faster convergence. Specifically, it\\nconverges faster by nearly 100 epochs, with a 0.7 bps improvement in weighted\\nsum rate and a 1.8 dBm power gain when compared with existing work.\",\"PeriodicalId\":501034,\"journal\":{\"name\":\"arXiv - EE - Signal Processing\",\"volume\":\"54 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - EE - Signal Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.11270\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Signal Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11270","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Geometry Aware Meta-Learning Neural Network for Joint Phase and Precoder Optimization in RIS
In reconfigurable intelligent surface (RIS) aided systems, the joint
optimization of the precoder matrix at the base station and the phase shifts of
the RIS elements involves significant complexity. In this paper, we propose a
complex-valued, geometry aware meta-learning neural network that maximizes the
weighted sum rate in a multi-user multiple input single output system. By
leveraging the complex circle geometry for phase shifts and spherical geometry
for the precoder, the optimization occurs on Riemannian manifolds, leading to
faster convergence. We use a complex-valued neural network for phase shifts and
an Euler inspired update for the precoder network. Our approach outperforms
existing neural network-based algorithms, offering higher weighted sum rates,
lower power consumption, and significantly faster convergence. Specifically, it
converges faster by nearly 100 epochs, with a 0.7 bps improvement in weighted
sum rate and a 1.8 dBm power gain when compared with existing work.