第五代(5G)无线通信网络采用的大规模多输入多输出(MIMO)技术需要大量空口资源估计和反馈MIMO信道。除优化导频、估计和反馈设计外，对信道衰落的预测也是节约空口资源的有效途径。运用相空间重构方法研究三维信道模型相关的相空间重构参数，提出一种基于经验知识的小样本在线学习方法，对MIMO信道系数和信道容量进行预测。研究发现无线信道数据具有混沌性，相空间延迟时间和嵌入维数服从一定分布，因此可以作为实时预测的先验参数进行设定。实验结果表明，该方法预测效率比传统ARMA方法提升6倍左右，信道容量的平均误差最小为5.91%。最后，采用某市区的实测数据验证相空间重构方法的有效性，信道容量的预测平均误差最小为0.91%。
The fifth generation (5G) wireless communication network uses multiple input multiple output (MIMO) technology, which requires a lot of air interface resource estimation and feedback MIMO channel. Besides optimizing pilot, estimation and feedback design, channel fading prediction is also an effective way to save air interface resources. In this paper, phase space reconstruction method is used to study the phase space reconstruction parameters related to three-dimensional channel model, and a small sample online learning method based on empirical knowledge is proposed to predict MIMO channel coefficients and channel capacity. It is found that the wireless channel data is chaotic, and the phase space delay time and embedding dimension obey a certain distribution, so it can be set as the prior parameters of real-time prediction. Experimental results show that the prediction efficiency of the proposed method is about six times higher than that of the traditional ARMA method, and the minimum average error of channel capacity is 5.91%. Finally, the effectiveness of the phase space reconstruction method is verified by the measured data of an urban area, and the minimum average error of channel capacity prediction is 0.91%.
[1] Chen C M, Volski V, van der Perre L, et al. Finite large antenna arrays for massive MIMO: characterization and system impact[J]. IEEE Transactions on Antennas and Propagation, 2017, 65(12): 6712-6720.DOI:10.1109/TAP.2017.2754444.
[2] Hassan M E, Falou A E, Langlais C. Performance assessment of linear precoding for multi-user massive MIMO systems on a realistic 5G mmWave channel[C]//2018 IEEE Middle East and North Africa Communications Conference (MENACOMM). April 18-20, 2018, Jounieh, Lebanon. IEEE, 2018:1-5.DOI:10.1109/MENACOMM.2018.8371025.
[3] Gao Y, Vinck H, Kaiser T. Massive MIMO antenna selection: switching architectures, capacity bounds, and optimal antenna selection algorithms[J]. IEEE Transactions on Signal Processing, 2018, 66(5): 1346-1360.DOI:10.1109/TSP.2017.2786220.
[4] Björnson E, Hoydis J, Sanguinetti L. Massive MIMO has unlimited capacity[J]. IEEE Transactions on Wireless Communications, 2018, 17(1): 574-590.DOI:10.1109/TWC.2017.2768423.
[5] Duel-Hallen A. Fading channel prediction for mobile radio adaptive transmission systems[J]. Proceedings of the IEEE, 2007, 95(12): 2299-2313.DOI:10.1109/JPRDC.2007.904443.
[6] Zhao J W, Xie H X, Gao F F, et al. Time varying channel tracking with spatial and temporal BEM for massive MIMO systems[J]. IEEE Transactions on Wireless Communications, 2018, 17(8): 5653-5666.DOI:10.1109/TWC.2018.2848259.
[7] He P, Yuan Y, Liu G. Web services quality prediction based on multivariate time series analysis[C]//2018 IEEE 9th International Conference on Software Engineering and Service Science (ICSESS), November 23-25, 2018, Beijing, China. IEEE, 2018: 881-884.DOI:10.1109/ICSESS.2018.8663771.
[8] Miao T. Research of regional drought forecasting based on phase space reconstruction and wavelet neural network model[C]//2018 7th International Conference on Agro-geoinformatics (Agro-geoinformatics). August 6-9, 2018, Hangzhou, China. IEEE, 2018: 1-4.DOI:10.1109/Agro-Geoinformatics.2018.8475999.
[9] Han Y J, Liu J. The online forecasting research of short-term wind speed and power generation at wind farm based on phase space reconstruction[C]//2015 Seventh International Conference on Measuring Technology and Mechatronics Automation. June 13-14, 2015, Nanchang, China. IEEE, 2015: 1234-1237.DOI:10.1109/ICMTMA.2015.300.
[10] Guo Z Q, Chen L J, Liu P. Short-term stock forecasting based on phase space reconstruction and cluster analysis[C]//2019 11th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC). August 24-25, 2019, Hangzhou, China. IEEE, 2019, 2:50-53.DOI:10.1109/IHMSC.2019.10107.
[11] Wang B Y, Zheng W X. Blind adaptive channel identification/equalization in chaotic communications by using nonlinear prediction technique[C]//Proceedings 7th International Conference on Signal Processing, 2014. Proceedings. ICSP '04.2004. August 31-September 4, 2004, Beijing, China. IEEE, 2004: 372-375.DOI:10.1109/ICDSP.2004.1452659.
[12] Zhou Y T, Wang R, Xia K W. Nonlinear prediction of fast fading channel based on minimax probability machine[C]//2011 6th IEEE Conference on Industrial Electronics and Applications. June 21-23, 2011, Beijing, China. IEEE, 2011: 451-454.DOI:10.1109/ICIEA.2011.5975626.
[13] Jiang W, Schotten H D. A comparison of wireless channel predictors: artificial intelligence versus Kalman filter[C]//ICC 2019—2019 IEEE International Conference on Communications (ICC). May 20-24, 2019, Shanghai, China. IEEE, 2019: 1-6.DOI:10.1109/ICC.2019.8761308.
[14] Nagashima R, Ohtsuki T, Jiang W J, et al. Channel prediction for massive MIMO with channel compression based on principal component analysis[C]//2016 IEEE 27th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC). September 4-8, 2016, Valencia, Spain. IEEE, 2016: 1-6.DOI:10.1109/PIMRC.2016.7794949.
[15] Gu S S, Jiao J, Huang Z X, et al. ARMA-based adaptive coding transmission over millimeter-wave channel for integrated satellite-terrestrial networks[J]. IEEE Access, 2018, 6: 21635-21645.DOI:10.1109/ACCESS.2018.2825256.
[16] Yang Y, Li Y, Zhang W X, et al. Generative-adversarial-network-based wireless channel modeling: challenges and opportunities[J]. IEEE Communications Magazine, 2019, 57(3): 22-27.DOI:10.1109/MCOM.2019.1800635.
[17] Huang J, Wang C X, Bai L, et al. A big data enabled channel model for 5G wireless communication systems[J]. IEEE Transactions on Big Data, 2020, 6(2): 211-222.DOI:10.1109/TBDATA.2018.2884489.
[18] Abarbanel H D I, Brown R, Kadtke J B. Prediction in chaotic nonlinear systems: methods for time series with broadband Fourier spectra[J]. Physical Review A, 1990, 41(4): 1782.
[19] Rosenstein M T, Collins J J, de Luca C J. A practical method for calculating largest Lyapunov exponents from small data sets[J]. Physica D: Nonlinear Phenomena, 1993,65(1/2):117-134.DOI:10.1016/0167-2789(93)90009-P.
[20] Abarbanel H D I, Brown R, Sidorowich J J, et al. The analysis of observed chaotic data in physical systems[J]. Reviews of Modern Physics, 1993, 65(4): 1331-1392.DOI:10.1103/REVMODPHYS.65.1331.
[21] Fraser A M, Swinney H L. Independent coordinates for strange attractors from mutual information[J]. Physical Review. A, General Physics, 1986, 33(2): 1134-1140.DOI:10.1103/physreva.33.1134.
[22] Ng W W, Panu U S, Lennox W C. Chaos based Analytical techniques for daily extreme hydrological observations[J]. Journal of Hydrology, 2007, 342(1/2): 17-41.DOI:10.1016/j.jhydrol.2007.04.023.
[23] Takens F. Detecting strange attractors in turbulence[M]//Lecture Notes in Mathematics. Berlin, Heidelberg: Springer Berlin Heidelberg, 1981：366-381.
[24] Abarbanel H D I, Brown R, Kadtke J B. Prediction and system identification in chaotic nonlinear systems: time series with broadband spectra[J]. Physics Letters A, 1989, 138(8):401-408.DOI:10.1016/0375-9601(89)90839-6.
