基于人工智能的大规模MIMO信道状态信息反馈综述

Survey of Massive MIMO Channel State Information Feedback Based on Artificial Intelligence

大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)技术凭借其高能量效率和高频谱效率的优势成为下一代移动通信的核心技术之一,其系统增益的基础在于基站能够精确获知信道状态信息(Channel State Information,CSI)。由于大规模MIMO系统中基站天线数量巨大,基站获取下行信道状态信息将造成巨大的系统开销,传统基于码本或矢量量化的反馈方法受到挑战,频分双工(Frequency Division Duplex,FDD)模式下5G通信的实际应用也受到制约,而人工智能技术尤其是深度学习(Deep Learning,DL)为解决大规模MIMO系统中的CSI反馈问题提供了新的思路。围绕大规模MIMO系统CSI反馈存在的问题,阐述了CSI反馈的背景,构建了FDD大规模MIMO系统模型,详细描述了代表性的国内外基于DL的CSI反馈方案,最后对基于人工智能的大规模MIMO信道状态信息反馈进行了展望和总结。

The massive multiple-input multiple-output(MIMO)technology has become one of the core technologies of the next generation mobile communication due to its advantages of high energy efficiency and high frequency spectrum efficiency.The basis of the system gain is that the base station can accurately know the channel state information(CSI).Due to the large number of base station antennas in massive MIMO systems,the acquisition of downlink CSI by base station will cause huge system overhead,which challenges the traditional feedback methods based on codebook or vector quantization,and restricts the practical application of 5G communication in frequency division duplex(FDD)mode.Artificial intelligence technology,especially deep learning(DL),provides a new way to solve CSI feedback problems in massive MIMO systems.Focusing on the existing problems of CSI feedback in massive MIMO systems,this paper introduces the background of CSI feedback,constructs an FDD massive MIMO system model,describes representative CSI feedback schemes based on deep learning at home and abroad,and finally presents the prospect and summary of massive MIMO CSI feedback based on artificial intelligence.