LLM4CP:Adapting Large Language Models for Channel PredictionOA
LLM4CP:Adapting Large Language Models for Channel Prediction
Channel prediction is an effective approach for reducing the feedback or estimation overhead in massive multi-input multi-output(m-MIMO)systems.However,existing channel prediction methods lack precision due to model mismatch errors or network gen-eralization issues.Large language models(LLMs)have demonstrated powerful modeling and generalization abil-ities,and have been successfully applied to cross-modal tasks,including the time series analysis.Leveraging the expressive power of LLMs,we propose a pre-trained LLM-empowered channel prediction(LLM4CP)method to predict the future downlink channel state informa-tion(CSI)sequence based on the historical uplink CSI sequence.We fine-tune the network while freezing most of the parameters of the pre-trained LLM for better cross-modality knowledge transfer.To bridge the gap between the channel data and the feature space of the LLM,preprocessor,embedding,and output modules are specifically tailored by taking into account unique channel characteristics.Simulations validate that the proposed method achieves state-of-the-art(SOTA)prediction per-formance on full-sample,few-shot,and generalization tests with low training and inference costs.
Boxun Liu;Xuanyu Liu;Shijian Gao;Xiang Cheng;Liuqing Yang
State Key Laboratory of Advanced Optical Communication Systems and Networks,School of Electronics,Peking University,Beijing 100871,ChinaInternet of Things Thrust,The Hong Kong University of Science and Technology(Guangzhou),Guangzhou 511400,ChinaInternet of Things Thrust and Intelligent Transportation Thrust,The Hong Kong University of Science and Technology(Guangzhou),Guangzhou 511400,China||Department of Electronic and Computer En-gineering and Department of Civil and Environmental Engineering,The Hong Kong University of Science and Technology,Hong Kong,China
channel predictionmassive multi-input multi-output(m-MIMO)large language models(LLMs)fine-tuningtime-series
《通信与信息网络学报（英文）》 2024 (002)
113-125 / 13
This work was supported in part by the National Natu-ral Science Foundation of China under Grants 62125101 and 62341101,in part by the New Cornerstone Science Foundation through the XPLORER PRIZE,in part by Guangdong Provincial Key Lab of Integrated Commu-nication,Sensing and Computation for Ubiquitous Internet of Things un-der Grant 2023B1212010007,in part by Guangzhou Municipal Science and Technology Project under Grant 2023A03J0011,and in part by Guangdong Provincial Department of Education Major Research Project under Grant 2023ZDZX1037.
评论