跳到主要导航 跳到搜索 跳到主要内容

Hardware Acceleration of Pattern Synthesis CNN for Array Antennas Based on FPGA

  • Zhikuo Li
  • , Xiang Li
  • , Ming Bai*
  • *此作品的通讯作者
  • Beihang University

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

This paper presents an FPGA-based accelerated neural network implementation for large-scale array antenna radiation pattern synthesis, addressing the critical challenges of real-time adaptive beam scanning. The methodology combines a convolutional neural network designed for radiation pattern synthesis with efficient FPGA deployment using the Vitis AI tools. Experimental validation on a 1024-element array antenna pattern synthesis demonstrates superb performance: sub-0.1° beam pointing accuracy, sidelobe level deviations constrained to within 2 dB, while reducing the synthesis time from 163.6 ms (GPU implementation) to 4.7 ms (35-fold speedup).

源语言英语
主期刊名IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization for RF, Microwave, and Terahertz Applications, NEMO 2025 - Proceedings
出版商Institute of Electrical and Electronics Engineers Inc.
ISBN(电子版)9798331597993
DOI
出版状态已出版 - 2025
活动2025 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization for RF, Microwave, and Terahertz Applications, NEMO 2025 - Tianjin, 中国
期限: 29 7月 20251 8月 2025

出版系列

姓名IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization for RF, Microwave, and Terahertz Applications, NEMO 2025 - Proceedings

会议

会议2025 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization for RF, Microwave, and Terahertz Applications, NEMO 2025
国家/地区中国
Tianjin
时期29/07/251/08/25

学术指纹

探究 'Hardware Acceleration of Pattern Synthesis CNN for Array Antennas Based on FPGA' 的科研主题。它们共同构成独一无二的学术指纹。

引用此