TY - JOUR
T1 - CV-3DCNN
T2 - Complex-Valued Deep Learning for CSI Prediction in FDD Massive MIMO Systems
AU - Zhang, Yibin
AU - Wang, Jie
AU - Sun, Jinlong
AU - Adebisi, Bamidele
AU - Gacanin, Haris
AU - Gui, Guan
AU - Adachi, Fumiyuki
N1 - Funding Information:
Manuscript received September 3, 2020; accepted September 27, 2020. Date of publication September 29, 2020; date of current version February 9, 2021. This work was supported in part by the Major Project of the Ministry of Industry and Information Technology of China under Grant TC190A3WZ-2; in part by National Natural Science Foundation of China under Grant 61901228; in part by the Six Top Talents Program of Jiangsu under Grant XYDXX-010; and in part by the 1311 Talent Plan of Nanjing University of Posts and Telecommunications. The associate editor coordinating the review of this article and approving it for publication was M. Derakhshani. (Corresponding author: Guan Gui.) Yibin Zhang, Jie Wang, Jinlong Sun, and Guan Gui are with the College of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China (e-mail: b16011024@njupt.edu.cn; 2018010223@njupt.edu.cn; sunjinlong@njupt.edu.cn; guiguan@njupt.edu.cn).
Publisher Copyright:
© 2012 IEEE.
PY - 2021/2
Y1 - 2021/2
N2 - In beyond fifth-generation (B5G) era, massive multiple-input multiple-output (M-MIMO) will be a key technology to offer higher network capacities. Due to the different frequency of uplink and downlink channels in FDD systems, the channel state information (CSI) feedback from user terminal to the base station is necessary, but this reduces the spectrum efficiency. This letter proposes a deep learning based solution to predict the downlink CSI in frequency division duplex (FDD) systems, which is termed as complex-valued three dimensional convolutional neural network (CV-3DCNN). The proposed network uses a complex-valued neural network in complex domain to deal with the complex CSI matrices, and adopts three-dimensional convolution operations for feature extraction. The proposed scheme aims to make full use of the hidden information of the complex matrices of the CSI data, and to minimize information loss caused by data processing. The experimental results demonstrate that the proposed architecture can improve accuracy of the downlink CSI prediction by approximately 6 dB.
AB - In beyond fifth-generation (B5G) era, massive multiple-input multiple-output (M-MIMO) will be a key technology to offer higher network capacities. Due to the different frequency of uplink and downlink channels in FDD systems, the channel state information (CSI) feedback from user terminal to the base station is necessary, but this reduces the spectrum efficiency. This letter proposes a deep learning based solution to predict the downlink CSI in frequency division duplex (FDD) systems, which is termed as complex-valued three dimensional convolutional neural network (CV-3DCNN). The proposed network uses a complex-valued neural network in complex domain to deal with the complex CSI matrices, and adopts three-dimensional convolution operations for feature extraction. The proposed scheme aims to make full use of the hidden information of the complex matrices of the CSI data, and to minimize information loss caused by data processing. The experimental results demonstrate that the proposed architecture can improve accuracy of the downlink CSI prediction by approximately 6 dB.
KW - FDD massive MIMO
KW - channel state information
KW - complex-valued neural network
KW - partial channel reciprocity
KW - three-dimensional convolutional layer
UR - http://www.scopus.com/inward/record.url?scp=85101452923&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85101452923&partnerID=8YFLogxK
U2 - 10.1109/LWC.2020.3027774
DO - 10.1109/LWC.2020.3027774
M3 - Article
AN - SCOPUS:85101452923
VL - 10
SP - 266
EP - 270
JO - IEEE Wireless Communications Letters
JF - IEEE Wireless Communications Letters
SN - 2162-2337
IS - 2
M1 - 9209095
ER -