Lightweight Differential Frameworks for CSI Feedback in Time-Varying Massive MIMO Systems

被引：2

作者：

Zhang, Yangyang ^{[1
]}

Zhang, Xichang ^{[1
]}

Liu, Yi ^{[1
]}

机构：

[1] Xidian Univ, State Key Lab Integrated Serv Network, Xian 710071, Peoples R China

来源：

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY | 2024年 / 73卷 / 05期

基金：

国家重点研发计划;

关键词：

Complexity theory; Image reconstruction; Correlation; Neural networks; Massive MIMO; Downlink; Mathematical models; Massive multiple-input multiple-output (MIMO); channel state information (CSI) feedback; deep learning; differential method; channel correlation; LIMITED FEEDBACK; NEURAL-NETWORKS;

D O I：

10.1109/TVT.2023.3345938

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Channel state information (CSI) is vital for massive multiple-input multiple-output (MIMO) systems to provide high channel capacity and energy efficiency. However, the massive antennas will lead to high feedback overhead in frequency division duplex (FDD) MIMO systems. To leverage the temporal correlation of the channel in CSI feedback, recent applications of recurrent neural networks have demonstrated promising results but with tremendous complexity. The MarkovNet reduces the overall complexity but owns the high complexity at the user equipment (UE), which is not suitable for practical deployment. In this work, we aim to improve the reconstruction performance and reduce the complexity of the UE. Leveraging the channel temporal correlation, we propose two differential frameworks called DIFNet1 and DIFNet2 to improve feedback accuracy and efficiency. Moreover, we explore the real part and imaginary part correlations of channel differential terms to reduce the complexity. Finally, we unfold the Iterative Shrinkage-Thresholding Algorithm (ISTA) to provide excellent reconstruction performance. Simulation results demonstrate that the proposed frameworks improve the reconstruction performance while greatly reducing the complexity of UEs.

引用

页码：6878 / 6893

页数：16

共 54 条

[1] Massive MIMO: survey and future research topics [J].

Araujo, Daniel C. ;

Maksymyuk, Taras ;

de Almeida, Andre L. F. ;

Maciel, Tarcisio ;

Mota, Joao C. M. ;

Jo, Minho .

IET COMMUNICATIONS, 2016, 10 (15) :1938-1946

[2] Lightweight Convolutional Neural Networks for CSI Feedback in Massive MIMO [J].

Cao, Zheng ;

Shih, Wan-Ting ;

Guo, Jiajia ;

Wen, Chao-Kai ;

Jin, Shi .

IEEE COMMUNICATIONS LETTERS, 2021, 25 (08) :2624-2628

[3] Viewing Channel as Sequence Rather Than Image: A 2-D Seq2Seq Approach for Efficient MIMO-OFDM CSI Feedback [J].

Chen, Zirui ;

Zhang, Zhaoyang ;

Xiao, Zhuoran ;

Yang, Zhaohui ;

Wong, Kai-Kit .

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2023, 22 (11) :7393-7407

[4]

Cheng P, 2014, 80 IEEE VEHICULAR TE

[5] A Scalable Limited Feedback Design for Network MIMO Using Per-Cell Product Codebook [J].

Cheng, Yong ;

Lau, Vincent K. N. ;

Long, Yi .

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2010, 9 (10) :3093-3099

[6]

Cui Q., 2023, IEEE Wireless Commun. Lett., early access, DOI [10.1109/LWC.2023.3249739.[30]X, DOI 10.1109/LWC.2023.3249739.[30]X]

[7] Unsupervised Online Learning in Deep Learning-Based Massive MIMO CSI Feedback [J].

Cui, Yiming ;

Guo, Jiajia ;

Wen, Chao-Kai ;

Jin, Shi ;

Han, Shuangfeng .

IEEE COMMUNICATIONS LETTERS, 2022, 26 (09) :2086-2090

[8] An iterative thresholding algorithm for linear inverse problems with a sparsity constraint [J].

Daubechies, I ;

Defrise, M ;

De Mol, C .

COMMUNICATIONS ON PURE AND APPLIED MATHEMATICS, 2004, 57 (11) :1413-1457

[9]

Howard AG, 2017, Arxiv, DOI [arXiv:1704.04861, 10.48550/arXiv.1704.04861]

[10] Deep Learning-Based Channel Estimation for Wideband Hybrid MmWave Massive MIMO [J].

Gao, Jiabao ;

Zhong, Caijun ;

Li, Geoffrey Ye ;

Soriaga, Joseph B. ;

Behboodi, Arash .

IEEE TRANSACTIONS ON COMMUNICATIONS, 2023, 71 (06) :3679-3693

← 1 2 3 4 5 6 →