基于编码—解码模型的D类功率放大器行为建模

doi:10.16180/j.cnki.issn1007-7820.2020.02.004

电子科技 ›› 2020, Vol. 33 ›› Issue (2): 20-24.doi: 10.16180/j.cnki.issn1007-7820.2020.02.004

基于编码—解码模型的D类功率放大器行为建模

赵一鹤,邵杰,程永亮

南京航空航天大学电子信息工程学院,江苏南京 210016

收稿日期:2019-01-10 出版日期:2020-02-15 发布日期:2020-03-12
作者简介:赵一鹤(1994-),男,硕士研究生。研究方向:信号检测与处理,深度学习。|邵杰(1963-),男,博士,副教授。研究方向:信号检测与处理,数字系统设计与计算机应用。
基金资助:
国家自然科学基金(61401198)

Behavior Modeling of Class-D Power Amplifier Based on Encoder-Decoder Model

ZHAO Yihe,SHAO Jie,CHENG Yongliang

School of Electronic Information Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China

Received:2019-01-10 Online:2020-02-15 Published:2020-03-12
Supported by:
Nation Natural Foundation of China(61401198)

摘要/Abstract

摘要：

D类功率放大器具有优异的传输效率,属于开关类功放,其输出信号存在较大的非线性失真。对D类功率放大器进行行为建模时要同时考虑其非线性和记忆特性。文中将小波变换引入到编码—解码神经网络模型中,提出了小波编码—解码神经网络模型。使用基于门限循环单元的编码—解码模型和小波编码—解码模型进行D类功率放大器的行为建模。实验结果表明,文中提出的D类功率放大器行为模型相比于传统的Voterra-Laguerre模型而言,在信号的时域和频域都具有更高的精度。

关键词: D类功率放大器, 非线性系统, 行为模型, 门限循环单元, 编码—解码神经网络;, 小波变换

Abstract:

Class D power amplifiers have excellent transmission efficiency and are classified as power amplifiers. Their output signals have large nonlinear distortion. The behavior modeling of calss-D power amplifier should take into account both nonlinearity and memory characteristics. This study introduced wavelet transform into the encoder-decoder neural network model, and proposed sequence to sequence wavelet neural network model. In this paper, the encoder-decoder model and sequence to sequence wavelet model based on gated recurrent unit were used in the behavior modeling of class-D power amplifier. Experiments results demonstrated that the proposed behavior model of class-D power amplifier had higher precision in time and frequency domain than the traditional Voterra-Laguerre model.

Key words: class-D power amplifier, nonlinearsystem, behaviormodeling, gated recurrent unit, encoder-decoder neural network, wavelet transform

中图分类号:

TP274

赵一鹤,邵杰,程永亮. 基于编码—解码模型的D类功率放大器行为建模[J]. 电子科技, 2020, 33(2): 20-24.

ZHAO Yihe,SHAO Jie,CHENG Yongliang. Behavior Modeling of Class-D Power Amplifier Based on Encoder-Decoder Model[J]. Electronic Science and Technology, 2020, 33(2): 20-24.

图/表 5

图1

图2

表1

图3

表2

参考文献 16

[1]	Roblin P, Root D E, Verspecht J , et al. Newtrends for the nonlinear measurement and modeling of high-power RF transistors and amplifiers with memory effects[J]. IEEE Transactions on Microwave Theory & Techniques, 2012,60(6):1964-1978.
[2]	Galaviz J A, Roblin P, Cárdenas J R , et al. Comparison of a genetic programming approach with ANFIS for power amplifier behavioral modeling and FPGA implementation[J]. Soft Computing, 2017,32(10):1-19.
[3]	Amhaimar L, Ahyoud S, Asselman A , et al. Peak-to-Average power ratio reduction based varied phase for MIMO-OFDM systems[J]. International Journal of Advanced Computer Science and Applications, 2016,7(1):432-437.
[4]	Simbélie F, Laurent S, Medrel P , et al. Non-linear electrical modeling of MASMOS structure to design power amplifiers for 4G applications[J]. International Journal of Microwave and Wireless Technologies, 2018,10(1):9-17.
[5]	Hao R, Zhang X, Gao H , et al. Avolterra-series-based design technique for the bias circuit of class-AB RF power amplifier[C]. Xi’an:2 ^nd IEEE Advanced Information Management,Communicates,Electronic and Automation Control Conference , 2018.
[6]	Jing X, Lang Z. Frequency domain analysis and design of nonlinear systems based onvolterra series expansion[M]. Berlin: Springer International Publishing, 2015.
[7]	Jie Shao, Zhao Weisong, Jin Xiangjun , et al. Underwater broadband acoustic scattering modelling based on FDTD[J]. ElektronikaIr Elektrotechnika, 2015,21(2):1-7.
[8]	Jin X, Shao J, Zhang X , et al. Modeling of nonlinear system based on deep learning framework[J]. Nonlinear Dynamics, 2016,84(3):1327-1340.
[9]	Che Z, Purushotham S, Cho K , et al. Recurrent neural networks for multivariate time series with missing values[J]. Scientific Reports, 2018,8(1):6085-6090.
[10]	Zhu Y, Imamura M, Nikovski D , et al. Introducing time series chains: a new primitive for time series data mining[J]. Knowledge and Information Systems, 2018,6(3):1-27.
[11]	Luongvinh D, Kwon Y . Behavioral modeling of power amplifiers using fully recurrent neural networks[C]. Long Beach:IEEE MTT-S International Microwave Symposium Digest, 2005.
[12]	Yan S, Zhang C, Zhang Q . Recurrent neural network technique for behavioral modeling of power amplifier with memory effects[J]. International Journal of RF and Microwave Computer-Aided Engineering, 2015,25(4):289-298.
[13]	周祥全, 张津 . 深层网络中的梯度消失现象[J]. 科技展望, 2017,6(27):284-285.
	Zhou Xiangquan, Zhang Jin . Gradient disappearance in deep neural networks[J]. Technology Outlook, 2017,6(27):284-285.
[14]	Greff K, Srivastava R K, Koutnik J , et al. LSTM:a search space odyssey[J]. IEEE Transactions on Neural Networks & Learning Systems, 2016,28(10):2222-2232.
[15]	尹光花, 刘小明, 张露 , 等. 基于LSTM特征模板的短文本情感要素分析与研究[J]. 电子科技, 2018,11(10):38-41.
	Yin Guanghua, Liu Xiaoming, Zhang Lu , et al. Sentiment elements of internet short texts for analysis and research based on LSTM network mode[J]. Electronic Science and Technology, 2018,11(10):38-41.
[16]	Zhou G B, Wu J, Zhang C L , et al. Minimal gated unit for recurrent neural networks[J]. International Journal of Automation & Computing, 2016,13(3):226-234.

模型	平均时域误差/V
VL	0.093 5
S2S	9.45×10^-4
S2S-WL	5.28×10^-4

模型	VL	Seq2Seq	Seq2Seq-WL
平均频域误差	6.7114	0.5904	0.3572
f₁误差/dB	-0.0653	0.0013	0.0005
f₂误差/dB	0.0038	0.0003	0.0002
f₃误差/dB	5.5144	-0.0046	-0.0087
f₄误差/dB	9.0907	-0.0004	-0.0010
f₅误差/dB	4.4487	-0.0156	0.0257
f₆误差/dB	8.0684	0.0278	-0.0065

基于编码—解码模型的D类功率放大器行为建模

Behavior Modeling of Class-D Power Amplifier Based on Encoder-Decoder Model

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 16

相关文章 15

Metrics

本文评价

推荐阅读 7

[1]	窦征立,王亚刚. 基于Elman神经网络的联合循环机组燃烧室温度模型建模[J]. 电子科技, 2021, 34(3): 60-64.
[2]	郭欣,高燕,蒋琳,张志姝. 基于事件触发和量化的非线性系统稳定性研究[J]. 电子科技, 2020, 33(3): 56-61.
[3]	张乐乐,苏前敏. 具有不匹配量化的非线性系统自适应滑模控制[J]. 电子科技, 2020, 33(12): 38-43.
[4]	刘露露,刘烨. 基于动态面的死区输入系统跟踪控制算法[J]. 电子科技, 2020, 33(11): 49-54.
[5]	朱先和，杨世平. 基于小波变换对信号噪声的处理研究[J]. , 2016, 29(6): 128-.
[6]	赵季红,贺丹,王伟华,曲桦,雷佩. 基于小波变换的酉ESPRIT波束赋形技术研究[J]. , 2015, 28(9): 6-.
[7]	邱政,汪锋,潘元璋. 一种基于频谱细化技术的水下目标识别方法[J]. , 2015, 28(7): 56-.
[8]	马立新,单宇. 基于小波概率网络的局部放电模式识别[J]. , 2015, 28(7): 94-.
[9]	方辰. 基于有理函数模型多源遥感数据网络服务设计[J]. , 2015, 28(6): 33-.
[10]	钱莉,姚恒,刘牮. 基于小波高频系数统计特征的电路故障诊断[J]. , 2015, 28(6): 118-.
[11]	胡亚军,陈青. 图像水印技术在电力系统中的应用[J]. , 2015, 28(4): 149-.
[12]	董佳鹏,何建忠. 基于冗余小波变换的灰度多聚焦图像融合方法[J]. , 2015, 28(12): 100-.
[13]	田谨,应捷,朱丹丹. 一种新的箭头形交通信号灯识别方法[J]. , 2015, 28(11): 150-.
[14]	黄如科,梁天生,余汉华. 基于小波变换与BP神经网络的模拟电路故障诊断[J]. , 2014, 27(8): 46-.
[15]	巩小磊,龚涛,朋杨琴. 基于免疫克隆选择的图像压缩算法研究[J]. , 2014, 27(8): 170-.