基于数据聚类的CSI反馈Transformer网络简化实现方法  

A Simplified Implementation Method of CSI Feedback Transformer Network Based on Data Clustering

作　　者：还冬锐 张逸帆 姜明[1,2] HUAN Dongrui;ZHANG Yifan;JIANG Ming(National Mobile Communications Research Laboratory,Southeast University,Nanjing 210096,China;Purple Mountain Laboratories,Nanjing 211111,China)

机构地区：[1]东南大学移动通信全国重点实验室,南京210096 [2]紫金山实验室,南京211111

出　　处：《数据采集与处理》2025年第2期431-445,共15页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金(62271137)。

摘　　要：为应对大规模多输入多输出(Multiple⁃input multiple⁃output,MIMO)系统中信道状态信息(Channel state information,CSI)反馈开销的日益增长,基于深度学习的CSI反馈网络(如Transformer网络)受到了广泛的关注,是一种非常有应用前景的智能传输技术。为此,本文提出了一种基于数据聚类的CSI反馈Transformer网络的简化方法,采用基于聚类的近似矩阵乘法(Approximate matrix multiplication,AMM)技术,以降低反馈过程中Transformer网络的计算复杂度。本文主要对Transformer网络的全连接层计算(等效为矩阵乘法),应用乘积量化(Product quantization,PQ)和MADDNESS等简化方法,分析了它们对计算复杂度和系统性能的影响,并针对神经网络数据的特点进行了算法优化。仿真结果表明,在适当的参数调整下,基于MADDNESS方法的CSI反馈网络性能接近精确矩阵乘法方法,同时可大幅降低计算复杂度。In order to cope with the increasing overhead of channel state information(CSI)feedback in massive multiple-input multiple-output(MIMO)systems,deep learning-based CSI feedback networks(such as Transformer)have received extensive attention and become very promising intelligent transmission technologies.To this end,this paper proposes a simplification method of CSI feedback Transformer network based on data clustering,which uses clustering-based approximate matrix multiplication(AMM)to reduce the computational complexity of the Transformer network in the feedback process.In this paper,we focus on the computation of the fully connected layer in the Transformer network(equivalent to matrix multiplication),adopt the simplification methods such as product quantization(PQ)and MADDNESS,analyze their influence on the computational complexity and system performance,and optimize the algorithm according to the characteristics of neural network data.Simulation results show that the performance of the CSI feedback network based on the MADDNESS method is close to that of the exact matrix multiplication method with an appropriate parameter adjustment,and the computational complexity can be greatly reduced.

关 键 词：信道状态信息反馈 多输入多输出 神经网络 近似矩阵乘法 聚类计算 

分 类 号：TN92[电子电信—通信与信息系统]