Full Text:  <1261>

Summary:  <226>

CLC number: TN92

On-line Access: 2023-02-27

Received: 2022-04-26

Revision Accepted: 2023-02-27

Crosschecked: 2022-09-28

Cited: 0

Clicked: 1335

Citations:  Bibtex RefMan EndNote GB/T7714

 ORCID:

Zunwen HE

https://orcid.org/0000-0002-5584-4855

Yan ZHANG

https://orcid.org/0000-0002-2168-9674

-   Go to

Article info.
Open peer comments

Frontiers of Information Technology & Electronic Engineering 

Accepted manuscript available online (unedited version)


Ensemble-transfer-learning-based channel parameter prediction in asymmetric massive MIMO systems


Author(s):  Zunwen HE, Yue LI, Yan ZHANG, Wancheng ZHANG, Kaien ZHANG, Liu GUO, Haiming WANG

Affiliation(s):  School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China; more

Corresponding email(s):  hezunwen@bit.edu.cn, zhangy@bit.edu.cn

Key Words:  Asymmetric massive multiple-input multiple-output (MIMO) system; Channel model; Ensemble learning; Instance transfer; Parameter prediction


Share this article to: More <<< Previous Paper|Next Paper >>>

Zunwen HE, Yue LI, Yan ZHANG, Wancheng ZHANG, Kaien ZHANG, Liu GUO, Haiming WANG. Ensemble-transfer-learning-based channel parameter prediction in asymmetric massive MIMO systems[J]. Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/FITEE.2200169

@article{title="Ensemble-transfer-learning-based channel parameter prediction in asymmetric massive MIMO systems",
author="Zunwen HE, Yue LI, Yan ZHANG, Wancheng ZHANG, Kaien ZHANG, Liu GUO, Haiming WANG",
journal="Frontiers of Information Technology & Electronic Engineering",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/FITEE.2200169"
}

%0 Journal Article
%T Ensemble-transfer-learning-based channel parameter prediction in asymmetric massive MIMO systems
%A Zunwen HE
%A Yue LI
%A Yan ZHANG
%A Wancheng ZHANG
%A Kaien ZHANG
%A Liu GUO
%A Haiming WANG
%J Frontiers of Information Technology & Electronic Engineering
%P 275-288
%@ 2095-9184
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/FITEE.2200169"

TY - JOUR
T1 - Ensemble-transfer-learning-based channel parameter prediction in asymmetric massive MIMO systems
A1 - Zunwen HE
A1 - Yue LI
A1 - Yan ZHANG
A1 - Wancheng ZHANG
A1 - Kaien ZHANG
A1 - Liu GUO
A1 - Haiming WANG
J0 - Frontiers of Information Technology & Electronic Engineering
SP - 275
EP - 288
%@ 2095-9184
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/FITEE.2200169"


Abstract: 
Asymmetric massive multiple-input multiple-output (MIMO) systems have been proposed to reduce the burden of data processing and hardware cost in sixth-generation mobile networks (6G). However, in the asymmetric massive MIMO system, reciprocity between the uplink (UL) and downlink (DL) wireless channels is not valid. As a result, pilots are required to be sent by both the base station (BS) and user equipment (UE) to predict double-directional channels, which consumes more transmission and computational resources. In this paper we propose an ensemble-transfer-learning-based channel parameter prediction method for asymmetric massive MIMO systems. It can predict multiple DL channel parameters including path loss (PL), multipath number, delay spread (DS), and angular spread. Both the UL channel parameters and environment features are chosen to predict the DL parameters. Also, we propose a two-step feature selection algorithm based on the SHapley Additive exPlanations (SHAP) value and the minimum description length (MDL) criterion to reduce the computation complexity and negative impact on model accuracy caused by weakly correlated or uncorrelated features. In addition, the instance transfer method is introduced to support the prediction model in new propagation conditions, where it is difficult to collect enough training data in a short time. Simulation results show that the proposed method is more accurate than the back propagation neural network (BPNN) and the 3GPP TR 38.901 channel model. Additionally, the proposed instance-transfer-based method outperforms the method without transfer learning in predicting DL parameters when the beamwidth or the communication sector changes.

在非对称大规模MIMO系统中基于集成-迁移学习的信道参数预测

何遵文1,李悦1,张焱1,张万成1,张恺恩1,郭柳1,王海明2
1北京理工大学信息与电子学院,中国北京市,100081
2东南大学毫米波国家重点实验室,中国南京市,210096
摘要:为降低第六代移动网络中的数据处理负担和硬件成本,非对称大规模多入多出(multiple-input multiple-output,MIMO)系统被提出。然而,在非对称大规模MIMO系统中,上行和下行无线信道之间的互易性是无效的。因此,需要基站和用户设备都发送导频来预测双向信道,这会消耗更多传输和计算资源。本文提出一种基于集成迁移学习的非对称大规模MIMO系统的信道参数预测方法,可以预测多个下行信道参数,包括路径损耗、多径数、时延扩展和角度扩展。选择上行信道参数和环境特征来预测下行参数。此外,提出一种基于SHAP(SHapley Additive exPlanations)值和最小描述长度标准的两步特征选择算法,以降低由弱相关或不相关特征引起的计算复杂度和对模型准确性的负面影响。引入实例迁移方法,以支持预测模型应对在新的传播条件下难以在短时间内收集足够训练数据的问题。仿真结果表明,该方法比反向传播神经网络和3GPP TR 38.901信道模型更准确。当波束宽度或通信扇区发生变化时,所提出的基于实例迁移的方法在预测下行参数方面优于没有迁移学习的方法。

关键词组:非对称大规模MIMO系统;信道模型;集成学习;实例迁移;参数预测

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

[1]3GPP, 2020. Study on Channel Model for Frequencies from 0.5 to 100 GHz. TR 38.901 V16.1.0. 3rd Generation Partnership Project (3GPP), France.

[2]Albreem MA, Juntti M, Shahabuddin S, 2019. Massive MIMO detection techniques: a survey. IEEE Commun Surv Tut, 21(4):3109-3132.

[3]Albreem MA, Habbash AH, Abu-Hudrouss AM, et al., 2021. Overview of precoding techniques for massive MIMO. IEEE Access, 9:60764-60801.

[4]Bazzi A, Slock DTM, Meilhac L, 2016. Detection of the number of superimposed signals using modified MDL criterion: a random matrix approach. Proc IEEE Int Conf on Acoustics, Speech and Signal Processing, p.4593-4597.

[5]Chen TQ, Guestrin C, 2016. XGBoost: a scalable tree boosting system. Proc 22nd ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining, p.785-794.

[6]Chen YS, Lai FP, You JW, 2019. Analysis of antenna radiation characteristics using a hybrid ray tracing algorithm for indoor WiFi energy-harvesting rectennas. IEEE Access, 7:38833-38846.

[7]Erden F, Ozdemir O, Guvenc I, 2020. 28 GHz mmWave channel measurements and modeling in a library environment. Proc IEEE Radio and Wireless Symp, p.52-55.

[8]Gallieri M, Maciejowski JM, 2012. lasso MPC: smart regulation of over-actuated systems. Proc American Control Conf, p.1217-1222.

[9]Gao DY, Hu BQ, Qin FJ, et al., 2021. A real-time gravity compensation method for INS based on BPNN. IEEE Sens J, 21(12):13584-13593.

[10]Han Y, Li MY, Jin S, et al., 2020. Deep learning-based FDD non-stationary massive MIMO downlink channel reconstruction. IEEE J Sel Areas Commun, 38(9):1980-1993.

[11]Hong W, Zhou JY, Chen JX, et al., 2020. Asymmetric full-digital beamforming mmWave massive MIMO systems for B5G/6G wireless communications. Proc IEEE Asia-Pacific Microwave Conf, p.31-32.

[12]Hong W, Jiang ZH, Yu C, et al., 2021. The role of millimeter-wave technologies in 5G/6G wireless communications. IEEE J Microw, 1(1):101-13584.

[13]Hu WM, Hu W, Maybank S, 2008. AdaBoost-based algorithm for network intrusion detection. IEEE Trans Syst Man Cybern B Cybern, 38(2):577-583.

[14]Huang L, Long T, Mao EK, 2009. Robust estimation of the number of sources using an MMSE-based MDL method. Proc IET Int Radar Conf, p.1-4.

[15]Huang L, Fu QB, Li GF, et al., 2019. Improvement of maximum variance weight partitioning particle filter in urban computing and intelligence. IEEE Access, 7:106527-106535.

[16]Jiang T, Zhang JH, Tang P, et al., 2021. A study of uplink and downlink channel spatial characteristics in an urban micro scenario at 28 GHz. Front Inform Technol Electron Eng, 22(4):488-502.

[17]Joo J, Park MC, Han DS, et al., 2019. Deep learning-based channel prediction in realistic vehicular communications. IEEE Access, 7:27846-27858.

[18]Kim MD, Liang JY, Lee J, et al., 2016. Directional multipath propagation characteristics based on 28GHz outdoor channel measurements. Proc 10th European Conf on Antennas and Propagation, p.1-5.

[19]Li SD, Liu YJ, Chen ZP, et al., 2017. Measurements and modelling of millimeter-wave channel at 28 GHz in the indoor complex environment for 5G radio systems. Proc 9th Int Conf on Wireless Communications and Signal Processing, p.1-6.

[20]Lin B, Gao FF, Zhang S, et al., 2021. Deep learning-based antenna selection and CSI extrapolation in massive MIMO systems. IEEE Trans Wirel Commun, 20(11):7669-7681.

[21]Lin CH, Chi CY, Chen LL, et al., 2018. Detection of sources in non-negative blind source separation by minimum description length criterion. IEEE Trans Neur Netw Learn Syst, 29(9):4022-4037.

[22]Liu XM, Tang JS, 2014. Mass classification in mammograms using selected geometry and texture features, and a new SVM-based feature selection method. IEEE Syst J, 8(3):910-920.

[23]Lundberg SM, Lee SI, 2017. A unified approach to interpreting model predictions. Proc 31st Int Conf on Neural Information Processing Systems, p.4768-4777.

[24]Luo XR, Zhang Y, He ZW, et al., 2019. A two-step environment-learning-based method for optimal UAV deployment. IEEE Access, 7:149328-149340.

[25]Medeđović P, Veletić M, Blagojević Ž, 2012. Wireless insite software verification via analysis and comparison of simulation and measurement results. Proc 35th Int Convention MIPRO, p.776-781.

[26]Pan X, Yan CX, Zhang JK, 2019. Nonlinearity-based single-channel monopulse tracking method for OFDM-aided UAV A2G communications. IEEE Access, 7:148485-148494.

[27]Pardoe D, Stone P, 2010. Boosting for regression transfer. Proc 27th Int Conf on Machine Learning, p.863-870.

[28]Qiu JH, Xu K, Xia XC, et al., 2022. Secure transmission scheme based on fingerprint positioning in cell-free massive MIMO systems. IEEE Trans Signal Inform Process Netw, 8:92-105.

[29]Radhakrishnan V, Taghizadeh O, Mathar R, 2021. Impairments-aware resource allocation for FD massive MIMO relay networks: sum rate and delivery-time optimization perspectives. IEEE Trans Signal Inform Process Netw, 7:177-191.

[30]Sterba J, Kocur D, 2009. Pilot symbol aided channel estimation for OFDM system in frequency selective Rayleigh fading channel. Proc 19th Int Conf on Radioelektronika, p.77-80.

[31]Tuan DM, Cheon Y, Aoki Y, et al., 2016. Performance comparison of millimeter-wave communications system with different antenna beam-width. Proc 10th European Conf on Antennas and Propagation, p.1-5.

[32]Wang D, Nie FP, Huang H, 2015. Feature selection via global redundancy minimization. IEEE Trans Knowl Data Eng, 27(10):2743-2755.

[33]Wang DH, Yang Z, Yi Z, 2017. LightGBM: an effective miRNA classification method in breast cancer patients. Proc Int Conf on Computational Biology and Bioinformatics, p.7-11.

[34]Wang F, Liang JL, Wang ZD, et al., 2018. A variance-constrained approach to recursive filtering for nonlinear 2-D systems with measurement degradations. IEEE Trans Cybern, 48(6):1877-1887.

[35]Wang ZG, Xiao X, Rajasekaran S, 2020. Novel and efficient randomized algorithms for feature selection. Big Data Min Anal, 3(3):208-224.

[36]Wax M, Ziskind I, 1989. Detection of the number of coherent signals by the MDL principle. IEEE Trans Acoust Speech Signal Process, 37(8):1190-1196.

[37]Wimalajeewa T, Varshney PK, Su W, 2017. Detection of single versus multiple antenna transmission systems using pilot data. IEEE Trans Signal Inform Process Netw, 3(1):159-171.

[38]Wu XY, Zhang Y, Wang CX, et al., 2015. 28 GHz indoor channel measurements and modelling in laboratory environment using directional antennas. Proc 9th European Conf on Antennas and Propagation, p.1-5.

[39]Yang GS, Zhang Y, He ZW, et al., 2019. Machine-learning-based prediction methods for path loss and delay spread in air-to-ground millimetre-wave channels. IET Microw Antenn Propag, 13(8):1113-1121.

[40]Yang YW, Gao FF, Ma XL, et al., 2019. Deep learning-based channel estimation for doubly selective fading channels. IEEE Access, 7:36579-36589.

[41]Yao JG, Liu X, Zhu XY, et al., 2015. Control of large-scale systems through dimension reduction. IEEE Trans Serv Comput, 8(4):563-575.

[42]Ye H, Li GY, Juang BH, 2018. Power of deep learning for channel estimation and signal detection in OFDM systems. IEEE Wirel Commun Lett, 7(1):114-117.

[43]Zhang S, Liu YS, Gao FF, et al., 2021. Deep learning based channel extrapolation for large-scale antenna systems: opportunities, challenges and solutions. IEEE Wirel Commun, 28(6):160-167.

[44]Zhang SB, Zhang S, Gao FF, et al., 2021. Machine-learning-based prediction methods for path loss and delay spread in air-to-ground millimeter wave channels. IEEE Trans Commun, 69(10):6691-6705.

[45]Zhao SY, Huang B, Liu F, 2018. Localization of indoor mobile robot using minimum variance unbiased FIR filter. IEEE Trans Autom Sci Eng, 15(2):410-419.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - 2024 Journal of Zhejiang University-SCIENCE