差分编码辅助的机器学习CSI反馈方法

Differential Coding-Aided CSI Feedback Using Machine Learning

在大规模多输入多输出(massive Multiple Input Multiple Output,mMIMO)系统中,信道状态信息(Channel State Information,CSI)反馈严重占用上行带宽资源。为此,提出差分编码辅助的机器学习CSI反馈,探索节省上行带宽资源,减小用户设备(User Equipment,UE)能量开销的途径。在UE端,通过差分编码方式反馈CSI,避免UE端发送用于上行信道估计的导引,从而减小上行带宽资源占用,并节省UE能量开销。在基站(Base Station,BS)端,首先对接收信号进行差分译码,随后构建轻量化神经网络恢复下行CSI。仿真结果表明,相对于上行导引信道估计的CSI反馈,所研究的差分编码辅助的机器学习CSI反馈方法在节省上行带宽资源、减小UE能量消耗的同时,提高了下行CSI的重构精度。

Channel state information(CSI)feedback heavily consumes uplink bandwidth resources in massive multiple input multiple output(mMIMO)systems.The differential coding-assisted ma-chine learning CSI feedback is investigated to explore an additional way to save uplink bandwidth resources and reduce the energy overhead of user equipment(UE).On the UE side,CSI is fed back by differential coding to avoid sending pilot at UE for uplink channel estimation,thereby re-ducing the occupation of uplink bandwidth resource and saving UE energy overhead.On the BS side,the received signal is differentially decoded at first,and then a lightweight neural network is built to recover the downlink CSI.Simulation results show that compared with the existing CSI feedback based on transmitting pilot,the proposed method saves uplink bandwidth resources,rereduces UE energy consumption,and improves the reconstruction accuracy of downlink CSI.