特征模理论与机器学习协同辅助的宽带小型化天线设计OA
Design of Broadband Miniaturized Antennas Synergistically Assisted by Characteristic Mode Theory and Machine Learning
基于特征模理论提出了一款宽带小型化微带贴片天线(Microstrip Patch Antenna,MPA),通过在天线辐射贴片中加载槽结构来延长电流路径并增强电抗加载效应,使其谐振频率降低,从而实现天线的小型化.采用特征模(Characteristic Mode,CM)理论对辐射单元工作机制进行分析并指导确定馈电位置的选择,使期望模式谐振;并在 2 个基板之间加入空气层,从而拓宽天线的带宽.同时,引入机器学习进行参数性能优化来克服传统工具扫参数据量大、耗时长等缺点,实现了天线的工作频率为 6.15～11.45 GHz(相对带宽为 60%),尺寸为 0.43λ0×0.43λ0×0.12λ0,与传统微带天线 0.6λ0×0.6λ0×0.17λ0 相比整体尺寸实现了 63%的缩减.
A broadband miniaturized Microstrip Patch Antenna(MPA)is proposed,which extends the current path and enhances the reactive loading effect by loading a slot structure on the antenna's radiating patch,thereby lowering its resonant frequency to achieve miniaturization.The Characteristic Mode(CM)theory is used to analyze the working mechanism of the radiating unit and guide the selection of the feed position to ensure the desired mode resonates.An air layer is introduced between the two substrates to broaden the antenna's bandwidth.Additionally,machine learning is introduced to optimize the parameter performance,overcoming the drawbacks of traditional tools,such as the large amount of parameter sweep data and high time consumption,ultimately achieving an operating frequency of 6.15～11.45 GHz(relative bandwidth of 60%)and a size of 0.43λ0×0.43λ0×0.12λ0.Compared to traditional microstrip antenna size of 0.6λ0×0.6λ0×0.17λ0,the overall size is reduced by 63%.
温乐乐;尤智源;高瑞;赵晔;何智慧
延安大学 物理与电子信息学院,陕西 延安 716000延安大学 物理与电子信息学院,陕西 延安 716000延安大学 物理与电子信息学院,陕西 延安 716000||先进光电材料与器件陕西省高等学校重点实验室,陕西 延安 716000延安大学 物理与电子信息学院,陕西 延安 716000||先进光电材料与器件陕西省高等学校重点实验室,陕西 延安 716000延安大学 物理与电子信息学院,陕西 延安 716000||先进光电材料与器件陕西省高等学校重点实验室,陕西 延安 716000
信息技术与安全科学
微带天线特征模宽带小型化机器学习
microstrip antennaCMbroadband miniaturizationmachine learning
《无线电工程》 2026 (5)
761-769,9
延安大学研究生实践创新计划(YSJ2026022)延安大学科研计划(GXYQ001) Graduate Practice Innovation Program of Yan'an Univer-sity(YSJ2026022)Scientific Research Program of Yan'an University(GXYQ001)
评论