基于卡尔曼滤波的极限学习机在线盲均衡算法

doi:10.12305/j.issn.1001-506X.2021.03.04

摘要/Abstract

摘要：

针对正交振幅调制(quadrature amplitude modulation, QAM)信号, 在预测方法的盲均衡框架下, 基于卡尔曼滤波(Kalman filter, KF)提出了一种新的神经网络在线盲均衡算法。采用复数型极限学习机(complex extreme learning machine, C-ELM)作为非线性预测滤波器(prediction filter, PF), 用KF实时更新C-ELM的输出权值以使预测误差达到最小, 再通过自动增益装置调整信号的幅度变化, 最后引入相位调整因子纠正信号的相位旋转。仿真结果表明, 所提算法实现了良好的实时均衡效果, 具有较快的收敛速率和较小的稳态均方误差, 不仅适用于方形, 同时也适用于十字形QAM信号的盲均衡。

关键词: 盲均衡, 预测方法, 复数型极限学习机, 卡尔曼滤波

Abstract:

For quadrature amplitude modulation (QAM) signals, a new neural network online blind equalization algorithm based on the Kalman filter (KF) is proposed under the blind equalization framework of the prediction method. For the purpose of minimizing the prediction error, this paper adopts the complex extreme learning machine(C-ELM) as the nonlinear prediction filter(PF) and sequentially updates the output weights of C-ELM using the KF.The amplitude of the signal is then adjusted by an automatic gain control device, and finally the phase rotation problem is corrected through a phase tuning factor. Simulation results show that the proposed algorithm achieves a good real-time equalization effect and has a faster convergence speed and lower steady mean square error. Moreover, the algorithm is suitable for blind equalization of the square, as well as the cross QAM signals.

Key words: blind equalization, prediction method, complex extreme learning machine, Kalman filter

中图分类号:

杨凌, 程丽, 韩琴, 赵傲男. 基于卡尔曼滤波的极限学习机在线盲均衡算法[J]. 系统工程与电子技术, 2021, 43(3): 623-630.

Ling YANG, Li CHENG, Qin HAN, Aonan ZHAO. Online blind equalization algorithm using extreme learning machine based on Kalman filter[J]. Systems Engineering and Electronics, 2021, 43(3): 623-630.

图/表 13

图1

图2

表1

表2

表3

表4

表5

表6

图3

表7

图4

图5

图6

参考文献 25

1	SATO Y . Two extensional applications of the zero-forcing equa-lization method[J]. IEEE Trans.on Communications, 1975, 23 (6): 684- 687. doi: 10.1109/TCOM.1975.1092850
2	李进, 冯大政, 刘文娟. 快速QAM信号多模盲均衡算法[J]. 电子与信息学报, 2013, 35 (2): 273- 279.
	LI J , FENG D Z , LIU W J . A fast multimodulus blind equalization algorithm for QAM signal[J]. Journal of Electronics & Information Technology, 2013, 35 (2): 273- 279.
3	马思扬, 王彬, 彭华. L0-范数约束的稀疏多径信道RLS常模盲均衡算法[J]. 电子学报, 2017, 45 (10): 2561- 2568. doi: 10.3969/j.issn.0372-2112.2017.10.035
	MA S Y , WANG B , PENG H . L0-norm constraint RLS constant modulus algorithm for sparse channel equalization[J]. Acta Electronica Sinica, 2017, 45 (10): 2561- 2568. doi: 10.3969/j.issn.0372-2112.2017.10.035
4	杨凌, 陈亮, 赵膑, 等. 基于复数支持向量回归机的盲均衡算法[J]. 通信学报, 2019, 40 (10): 180- 188.
	YANG L , CHEN L , ZHAO B , et al. Blind equalization algorithm based on complex support vector regression[J]. Journal on Communications, 2019, 40 (10): 180- 188.
5	JAGANATHAN K , HASSIBI B . Reconstruction of signals from their autocorrelation and cross-correlation vectors, with applications to phase retrieval and blind channel estimation[J]. IEEE Trans.on Signal Processing, 2019, 67 (11): 2937- 2946. doi: 10.1109/TSP.2019.2911254
6	ZHENG F C , MCLAUGHLIN S , MULGREW B . Blind equalization of nonminimum phase channels: higher order cumulant based algorithm[J]. IEEE Trans.on Signal Processing, 1993, 41 (2): 681- 691. doi: 10.1109/78.193209
7	SLOCK D T M. Blind fractionally-spaced equalization, perfect-reconstruction filter banks and multichannel linear prediction[C]//Proc.of the IEEE International Conference on Acoustics, Speech, & Signal Processing, 1994.
8	FERRARI R, PANAZIO C M, ATTUX R R F, et al. Unsupervised channel equalization using fuzzy prediction-error filters[C]//Proc.of the IEEE Workshop on Neural Networks for Signal Processing, 2003.
9	CAVALCANTE C C, MONTALVAO J R F, DORIZZI B, et al. A neural predictor for blind equalization of digital communication systems: is it plausible?[C]//Proc. of the IEEE Signal Processing Society Workshop, 2000.
10	WADA C, CONSOLARO D M, FERRARI R, et al. Nonlinear blind source deconvolution using recurrent prediction-error filters and an artificial immune system[C]//Proc.of the 8th International Conference on Independent Component Analysis and Signal Separation, 2009.
11	CHEN L , YANG L , DU J , et al. An extreme learning machine architecture based on volterra filtering and PCA[J]. IEICE Trans.on Information and Systems, 2017, E100-D (11): 2690- 2701. doi: 10.1587/transinf.2017EDP7089
12	HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: a new learning scheme of feedforward neural networks[C]//Proc.of the IEEE International Joint Conference on Neural Networks, 2004.
13	LIANG N Y , HUANG G B , SARATCHANDRAN P , et al. A fast and accurate online sequential learning algorithm for feedforward networks[J]. IEEE Trans.on Neural Networks, 2006, 17 (6): 1411- 1423. doi: 10.1109/TNN.2006.880583
14	FOUCART T . Stability of the inverse correlation matrix partial ridge regression[J]. Journal of Statistical Planning and Infe-rence, 1999, 77 (1): 141- 154. doi: 10.1016/S0378-3758(98)00195-5
15	HUYNH H T , WON Y . Regularized online sequential learning algorithm for single-hidden layer feed forward neural networks[J]. Pattern Recognition Letters, 2011, 32 (14): 1930- 1935. doi: 10.1016/j.patrec.2011.07.016
16	LUO X , YANG X , JIANG C , et al. Timeliness online regula-rized extreme learning machine[J]. International Journal of Machine Learning and Cybernetics, 2018, 9, 465- 476. doi: 10.1007/s13042-016-0544-9
17	ZHANG X , MA H . Data-driven model-free adaptive control based on error minimized regularized online sequential extreme learning machine[J]. Energies, 2019, 12, 3241. doi: 10.3390/en12173241
18	COSMO D L , SALLES E O T . Multiple sequential regularized extreme learning machines for single image super resolution[J]. IEEE Signal Processing Letters, 2019,
19	GU Y , LIU J , CHEN Y , et al. TOSELM: timeliness online sequential extreme learning machine[J]. Neurocomputing, 2014, 128, 119- 127. doi: 10.1016/j.neucom.2013.02.047
20	WATSON P K . Kalman filtering as an alternative to ordinary least squares-some theoretical considerations and empirical results[J]. Empirical Economics, 1983, 8 (2): 71- 85. doi: 10.1007/BF01973191
21	杨凌, 韩琴, 程丽, 等. 基于预测原理的回声状态网络在线盲均衡算法[J]. 通信学报, 2020, 41 (3): 145- 153.
	YANG L , HAN Q , CHENG L , et al. Online blind equalization algorithm based on prediction principle[J]. Journal on Communications, 2020, 41 (3): 145- 153.
22	HUANG G B , ZHU Q Y , SIEW C K . Extreme learning machine: theory and applications[J]. Neurocomputing, 2006, 70 (1/3): 489- 501.
23	CHU Y H , LIN H F , YANG L , et al. Discriminative globality-locality preserving extreme learning machine for image classification[J]. Neurocomputing, 2019, doi: 10.1016/j.neucom.2019.09.013
24	LI M B , HUANG G B , SARATCHANDRAN P , et al. Fully complex extreme learning machine[J]. Neurocomputing, 2005, 68, 306- 314. doi: 10.1016/j.neucom.2005.03.002
25	唐成凯, 张玲玲, 廉保旺. 卫星高阶调制信号通信下非线性误差修正均衡方法[J]. 通信学报, 2017, 38 (1): 117- 125.
	TANG C K , ZHANG L L , LIAN B W . Nonlinear error modified equalization algorithm in high-order modulation of satellite communication[J]. Journal of Communications, 2017, 381 (1): 117- 125.

隐含层节点数	SNR/dB
隐含层节点数	10	15	20	25	30	35
5	0.6	-4.0	-8.2	-11.4	-12.8	-13.5
15	0.5	-3.9	-7.9	-11.1	-12.5	-13.1
50	0.7	-3.8	-7.8	-11.0	-11.9	-12.5

隐含层节点数	迭代次数/次
5	2 500~2 600
15	2 500~2 600
50	2 500~2 600

隐层激活函数	SNR/dB
隐层激活函数	10	15	20	25	30	35
arctan	0.7	-4.0	-8.1	-11.4	-12.9	-13.5
sin	0.8	-4.0	-8.1	-11.3	-12.8	-13.3
tanh	0.6	-3.8	-8.0	-11.2	-12.5	-13.2

隐层激活函数g(·)	迭代次数/次
arctan	2 500~2 600
sin	3 100~3 200
tanh	2 500~2 600

输出函数f_readout(·)	SNR/dB
输出函数f_readout(·)	10	15	20	25	30	35
线性	0.6	-4.0	-8.2	-11.4	-12.9	-13.5
非线性	0.5	-7.8	-19.0	-25.5	-28.5	-28.8