Efficient Spatio-Temporal Predictive Learning for Massive MIMO CSI Prediction

doi:10.12142/ZTECOM.202501002

ZTE Communications ›› 2025, Vol. 23 ›› Issue (1): 3-10.DOI: 10.12142/ZTECOM.202501002

• Special Topic • Previous Articles Next Articles

Efficient Spatio-Temporal Predictive Learning for Massive MIMO CSI Prediction

CHENG Jiaming¹, CHEN Wei¹(), LI Lun²^,³, AI Bo¹

^1.School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China
^2.State Key Laboratory of Mobile Network and Mobile Multimedia Technology, Shenzhen 518055, China
^3.ZTE Corporation, Shenzhen 518057, China

Received:2025-02-23 Online:2025-03-25 Published:2025-03-25
About author:CHENG Jiaming received his BE degree from Beijing Jiaotong University, China in 2024, where he is currently pursuing his PhD degree. His current research interests include massive MIMO and intelligent communications.
CHEN Wei (weich@bjtu.edu.cn) received his BE and ME degrees from the Beijing University of Posts and Telecommunications, China in 2006 and 2009, respectively, and PhD degree in computer science from the University of Cambridge, UK in 2013. Later, he was a research associate with the Computer Laboratory, University of Cambridge, from 2013 to 2016. He is currently a professor with Beijing Jiaotong University, China. He has published over 130 articles and won several international awards. His current research interests include intelligent wireless communication systems and multimedia processing.
LI Lun received his MS degree in electronics and communication engineering from Harbin Institute of Technology, China in 2018. He joined ZTE Corporation in 2018, where he is currently a technical pre-research engineer. His research interests include artificial intelligence/machine learning for wireless communications.
AI Bo received his MS and PhD degrees from Xidian University, China in 2002 and 2004, respectively. He is currently a full professor with Beijing Jiaotong University, China. He has authored/coauthored eight books and published over 300 academic research articles. His research interests include the research and applications of channel measurement and channel modeling, and dedicated mobile communications for rail traffic systems. He has received many awards, such as the Distinguished Youth Foundation and the Excellent Youth Foundation from the National Natural Science Foundation of China, the Qiushi Outstanding Youth Award by Hong Kong Qiushi Foundation, the New Century Talents by the Chinese Ministry of Education, the Zhan Tianyou Railway Science and Technology Award by the Chinese Ministry of Railways, and the Science and Technology New Star Award by the Beijing Municipal Science and Technology Commission.
Supported by:
the Natural Science Foundation of China(U2468201);ZTE Industry?University?Institute Cooperation Funds(IA20240420002)

Abstract

Abstract:

Accurate channel state information (CSI) is crucial for 6G wireless communication systems to accommodate the growing demands of mobile broadband services. In massive multiple-input multiple-output (MIMO) systems, traditional CSI feedback approaches face challenges such as performance degradation due to feedback delay and channel aging caused by user mobility. To address these issues, we propose a novel spatio-temporal predictive network (STPNet) that jointly integrates CSI feedback and prediction modules. STPNet employs stacked Inception modules to learn the spatial correlation and temporal evolution of CSI, which captures both the local and the global spatio-temporal features. In addition, the signal-to-noise ratio (SNR) adaptive module is designed to adapt flexibly to diverse feedback channel conditions. Simulation results demonstrate that STPNet outperforms existing channel prediction methods under various channel conditions.

Key words: massive MIMO, deep learning, CSI prediction, CSI feedback

CHENG Jiaming, CHEN Wei, LI Lun, AI Bo. Efficient Spatio-Temporal Predictive Learning for Massive MIMO CSI Prediction[J]. ZTE Communications, 2025, 23(1): 3-10.

Figures/Tables 8

Figure 1 (a) Structure of AI-based CSI feedback; (b) Our proposed AI-based joint CSI feedback and prediction framework

Figure 2 Network architecture of STPNet

Figure 3 Architecture of SAM

Table 1 Simulation parameters

Parameter	Value
Channel type	3GPP CDL-C and UMa^[23]
Carrier frequency	2 GHz
Bandwidth	10 MHz
$N_{t}$	32
$N_{r}$	1
Number of subcarriers	32
Feedback interval	1 ms
UE speed	30 km/h

Table 1 Simulation parameters

Parameter	Value
Channel type	3GPP CDL-C and UMa^[23]
Carrier frequency	2 GHz
Bandwidth	10 MHz
$N_{t}$	32
$N_{r}$	1
Number of subcarriers	32
Feedback interval	1 ms
UE speed	30 km/h

Figure 4 A sample from the CDL-C channel model CSI dataset and the temporal autocorrelation

Figure 5 NMSE performance in the CDL-C channel model with CR=1/4 and 1/8

Figure 6 NMSE performance in the UMa channel model with CR=1/4

Figure 7 NMSE and achievable sum-rate performance of different architectures in the CDL-C channel model with CR=1/4

References 26

1	CHEN W S, LIN X Q, LEE J, et al. 5G-advanced toward 6G: past, present, and future [J]. IEEE journal on selected areas in communications, 2023, 41(6): 1592–1619. DOI: 10.1109/JSAC.2023.3274037
2	CHEN W, LIU Y W, JAFARKHANI H, et al. Signal processing and learning for next generation multiple access in 6G [J]. IEEE journal of selected topics in signal processing, 2024, 18(7): 1146–1177. DOI: 10.1109/JSTSP.2024.3511403
3	WEN C K, SHIH W T, JIN S. Deep learning for massive MIMO CSI feedback [J]. IEEE wireless communications letters, 2018, 7(5): 748–751. DOI: 10.1109/LWC.2018.2818160
4	GAO Y CHEN J J, LI D P. Intelligence driven wireless networks in B5G and 6G era: a survey [J]. ZTE communications, 2024, 22(3): 99–105. DOI: 10.12142/ZTECOM.202403012
5	YANG B LIANG X, LIU S N, et al. Intelligent 6G wireless network with multi-dimensional information perception [J]. ZTE communications, 2023, 21(2): 3–10. DOI: 10.12142/ZTECOM.202302002
6	GUO Y R, CHEN W, SUN F F, et al. Deep learning for CSI feedback: one-sided model and joint multi-module learning perspectives [EB/OL]. (2024-05-09)[2024-12-12].
7	CHENG J M, CHEN W, XU J L, et al. Swin Transformer-based CSI feedback for massive MIMO [C]//The 23rd International Conference on Communication Technology (ICCT). IEEE, 2023: 809–814. DOI: 10.1109/ICCT59356.2023.10419637
8	YI X P, YANG S, GESBERT D, et al. The degrees of freedom region of temporally correlated MIMO networks with delayed CSIT [J]. IEEE transactions on information theory, 2014, 60(1): 494–514. DOI: 10.1109/TIT.2013.2284500
9	MIHAI S, YAQOOB M, HUNG D V, et al. Digital twins: a survey on enabling technologies, challenges, trends and future prospects [J]. IEEE communications surveys & tutorials, 2022, 24(4): 2255–2291. DOI: 10.1109/COMST.2022.3208773
10	TAN J SHA X B, DAI B, et al. Analysis of industrial Internet of Things and digital twins [J]. ZTE communications, 2021, 19(2): 53–60. DOI: 10.12142/ZTECOM.202102007
11	YIN H F, WANG H Q, LIU Y Z, et al. Addressing the curse of mobility in massive MIMO with prony-based angular-delay domain channel predictions [J]. IEEE journal on selected areas in communications, 2020, 38(12): 2903–2917. DOI: 10.1109/JSAC.2020.3005473
12	BADDOUR K E, BEAULIEU N C. Autoregressive modeling for fading channel simulation [J]. IEEE transactions on wireless communications, 2005, 4(4): 1650–1662. DOI: 10.1109/TWC.2005.850327
13	JIANG W, SCHOTTEN H D. Neural network-based fading channel prediction: A comprehensive overview [J]. IEEE access, 2019, 7: 118112–118124. DOI: 10.1109/ACCESS.2019.2937588
14	JIANG W, SCHOTTEN H D. Deep learning for fading channel prediction [J]. IEEE open journal of the communications society, 2020, 1: 320–332. DOI: 10.1109/OJCOMS.2020.2982513
15	JIANG H, CUI M Y, NG D W K, et al. Accurate channel prediction based on transformer: making mobility negligible [J]. IEEE journal on selected areas in communications, 2022, 40(9): 2717–2732. DOI: 10.1109/JSAC.2022.3191334
16	REN Z Z, ZHANG X D, WANG J T. Joint CSI feedback and prediction with deep learning in high-speed scenarios [C]//Proceedings of IEEE/CIC International Conference on Communications in China (ICCC). IEEE, 2024: 1910–1915. DOI: 10.1109/ICCC62479.2024.10681972
17	SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions [C]//Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2015: 1–9. DOI: 10.1109/CVPR.2015.7298594
18	GAO Z Y, TAN C, WU L R, et al. SimVP: Simpler yet better video prediction [C]//Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2022: 3160–3170. DOI: 10.1109/CVPR52688.2022.00317
19	XU J L, AI B, CHEN W, et al. Wireless image transmission using deep source channel coding with attention modules [J]. IEEE transactions on circuits and systems for video technology, 2022, 32(4): 2315–2328. DOI: 10.1109/TCSVT.2021.3082521
20	XU J L, AI B, WANG N, et al. Deep joint source-channel coding for CSI feedback: an end-to-end approach [J]. IEEE journal on selected areas in communications, 2023, 41(1): 260–273. DOI: 10.1109/JSAC.2022.3221963
21	YANG K, WANG S X, DAI J C, et al. WITT: a wireless image transmission transformer for semantic communications [C]//IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2023: 1–5. DOI: 10.1109/ICASSP49357.2023.10094735
22	DENG L T ZHAO Y R. Deep learning-based semantic feature extraction: a literature review and future directions [J]. ZTE communications, 2023, 21(2): 11–17. DOI: 10.12142/ZTECOM.202302003
23	3GPP. Study on channel model for frequencies from 0.5 to 100 GHz: TR 38.901 V18.0.0 [S]. 2024
24	WU C, YI X P, ZHU Y M, et al. Channel prediction in high-mobility massive MIMO: from spatio-temporal autoregression to deep learning [J]. IEEE journal on selected areas in communications, 2021, 39(7): 1915–1930. DOI: 10.1109/JSAC.2021.3078503
25	YUAN J D, NGO H Q, MATTHAIOU M. Machine learning-based channel prediction in massive MIMO with channel aging [J]. IEEE transactions on wireless communications, 2020, 19(5): 2960–2973. DOI: 10.1109/TWC.2020.2969627
26	JAECKEL S, RASCHKOWSKI L, BÖRNER K, et al. Quasi deterministic radio channel generator, user manual and documentation [R]. Berlin, Germany: QuaDRiGa, 2021

[1]	TANG Chenyue, LI Zeshen, CHEN Zihan, YANG Howard H.. Device Activity Detection and Channel Estimation Using Score-Based Generative Models in Massive MIMO [J]. ZTE Communications, 2025, 23(1): 53-62.
[2]	WANG Chongchong, LI Yao, WANG Beibei, CAO Hong, ZHANG Yanyong. Point Cloud Processing Methods for 3D Point Cloud Detection Tasks [J]. ZTE Communications, 2023, 21(4): 38-46.
[3]	GONG Panyin, ZHANG Guidong, ZHANG Zhigang, CHEN Xiao, DING Xuan. Research on Fall Detection System Based on Commercial Wi-Fi Devices [J]. ZTE Communications, 2023, 21(4): 60-68.
[4]	XIE Xinyu, WU Yongpeng, YUAN Zhifeng, MA Yihua. Massive Unsourced Random Access Under Carrier Frequency Offset [J]. ZTE Communications, 2023, 21(3): 45-53.
[5]	DENG Letian, ZHAO Yanru. Deep Learning-Based Semantic Feature Extraction: A Literature Review and Future Directions [J]. ZTE Communications, 2023, 21(2): 11-17.
[6]	LU Ping, SHENG Bin, SHI Wenzhe. Scene Visual Perception and AR Navigation Applications [J]. ZTE Communications, 2023, 21(1): 81-88.
[7]	FAN Guotian, WANG Zhibin. Intelligent Antenna Attitude Parameters Measurement Based on Deep Learning SSD Model [J]. ZTE Communications, 2022, 20(S1): 36-43.
[8]	GAO Zhengguang, LI Lun, WU Hao, TU Xuezhen, HAN Bingtao. A Unified Deep Learning Method for CSI Feedback in Massive MIMO Systems [J]. ZTE Communications, 2022, 20(4): 110-115.
[9]	ZHANG Jintao, HE Zhenqing, RUI Hua, XU Xiaojing. Spectrum Sensing for OFDMA Using Multicarrier Covariance Matrix Aware CNN [J]. ZTE Communications, 2022, 20(3): 61-69.
[10]	SHAO Zhichao, YAN Wenjing, YUAN Xiaojun. Markovian Cascaded Channel Estimation for RIS Aided Massive MIMO Using 1‑Bit ADCs and Oversampling [J]. ZTE Communications, 2022, 20(1): 48-56.
[11]	HE Hongye, YANG Zhiguo, CHEN Xiangning. Payload Encoding Representation from Transformer for Encrypted Traffic Classification [J]. ZTE Communications, 2021, 19(4): 90-97.
[12]	XUE Songyan, LI Ang, WANG Jinfei, YI Na, MA Yi, Rahim TAFAZOLLI, Terence DODGSON. To Learn or Not to Learn:Deep Learning Assisted Wireless Modem Design [J]. ZTE Communications, 2019, 17(4): 3-11.
[13]	ZHENG Xiaoqing, LU Yaping, PENG Haoyuan, FENG Jiangtao, ZHOU Yi, JIANG Min, MA Li, ZHANG Ji, JI Jie. Detecting Abnormal Start-Ups, Unusual Resource Consumptions of the Smart Phone: A Deep Learning Approach [J]. ZTE Communications, 2019, 17(2): 38-43.
[14]	ZHENG Xiaoqing, CHEN Jun, SHANG Guoqiang. Deep Neural Network-Based Chinese Semantic Role Labeling [J]. ZTE Communications, 2017, 15(S2): 58-64.
[15]	Stefano Buzzi, Carmen D’Andrea. Massive MIMO 5G Cellular Networks: mm-Wave vs. μ-Wave Frequencies [J]. ZTE Communications, 2017, 15(S1): 41-49.

Efficient Spatio-Temporal Predictive Learning for Massive MIMO CSI Prediction

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 26

Related Articles 15

Recommended Articles 0

Metrics