A Basis Function Generation Based Digital Predistortion Concurrent Neural Network Model for RF Power Amplifiers

doi:10.12142/ZTECOM.202501009

ZTE Communications ›› 2025, Vol. 23 ›› Issue (1): 71-77.DOI: 10.12142/ZTECOM.202501009

• Research Papers • 上一篇下一篇

收稿日期:2023-09-23 出版日期:2025-03-25 发布日期:2025-03-25

A Basis Function Generation Based Digital Predistortion Concurrent Neural Network Model for RF Power Amplifiers

SHAO Jianfeng¹(), HONG Xi¹, WANG Wenjie¹, LIN Zeyu², LI Yunhua²

^1.Xi’an Jiaotong University, Xi’an 710049, China
^2.ZTE Corporation, Shenzhen 518057, China

Received:2023-09-23 Online:2025-03-25 Published:2025-03-25
About author:SHAO Jianfeng (sjf1996717@stu.xjtu.edu.cn) received his BS degree in electronic information engineering from the School of Information Engineering, North China University of Water Resources and Electric Power in 2018, MS degree in circuits and systems from the School of Physics and Electronic Engineering, Ningxia University, China in 2021. He is pursuing a PhD degree in information and communication engineering at Xi'an Jiaotong University, China. His research interests include array signal processing and digital pre-distortion.
HONG Xi received his BS degree in information engineering from the School of Electronics and Information Engineering, Xi'an Jiaotong University, China in 2012, and PhD degree in information and communication engineering from the School of Information and Communication Engineering, Xi'an Jiaotong University in 2022. He is currently an engineer at the School of Information and Communication Engineering, Xi'an Jiaotong University. His research interests include array signal processing, signal processing in communication systems, multipath mitigation in navigation, and GNSS physical layer interference detection.
WANG Wenjie received his BS, MS, and PhD degrees in information and communication engineering from Xi'an Jiaotong University, China in 1993, 1998, and 2001, respectively, where he is currently a professor. His main research interests include information theory, broadband wireless communications, signal processing in communication systems, and array signal processing.
LIN Zeyu received his BS and MS degrees at the School of Information and Communication Engineering, Xi'an Jiaotong University, China in 2018 and 2021, respectively. He is currently a senior RF algorithm architect in the RHP department of ZTE Corporation. His research interests include PA nonlinear system behavioral modeling and RF systems linearization.
LI Yunhua received his BE degree in communications engineering from Shandong University (Weihai), China in 2012 and PhD degree in military communications from the School of Telecommunications Engineering, Xidian University, China in 2017. He is currently a senior engineer of RF algorithms in ZTE Corporation. His research interests include communication signal processing, wireless communications, digital predistortion modeling of nonlinear systems, and more.
Supported by:
ZTE Industry?University?Institute Cooperation Funds(HC?CN?20220722010)

摘要/Abstract

Abstract:

This paper proposes a concurrent neural network model to mitigate non-linear distortion in power amplifiers using a basis function generation approach. The model is designed using polynomial expansion and comprises a feedforward neural network (FNN) and a convolutional neural network (CNN). The proposed model takes the basic elements that form the bases as input, defined by the generalized memory polynomial (GMP) and dynamic deviation reduction (DDR) models. The FNN generates the basis function and its output represents the basis values, while the CNN generates weights for the corresponding bases. Through the concurrent training of FNN and CNN, the hidden layer coefficients are updated, and the complex multiplication of their outputs yields the trained in-phase/quadrature (I/Q) signals. The proposed model was trained and tested using 300 MHz and 400 MHz broadband data in an orthogonal frequency division multiplexing (OFDM) communication system. The results show that the model achieves an adjacent channel power ratio (ACPR) of less than –48 dB within a 100 MHz integral bandwidth for both the training and test datasets.

Key words: basis function generation, digital predistortion, generalized memory polynomial, dynamic deviation reduction, neural network

. [J]. ZTE Communications, 2025, 23(1): 71-77.

SHAO Jianfeng, HONG Xi, WANG Wenjie, LIN Zeyu, LI Yunhua. A Basis Function Generation Based Digital Predistortion Concurrent Neural Network Model for RF Power Amplifiers[J]. ZTE Communications, 2025, 23(1): 71-77.

图/表 7

参考文献 18

1	LIU Z J, HU X, WANG W D, et al. A joint PAPR reduction and digital predistortion based on real-valued neural networks for OFDM systems [J]. IEEE transactions on broadcasting, 2022, 68(1): 223–231. DOI: 10.1109/TBC.2021.3132158
2	LIU Z J, HU X, XU L X, et al. Low computational complexity digital predistortion based on convolutional neural network for wideband power amplifiers [J]. IEEE transactions on circuits and systems II: express briefs, 2022, 69(3): 1702–1706. DOI: 10.1109/TCSII.2021.3109973
3	LIU X, CHEN W H, WANG D H, et al. Robust digital predistortion for LTE/5G power amplifiers utilizing negative feedback iteration [J]. ZTE communications, 2020, 18(3): 49–56. DOI: 10.12142/ZTECOM.202003008
4	LIU Y J, ZHOU J, CHEN W H, et al. A robust augmented complexity-reduced generalized memory polynomial for wideband RF power amplifiers [J]. IEEE transactions on industrial electronics, 2014, 61(5): 2389–2401. DOI: 10.1109/TIE.2013.2270217
5	ZHU A D, DRAXLER P J, YAN J J, et al. Open-loop digital predistorter for RF power amplifiers using dynamic deviation reduction-based Volterra series [J]. IEEE transactions on microwave theory and techniques, 2008, 56(7): 1524–1534. DOI: 10.1109/TMTT.2008.925211
6	BRAITHWAITE R N. Adaptation of a digitally predistorted RF amplifier using selective sampling [J]. ZTE communications, 2011, 9(3): 3–12
7	HU X, LIU Z J, WANG W D, et al. Low-feedback sampling rate digital predistortion using deep neural network for wideband wireless transmitters [J]. IEEE transactions on communications, 2020, 68(4): 2621–2633. DOI: 10.1109/TCOMM.2020.2966718
8	ROSOŁOWSKI D W, JĘDRZEJEWSKI K. Experimental evaluation of PA digital predistortion based on simple feedforward neural network [C]//Proc. 23rd International Microwave and Radar Conference (MIKON). IEEE, 2020: 293–296. DOI: 10.23919/ MIKON48703.2020.9253814
9	FENG X, FEUVRIE B, DESCAMPS A S, et al. Digital predistortion method combining memory polynomial and feed-forward neural network [J]. Electronics letters, 2015, 51(12): 943–945. DOI: 10.1049/el.2015.0276
10	WU H B, CHEN W H, LIU X, et al. A uniform neural network digital predistortion model of RF power amplifiers for scalable applications [J]. IEEE transactions on microwave theory and techniques, 2022, 70(11): 4885–4899. DOI: 10.1109/TMTT.2022.3205930
11	JUNG S, KIM Y, WOO Y, et al. A two-step approach for DLA-based digital predistortion using an integrated neural network [J]. Signal processing, 2020, 177: 107736. DOI: 10.1016/j.sigpro.2020.107736
12	WU Y B, GUSTAVSSON U, AMAT A G I, et al. Residual neural networks for digital predistortion [C]//Proc. 2020 IEEE Global Communications Conference. IEEE, 2020: 1-6. DOI: 10.1109/globecom42002.2020.9322327
13	LÓPEZ-BUENO D, MONTORO G, GILABERT P L. Training data selection and dimensionality reduction for polynomial and artificial neural network MIMO adaptive digital predistortion [J]. IEEE transactions on microwave theory and techniques, 2022, 70(11): 4940–4954. DOI: 10.1109/TMTT.2022.3209214
14	JIANG C Y, YANG G C, HAN R L, et al. Gated dynamic neural network model for digital predistortion of RF power amplifiers with varying transmission configurations [J]. IEEE transactions on microwave theory and techniques, 2023, 71(8): 3605–3616. DOI: 10.1109/TMTT.2023.3241612
15	JARAUT P, ABDELHAFIZ A, CHENINI H, et al. Augmented convolutional neural network for behavioral modeling and digital predistortion of concurrent multiband power amplifiers [J]. IEEE transactions on microwave theory and techniques, 2021, 69(9): 4142–4156. DOI: 10.1109/TMTT.2021.3075689
16	JIANG C Y, LI H M, QIAO W, et al. Block-oriented time-delay neural network behavioral model for digital predistortion of RF power amplifiers [J]. IEEE transactions on microwave theory and techniques, 2022, 70(3): 1461–1473. DOI: 10.1109/TMTT.2021.3124211
17	KOBAL T, LI Y, WANG X Y, et al. Digital predistortion of RF power amplifiers with phase-gated recurrent neural networks [J]. IEEE transactions on microwave theory and techniques, 2022, 70(6): 3291–3299. DOI: 10.1109/TMTT.2022.3161024
18	MONDAL R, RISTANIEMI T, DOULA M. Genetic algorithm optimized memory polynomial digital pre-distorter for RF power amplifiers[C]//Proc. International Conference on Wireless Communications and Signal Processing. IEEE, 2013: 1–5. DOI: 10.1109/WCSP.2013.6677117

编辑推荐 0

Metrics

阅读次数

全文

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	2	0	0	66

来源	本网站	其他网站

次数	61	7
比例	90%	10%

摘要

176

最新录用	在线预览	正式出版

0	0	176

	来源	本网站

	次数	176
	比例	100%

Datasets	ACPR of Left Band/dB		ACPR of Right Band/dB		NMSE
Datasets	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	NMSE
300 MHz valset	-54.17	-46.18	-58.04	-48.65
300 MHz datasets	Max=-21.45 Min=-21.49 Average=-21.47	Max=-21.00 Min=-21.02 Average=-21.00	Max=-21.94 Min=-21.98 Average=-21.96	Max=-21.01 Min=-21.01 Average=-21.01	Max=-16.40 Min=-16.41 Average=-16.40
400 MHz valset	-54.90	-43.29	-53.61	-46.86
400 MHz datasets	Max=-21.36 Min=-21.40 Average=-21.38	Max=-19.39 Min=-19.48 Average=-9.45	Max=-22.61 Min=-22.88 Average=-22.66	Max=-20.13 Min=-20.22 Average=-20.19	Max=-17.02 Min=-17.05 Average=-17.04

Datasets	ACPR of Left Band/dB		ACPR of Right Band/dB		NMSE
Datasets	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	NMSE
300 MHz valset	-54.17	-46.18	-58.04	-48.65
300 MHz datasets	Max=-21.45 Min=-21.49 Average=-21.47	Max=-21.00 Min=-21.02 Average=-21.00	Max=-21.94 Min=-21.98 Average=-21.96	Max=-21.01 Min=-21.01 Average=-21.01	Max=-16.40 Min=-16.41 Average=-16.40
400 MHz valset	-54.90	-43.29	-53.61	-46.86
400 MHz datasets	Max=-21.36 Min=-21.40 Average=-21.38	Max=-19.39 Min=-19.48 Average=-9.45	Max=-22.61 Min=-22.88 Average=-22.66	Max=-20.13 Min=-20.22 Average=-20.19	Max=-17.02 Min=-17.05 Average=-17.04

Datasets	ACPR of Left Band/dB		ACPR of Right Band/dB		NMSE
Datasets	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	NMSE
300 MHz trainsets	Max=-50.27 Min=-50.77 Average=-50.48	Max=-42.53 Min=-43.15 Average=-42.94	Max=-51.29 Min=-51.99 Average=-51.70	Max=-44.25 Min=-45.14 Average=-44.73	Max=-43.62 Min=-46.12 Average=-46.69
300 MHz testsets	-49.65, -49.48	-42.77, -42.57	-50.71, -50.69	-44.35, -44.25	-44.77, -44.79
400 MHz trainsets	Max=-48.93 Min=-49.36 Average=-49.17	Max=-40.99 Min=-41.49 Average=-41.24	Max=-48.66 Min=-49.10 Average=-48.82	Max=-42.45 Min=-43.40 Average=-43.10	Max=-43.86 Min=-44.31 Average=-44.15
400 MHz testsets	-48.02, -48.17	-40.50, -40.64	-48.13, -48.12	-42.44, -42.54	-42.92, -42.85

Datasets	ACPR of Left Band/dB		ACPR of Right Band/dB		NMSE
Datasets	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	100 MHz Integral Bandwidth	20 MHz Integral Bandwidth	NMSE
300 MHz trainsets	Max=-50.27 Min=-50.77 Average=-50.48	Max=-42.53 Min=-43.15 Average=-42.94	Max=-51.29 Min=-51.99 Average=-51.70	Max=-44.25 Min=-45.14 Average=-44.73	Max=-43.62 Min=-46.12 Average=-46.69
300 MHz testsets	-49.65, -49.48	-42.77, -42.57	-50.71, -50.69	-44.35, -44.25	-44.77, -44.79
400 MHz trainsets	Max=-48.93 Min=-49.36 Average=-49.17	Max=-40.99 Min=-41.49 Average=-41.24	Max=-48.66 Min=-49.10 Average=-48.82	Max=-42.45 Min=-43.40 Average=-43.10	Max=-43.86 Min=-44.31 Average=-44.15
400 MHz testsets	-48.02, -48.17	-40.50, -40.64	-48.13, -48.12	-42.44, -42.54	-42.92, -42.85

A Basis Function Generation Based Digital Predistortion Concurrent Neural Network Model for RF Power Amplifiers

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 18

相关文章 0

编辑推荐 0

Metrics