采用卷积神经网络的双频功分器优化设计

孙思悦; 周文颖; 逯迈

doi:doi:10.11977/j.issn.1004-2474.2023.01.030

压电与声光, 2023, 45 (1): 158, 网络出版: 2023-04-07

采用卷积神经网络的双频功分器优化设计

Optimal Design of Dual-Band Power Divider Using Convolutional Neural Network

论文大纲

孙思悦周文颖逯迈

作者单位

兰州交通大学光电技术与智能控制教育部重点实验室, 甘肃兰州 730000

双频功分器卷积神经网络自组织映射神经网络多参数目标优化 dual-band power divider convolutional neural network self-organizing map neural network optimization of multi-parameter objectives

摘要

功率分配器(简称功分器)作为微波电路中常用的射频器件, 是构建5G系统多输入多输出(MIMO)馈电网络的重要组成单元。为了对已有固定频率的功分器结构进行重新快速地优化设计, 以适用于包括5G工作频段在内的任意实际所需的工作频段, 该文以预先设计的一种双频功分器作为优化设计目标, 提出了一种基于改进后的一维卷积神经网络的深度学习方案。预测功分器在其他任意双谐振频率处拥有良好性能的几何结构参数, 运用自组织映射神经网络进行样本的选取, 提高卷积神经网络的训练效率。预测出的功分器在电磁仿真软件中进行验证, 仿真结果显示功分器在工作频率处的回波损耗高于20 dB, 隔离度高于25 dB, 插入损耗低于3.4 dB, 工作带宽约为450~600 MHz, 证明了利用神经网络实现多参数目标功分器的优化设计是一种快速有效的方法。

Abstract

As a common RF device in the microwave circuits, the power divider is an important building block for building the multiple-input multiple-output (MIMO) feed network in 5G communication systems. In order to optimize and quickly redesign the existing fixed frequency power divider structure so as to apply to any actually desired operating band, including the 5G operating band, in this paper, we propose a deep learning scheme based on a modified one-dimensional convolutional neural network taking a pre-designed dual-band power divider as the optimized design target. The one-dimensional convolutional neural network could predict that the geometric structure parameter of the power divider has good performance at other arbitrary double resonant frequencies. The self-organizing mapping neural network is used to select samples to improve the training efficiency of convolutional neural network. The predicted power divider performance is verified in the electromagnetic(EM) simulation software. The simulation results show that the return loss of the power divider is higher than 20 dB at the resonant frequency, the isolation is higher than 25 dB, the insertion loss is lower than 3.4 dB, and the working bandwidth is about 450~600 MHz, which proves that the optimized design of multi-parameter objective power divider is a fast and effective method.

参考文献

[1] 刘文豹,杨自强,陈涛.宽带波导-微带一分四功分器设计［J］.压电与声光,2014,36(2): 289-290.

[2] ABDELRAHMAN B M,AHMED H N,NASHEDA I.A novel tri-band Wilkinson power divider for multiband wireless applications［J］.IEEE Microwave & Wireless Components Letters,2017,27(10): 891-893.

[3] WANG X,SAKAGAMI I,MA Z,et al.Miniaturized dual-band Wilkinson power divider with self-compensationstructure［J］.IEEE Transactions on Components Packaging & Manufacturing Technology,2015,5(3): 389-397.

[4] DARDEER O,ABOUELNAGA T,MOHRA A,et al.Compact UWB power divider,analysis and design［J］.Journal of Electromagnetic Analysis & Applications,2017,9(2): 10-21.

[5] BHATTACHARYA A,ROY B,ISLAM M,et al.An UWB Monopole antenna with hexagonal patch structure designed using particle swarm optimization algorithm for wireless applications［C］//［S.l.］:2016 International Conference on Microelectronics,Computing and Communications (MicroCom),IEEE,2016.

[6] 武欢欢,陈明,张延冬,等.宽带Wilkinson功分器的粒子群优化设计［J］.强激光与粒子束,2018,30(3): 37-41.

[7] KHAJEH-KHALILI F,HONARVAR M A.A design of triple lines Wilkinson power divider for application in wireless communicationsystems［J］.Journal of Electromagnetic Waves and Applications,2016,30(16): 2110-2124.

[8] 陈晓辉,郭欣欣,裴进明.一种基于Kriging模型和受限差分进化的电磁结构快速优化算法［J］.电波科学学报,2017,32(3): 273-278.

[9] 舒佩文,麦健业,褚庆昕.基于Harris Hawks优化算法的介质波导滤波器优化设计［J］.电波科学学报,2021,36(5): 787-796.

[10] STANKOVIC＇ Z ,OLC＇AN D I,DONCˇOV N S,et al.Consensus deep neural networks for antenna design and optimization［J］.IEEE Transactions on Antennas and Propagation,2021,70(7): 5015-5023.

[11] ZHOU W Y,MEI Z L,LU M,et al.Improved fully-connected neural network approach for decoupling microstrip antenna array design［J］.Journal of Electromagnetic Waves and Applications,2022,36(14): 1996-2009.

[12] NAN J C,XIE H,GAO M M,et al.Design of UWB antenna based on improved deep belief network and extreme learning machine surrogate models［J］.IEEE Access,2021,9: 126541-126549.

[13] XIAO L Y,SHAO W,JIN F L,et al.Inverse artificial neural network formultiobjective antenna design［J］.IEEE Transactions on Antennas and Propagation,2021,69(10): 6651-6659.

[14] KIM J H,BANG J.Antenna impedance matching using deep learning［J］.Sensors,2021,21(20): 6766-6775.

[15] ZKAYA U,SEYFI L,ZTRK A.Dimension optimization ofmulti-band microstrip antennas using deep learning methods［J］.Pamukkale University Journal of Engineering Sciences,2021,27(2): 229-233.

[16] Shukla N.TensorFlow机器学习［M］.北京: 机械工业出版社,2020.

[17] SARMAH K,SARMA K K.Design optimization of loop antenna using competitive learning ANN［C］//［S.l.］:2011 2nd National Conference on Emerging Trends and Applications in Computer Science,IEEE,2011.

[18] KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks［J］.Communications of the ACM,2017,60(6): 84-90.

孙思悦, 周文颖, 逯迈. 采用卷积神经网络的双频功分器优化设计[J]. 压电与声光, 2023, 45(1): 158. SUN Siyue, ZHOU Wenying, LU Mai. Optimal Design of Dual-Band Power Divider Using Convolutional Neural Network[J]. Piezoelectrics & Acoustooptics, 2023, 45(1): 158.

采用卷积神经网络的双频功分器优化设计

关于本站 Cookie 的使用提示

全站搜索

采用卷积神经网络的双频功分器优化设计

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索