基于自注意力机制的大规模MIMO信道状态信息特征向量反馈方法

doi:10.11959/j.issn.1000-0801.2023247

Abstract

Abstract:

Massive multiple-input multiple-output (MIMO) system can provide satisfying gain of spectrum efficiency for 5G and future wireless communication systems.In frequency-division duplex (FDD) mode, downlink channel state information (CSI) needs to be accurately fed back to the base station side to obtain this gain.To improve the feedback accuracy of downlink CSI eigenvector, a self-attention mechanism-based CSI feedback method named SA-CsiNet was proposed.SA-CsiNet respectively deployed self-attention modules at the encoder and the decoder to achieve feature extraction and reconstruction of CSI.Experimental results show that compared with codebook-based and conventional deep learning-based CSI feedback approaches, SA-CsiNet provides higher reconstruction accuracy of CSI.

Key words: channel feedback, massive MIMO, self-attention mechanism, deep learning

CLC Number:

TN929.5

Bei YANG, Xin LIANG, Hang YIN, Zheng JIANG, Xiaoming SHE. Self-attention mechanism-based CSI eigenvector feedback for massive MIMO[J]. Telecommunications Science, 2023, 39(11): 128-136.

Figures/Tables 6

References 21

[1]	骆胜军, 张申科 . 5G大规模天线技术发展及创新方向[J]. 电信科学, 2020,36(11): 141-148.
	LUO S J , ZHANG S K . 5G large-scale array antenna technology development and innovation direction[J]. Telecommunications Science, 2020,36(11): 141-148.
[2]	王正强, 杨晓娜, 万晓榆 ,等. 大规模 MIMO 系统能效优化算法研究综述[J]. 重庆邮电大学学报(自然科学版), 2019,31(6): 743-752.
	WANG Z Q , YANG X N , WAN X Y ,et al. Energy efficiency optimization algorithm in massive MIMO systems:a survey[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2019,31(6): 743-752.
[3]	张冬雪, 杜洁汝, 何世彪 ,等. 面向大规模 MIMO 系统的下行预编码技术研究进展[J]. 电讯技术, 2023,63(2): 291-299.
	ZHANG D X , DU J R , HE S B ,et al. Research progress of downlink precoding techniques for massive MIMO systems[J]. Telecommunication Engineering, 2023,63(2): 291-299.
[4]	陈成瑞, 程港, 何世彪 ,等. 基于人工智能的大规模 MIMO信道状态信息反馈综述[J]. 电讯技术, 2021,61(9): 1181-1190.
	CHEN C R , CHENG G , HE S B ,et al. Survey of massive MIMO channel state information feedback based on artificial intelligence[J]. Telecommunication Engineering, 2021,61(9): 1181-1190.
[5]	ETSI. 5G; NR; Physical layer procedures for data (V16.1.0,Release 16):TS 38.214[S]. 2020.
[6]	毛攀, 黄小光, 汪伟 ,等. 基于压缩感知的大规模 MIMO 信道估计与反馈[J]. 电信科学, 2018,34(12): 46-52.
	MAO P , HUANG X G , WANG W ,et al. Massive MIMO channel estimation and feedback based on compressed sensing[J]. Telecommunications Science, 2018,34(12): 46-52.
[7]	黄秋萍, 刘晓峰, 高秋彬 ,等. 基于人工智能的大规模天线信道状态信息反馈研究[J]. 电信科学, 2022,38(3): 74-83.
	HUANG Q P , LIU X F , GAO Q B ,et al. Study of channel state information feedback based on artificial intelligence[J]. Telecommunications Science, 2022,38(3): 74-83.
[8]	Qualcomm. New SI:study on artificial intelligence (AI)/machine learning (ML) for NR air interface:RP-213599[R]. 2021.
[9]	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.
[10]	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 (ICC). Piscataway:IEEE Press, 2020: 1-6.
[11]	WANG T , WEN C K , JIN S ,et al. Deep learning-based CSI feedback approach for time-varying massive MIMO channels[J]. IEEE Wireless Communications Letters, 2018,8(2): 416-419.
[12]	GUO J , YANG X , WEN C K ,et al. DL-based CSI feedback and cooperative recovery in massive MIMO[J]. arXiv preprint arXiv:2003.03303, 2020.
[13]	CHEN T , GUO J J , JIN S ,et al. A novel quantization method for deep learning-based massive MIMO CSI feedback[C]// Proceedings of 2019 IEEE Global Conference on Signal and Information Processing (GlobalSIP). Piscataway:IEEE Press, 2019: 1-5.
[14]	CHANG H , LIANG X , LI H ,et al. Deep learning-based bitstream error correction for CSI feedback[J]. IEEE Wireless Communications Letters, 2021,10(12): 2828-2832.
[15]	VASWANI A , SHAZEER N , PARMAR N ,et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. New York:ACM Press, 2017: 6000-6010.
[16]	GUO J , WEN C K , JIN S ,et al. Convolutional neural network-based multiple-rate compressive sensing for massive MIMO CSI feedback:design,simulation,and analysis[J]. IEEE Transactions on Wireless Communications, 2020,19(4): 2827-2840.
[17]	KINGMA D P , BA J . Adam:a method for stochastic optimization[J]. arXiv preprint arXiv:1412.6980, 2014.
[18]	SHLEZINGER N , ELDAR Y C . Deep task-based quantization[J]. Entropy, 2021,23(1): 104.
[19]	GUO J , LI X , CHEN M ,et al. AI enabled wireless communications with real channel measurements:channel feedback[J]. Journal of Communications and Information Networks, 2020,5(3): 310-317.
[20]	LIU W , TIAN W , XIAO H ,et al. EVCsiNet:eigenvector-based CSI feedback under 3GPP link-level channels[J]. IEEE Wireless Communications Letters, 2021,10(12): 2688-2692.
[21]	ETSI. 5G; NR; Multiplexing and channel coding (V16.1.0;Release 16):3GPP TS 38.212[S]. 2020.

Metrics

Recommended 0

No Suggested Reading articles found!

参数	设定
场景	密集城区
中心频率	4.0 GHz
用户分布范围	200 m
带宽	10 MHz
子载波间隔	15 kHz
RB数	48
子带数	12
子带颗粒度	4 RB
天线配置	32根发送天线，4根接收天线，均匀线性阵列
信道模型	满足3GPP TS 38.901
Rank数	1
信道估计	具有理想信道估计

方法	反馈开销/bit	余弦相似度（ρ）
Enhanced Type II码本	49	0.793
EVCsiNet	48	0.741
CRNet	48	0.744
SA-CsiNet	48	0.946
Enhanced Type II码本	71	0.829
EVCsiNet	60	0.749
CRNet	60	0.754
SA-CsiNet	60	0.949
Enhanced Type II码本	130	0.882
EVCsiNet	120	0.792
CRNet	120	0.799
SA-CsiNet	120	0.959
Enhanced Type II码本	250	0.914
EVCsiNet	240	0.835
CRNet	240	0.854
SA-CsiNet	240	0.968

反馈开销(O=M×B)/bit	方法	M	B	余弦相似度（ρ）
48	EVCsiNet	24	2	0.741
		16	3	0.719
	CRNet	24	2	0.744
		16	3	0.728
	SA-CsiNet	24	2	0.946
		16	3	0.950
60	EVCsiNet	30	2	0.749
		20	3	0.727
	CRNet	30	2	0.754
		20	3	0.734
	SA-CsiNet	30	2	0.949
		20	3	0.954
120	EVCsiNet	60	2	0.792
		40	3	0.758
	CRNet	60	2	0.799
		40	3	0.769
	SA-CsiNet	60	2	0.959
		40	3	0.961
240	EVCsiNet	120	2	0.835
		80	3	0.817
	CRNet	120	2	0.854
		80	3	0.823
	SA-CsiNet	120	2	0.968
		80	3	0.970

方法	M	参数数量	运行时间/s
EVCsiNet	24	41 848	0.000 384
CRNet		43 000	0.000 950
SA-CsiNet		21 431 960	0.000 939
EVCsiNet	30	51 070	0.000 403
CRNet		52 222	0.001 022
SA-CsiNet		21 441 182	0.000 989
EVCsiNet	60	97 180	0.000 410
CRNet		98 332	0.000 983
SA-CsiNet		21 487 292	0.001 007
EVCsiNet	120	189 400	0.000 425
CRNet		190 552	0.000 834
SA-CsiNet		21 579 512	0.000 908

Self-attention mechanism-based CSI eigenvector feedback for massive MIMO

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 6

References 21

Related Articles 15

Metrics

Recommended 0

[1]	Zhen YE, Guoxiang WANG, Junfeng SONG, Haokun LIU, Tiansong LI. A fast block partitioning algorithm for VVC intra coding based on depthwise separable convolution [J]. Telecommunications Science, 2023, 39(7): 99-108.
[2]	Shao GUO, Peng PAN, Yaozong FAN. A low complexity pilot assignment algorithm based on user polar coordinates in CF-mMIMO systems [J]. Telecommunications Science, 2023, 39(7): 1-10.
[3]	Min LU, Juan HU, Xianchao ZHANG, Weijian DING, Guangxue YUE. Personalized recommendation model based on users multi-features fusion [J]. Telecommunications Science, 2023, 39(5): 101-115.
[4]	Bo HU, Anding WANG, Guiyi WEI. Low complexity hybrid precoding algorithm for massive MIMO based on modified Newton method [J]. Telecommunications Science, 2023, 39(11): 80-95.
[5]	Weichen HU, Congyuan XU, Yong ZHAN, Guanghui CHEN, Siqing LIU, Zhiqiang WANG, Xiaolin WANG. A network intrusion detection method designed for few-shot scenarios [J]. Telecommunications Science, 2023, 39(10): 85-100.
[6]	Hui WANG, Qixin TAI, Liu LIU, Hong WANG, Rongfang SONG. Low-power transmission method for uplink millimeter-wave massive MIMO-NOMA system based on GSIC [J]. Telecommunications Science, 2023, 39(1): 51-59.
[7]	Bin ZHUGE, Zhenghu YIN, Wenxue SI, Lei YAN, Ligang DONG, Xian JIANG. Student knowledge tracking based multi-indicator exercise recommendation algorithm [J]. Telecommunications Science, 2022, 38(9): 129-143.
[8]	Haiyan CAO, Zhongliang WANG, Hao XU, Qianhong CHEN, Fangmin XU. Energy efficiency joint optimization algorithm for mixed-ADC massive MIMO systems [J]. Telecommunications Science, 2022, 38(8): 65-74.
[9]	Danhao DENG, Chaowei WANG, Fan JIANG, Weidong WANG. Coordinated air-ground scheduling for UAV-assisted cell-free massive MIMO [J]. Telecommunications Science, 2022, 38(8): 37-44.
[10]	Jie ZHOU, Bernardo Esono Esono Mikue, Xueying WANG, Huiting ZHOU, Hong LUO. PAPR optimization based on SLM and PTS algorithms in NC-OFDM systems [J]. Telecommunications Science, 2022, 38(7): 63-74.
[11]	Yue WANG, Fangmin XU, Haiyan CAO. Pilot interference mitigation for TDD cell-free massive MIMO systems [J]. Telecommunications Science, 2022, 38(5): 87-94.
[12]	Qiuping HUANG, Xiaofeng LIU, Qiubin GAO, Zhengxuan LIU, Liqiang JIN, Shaohui SUN. Study of channel state information feedback based on artificial intelligence [J]. Telecommunications Science, 2022, 38(3): 74-83.
[13]	Qing SHEN, Wenbin GUO, Jungang LOU, Qiangguo YU. Personalized recommendation model with multi-level latent features [J]. Telecommunications Science, 2022, 38(2): 71-83.
[14]	Panpan LI, Zhengxia XIE, Guangxue YUE, Xin LIU. Research progress and trends of deep learning based wireless communication receiving method [J]. Telecommunications Science, 2022, 38(2): 1-17.
[15]	Mengting LOU, Jing JIN, Hanning WANG, Qixing WANG, Jiangzhou WANG. Research and evaluation of sparse array for massive MIMO [J]. Telecommunications Science, 2021, 37(9): 48-56.