A Unified Deep Learning Method for CSI Feedback in Massive MIMO Systems

doi:10.12142/ZTECOM.202204013

ZTE Communications ›› 2022, Vol. 20 ›› Issue (4): 110-115.DOI: 10.12142/ZTECOM.202204013

• Research Paper • Previous Articles Next Articles

A Unified Deep Learning Method for CSI Feedback in Massive MIMO Systems

GAO Zhengguang¹(), LI Lun¹, WU Hao¹, TU Xuezhen², HAN Bingtao¹

^1.State Key Laboratory of Mobile Network and Mobile Multimedia Technology, Shenzhen 518055, China
^2.The College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China

Received:2021-11-15 Online:2022-12-31 Published:2022-12-30
About author:GAO Zhengguang (gao.zhengguang@zte.com.cn) received his BS degree from Hubei Engineering University, China in 2013, MS degree from South China Normal University, China in 2016, and PhD degree from Beijing University of Posts and Telecommunications, China in 2020. In his doctor’s degree program, he was a visiting PhD student in High Performance Networks group, University of Bristol, UK from Nov. 1, 2018 to Nov. 1, 2019. After graduation, he was selected for “LAN JIAN” program of ZTE Corporation as an algorithm researcher. His current research interests include 5G/6G communication technologies, mobile networks, and machine learning for future communications.|LI Lun received his MS degree in electronics and communication engineering from Harbin Insititute of Technology, China in 2018. He joined ZTE Corporation, China in 2018, where he is currently a technical pre-research engineer. His research interests include artificial intelligence/machine learning for wireless communications.|WU Hao received his BS degree from Beijing University of Posts and Telecommunications, China in 2010, and PhD degree from Southeast University, China in 2015, both in electrical engineering. He is now with the State Key Laboratory of Mobile Network and Mobile Multimedia Technology, ZTE Corporation, where he is a senior expert on wireless communication research and standardization. During 2011–2012, he was a visiting student at Columbia University, USA. Since 2016, Dr. WU has been a delegate representing ZTE Corporation in 3GPP RAN and RAN1, to which he has submitted numerous contributions on 4G and 5G technologies including MIMO, UE power saving, positioning and so on. His research interests include MIMO wireless communications, antenna array systems, and signal processing.|TU Xuezhen is currently pursuing her master’s degree with the College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, China. She received her bachelor’s degree in computer science and technology from Henan University, China in 2020. Her research interest is mainly communication-efficient distributed learning.|HAN Bingtao received his BS degree from Tianjin University, China in 2001, and MS degree from Nankai University, China in 2004. He is the deputy director of the State Key Laboratory of Mobile Network and Mobile Multimedia Technology, and the leader for “Adlik” project of the LF AI & Data Foundation. Currently, he is the AI system architect of Central R&D Institute, ZTE Corporation. His current research interests include deep learning algorithms, AI systems, and network intelligence. He is the author and co-author for numerous patents and related monographs.

Abstract

Abstract:

A unified deep learning (DL) based algorithm is proposed for channel state information (CSI) compression in massive multiple-input multiple-output (MIMO) systems. More importantly, the element filling strategy is investigated to address the problem of model redesigning and retraining for different antenna typologies in practical systems. The results show that the proposed DL-based algorithm achieves better performance than the enhanced Type Ⅱ algorithm in Release 16 of 3GPP. The proposed element filling strategy enables one-time training of a unified model to compress and reconstruct different channel state matrices in a practical MIMO system.

Key words: deep learning, channel state information, element filling strategy

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.

Figures/Tables 5

Figure 1 Pipeline of ACRNet+ model for channel state information (CSI) compression

Figure 2 Proposed ACRNet+ architecture

Figure 3 GCS of ACRNet+ and eType Ⅱ for the scenario with 1×16 multiple-input multiple-output (MIMO)

Figure 4 GCS of ACRNet+ and eType Ⅱ for the scenario with 2×8 multiple-input multiple-output (MIMO)

Table 1 Comparison between ACRNet+ and ACRNetH under the same compression ratio

Dataset	Method	1/4		1/8		1/15
Dataset	Method	NMSE	GCS	NMSE	GCS	NMSE	GCS
Data1	ACRNet+	-7.22	0.887 9	-5.10	0.813 4	-3.865	0.746 2
Data1	ACRNetH	-6.59	0.870 0	-4.70	0.795 8	-3.44	0.722 8
Data2	ACRNet+	-14.10	0.976 5	-9.78	0.937 5	-6.61	0.875 4
Data2	ACRNetH	-12.38	0.966 1	-8.07	0.910 5	-5.82	0.850 9
Data3	ACRNet+	-12.62	0.969 8	-8.001	0.911 4	-5.678	0.844 9
Data3	ACRNetH	-9.84	0.943	-6.812	0.882 5	-4.795	0.810 5

References 26

1	LU L, LI G Y, SWINDLEHURST A L, et al. An overview of massive MIMO: benefits and challenges [J]. IEEE journal of selected topics in signal processing, 2014, 8(5): 742–758. DOI: 10.1109/JSTSP.2014.2317671 DOI
2	MARZETTA T L. Massive MIMO: an introduction [J]. Bell labs technical journal, 2015, 20: 11–22. DOI: 10.15325/BLTJ.2015.2407793 DOI
3	WU H Q. Ten reflections on 5G [J]. ZTE technology journal, 2020, 26(1): 2–4. DOI: 10.12142/ZTECOM.202001001 DOI
4	FANG M, DUAN X Y, HU L J. Challenges, innovations and perspectives towards 6G [J]. ZTE technology journal, 2020, 26(3): 61–70. DOI: 10.12142/ ZTETJ.202003012 DOI
5	WANG X Y. 5G: striving for sustainable growth amid expectations [J]. ZTE technology journal, 2020, 26(1): 64–66. DOI: 10.12142/ZTETJ.202001014 DOI
6	GAO Z, DAI L L, WANG Z C, et al. Spatially common sparsity based adaptive channel estimation and feedback for FDD massive MIMO [J]. IEEE transactions on signal processing, 2015, 63(23): 6169–6183. DOI: 10.1109/TSP.2015.2463260 DOI
7	KUO P H, KUNG H T, TING P G. Compressive sensing based channel feedback protocols for spatially-correlated massive antenna arrays [C]//Proceedings of 2012 IEEE Wireless Communications and Networking Conference. IEEE, 2012: 492–497. DOI: 10.1109/WCNC.2012.6214417 DOI
8	LU L, LI G Y, QIAO D L, et al. Sparsity-enhancing basis for compressive sensing based channel feedback in massive MIMO systems [C]//Proceedings of 2015 IEEE Global Communications Conference. IEEE, 2015: 1–6. DOI: 10.1109/GLOCOM.2015.7417036 DOI
9	DAUBECHIES I, DEFRISE M, DE MOL C. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint [J]. Communications on pure and applied mathematics, 2004, 57(11): 1413–1457. DOI: 10.1002/cpa.20042 DOI
10	KONG Q L, GONG R, LIU J T, et al. Investigation on reconstruction for frequency domain photoacoustic imaging via TVAL3 regularization algorithm [J]. IEEE photonics journal, 2018, 10(5): 1–15. DOI: 10.1109/JPHOT.2018.2869815 DOI
11	METZLER C A, MALEKI A, BARANIUK R G. From denoising to compressed sensing [J]. IEEE transactions on information theory, 2016, 62(9): 5117–5144. DOI: 10.1109/TIT.2016.2556683 DOI
12	GAO Z, YAN S, ZHANG J, et al. ANN-based multi-channel QoT-prediction over a 563.4 km field-trial testbed [J]. Journal of lightwave technology, 2020, 38 (9): 2646–2655
13	GAO Z G, ZHANG J W, YAN S Y, et al. Deep reinforcement learning for BBU placement and routing in C-RAN [C]//Proceedings of Optical Fiber Communication Conference (OFC). OSA, 2019: 1–3. DOI: 10.1364/ofc.2019.w2a.22 DOI
14	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 DOI
15	WANG T Q, WEN C K, JIN S, et al. Deep learning-based CSI feedback approach for time-varying massive MIMO channels [J]. IEEE wireless communications letters, 2019, 8(2): 416–419. DOI: 10.1109/LWC.2018.2874264 DOI
16	LIU F, HE X C, LI C G, et al. CsiNet-plus model with truncation and noise on CSI feedback [J]. IEICE transactions on fundamentals of electronics, communications and computer sciences, 2020, E103.A(1): 376–381. DOI: 10.1587/transfun.2019eal2123 DOI
17	LU Z L, WANG J T, SONG J. Multi-resolution CSI feedback with deep learning in massive MIMO system [C]//Proceedings of ICC 2020–2020 IEEE International Conference on Communications. IEEE, 2020: 1–6. DOI: 10.1109/ICC40277.2020.9149229 DOI
18	LU Z L, ZHANG X D, HE H Y, et al. Binarized aggregated network with quantization: flexible deep learning deployment for CSI feedback in massive MIMO system [EB/OL]. [2021-10-01]. . DOI: 10.1109/TWC.2022.3141653 DOI URL
19	LIU L F, OESTGES C, POUTANEN J, et al. The COST 2100 MIMO channel model [J]. IEEE wireless communications, 2012, 19(6): 92–99. DOI: 10.1109/mwc.2012.6393523 DOI
20	STUBER G L, BARRY J R, MCLAUGHLIN S W, et al. Broadband MIMO-OFDM wireless communications [J]. Proceedings of the IEEE, 2004, 92(2): 271–294. DOI: 10.1109/JPROC.2003.821912 DOI
21	RASHEED M H, SALIH O M, SIDDEQ M M, et al. Image compression based on 2D discrete Fourier transform and matrix minimization algorithm [EB/OL]. [2021-10-01].
22	WANG Y S, YAO H X, ZHAO S C. Auto-encoder based dimensionality reduction [J]. Neurocomputing, 2016, 184: 232–242. DOI: 10.1016/j.neucom.2015.08.104 DOI
23	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks [J]. Communications of the ACM, 2017, 60(6): 84–90. DOI: 10.1145/3065386 DOI
24	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition [C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2016: 770–778. DOI: 10.1109/CVPR.2016.90 DOI
25	WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module [C]//Proceedings of the European Conference on Computer Vision. ECCV, 2018: 3–19. DOI: 10.1007/978-3-030-01234-2_1 DOI
26	3GPP. Study on channel model for frequencies from 0.5 to 100 GHz: TR 38.901 [S]. 2017

[1]	DENG Letian, ZHAO Yanru. Deep Learning-Based Semantic Feature Extraction: A Literature Review and Future Directions [J]. ZTE Communications, 2023, 21(2): 11-17.
[2]	FAN Guotian, WANG Zhibin. Intelligent Antenna Attitude Parameters Measurement Based on Deep Learning SSD Model [J]. ZTE Communications, 2022, 20(S1): 36-43.
[3]	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.
[4]	HE Hongye, YANG Zhiguo, CHEN Xiangning. Payload Encoding Representation from Transformer for Encrypted Traffic Classification [J]. ZTE Communications, 2021, 19(4): 90-97.
[5]	CHEN Liangqin, TIAN Liping, XU Zhimeng, CHEN Zhizhang. A Survey of Wi-Fi Sensing Techniques with Channel State Information [J]. ZTE Communications, 2020, 18(3): 57-63.
[6]	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.
[7]	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.
[8]	ZHENG Xiaoqing, CHEN Jun, SHANG Guoqiang. Deep Neural Network-Based Chinese Semantic Role Labeling [J]. ZTE Communications, 2017, 15(S2): 58-64.
[9]	Biao He, Xiangyun Zhou, and Thushara D. Abhayapala. Wireless Physical Layer Security with Imperfect Channel State Information: A Survey [J]. ZTE Communications, 2013, 11(3): 11-19.

A Unified Deep Learning Method for CSI Feedback in Massive MIMO Systems

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 5

References 26

Related Articles 9

Recommended Articles

Metrics