@inproceedings{7b1b16db553c4f938d50be723d591801,
title = "Deep Learning-Driven Rapid Correction of Beam-Control Codes for Radome-Enclosed Phased Arrays",
abstract = "This paper presents a deep learning approach for rapid correction of beam-control codes in radome-enclosed phased arrays. An end-to-end neural network predicts optimal phase-amplitude codes from sparse active element pattern (AEP) measurements using an Efficientunet with feature-wise linear modulation (FiLM) conditioning. A multi-task logarithmic domain loss term ensures main lobe fidelity, boresight error (BSE) and null depth (ND). Validated on an 8{\texttimes}8 Ka-band array with a multi-layer PCB radome, the 32-element sampling achieves 0° average BSE and optimal ND with millisecond-level prediction time per scanning angle.",
keywords = "Antenna radomes, Efficientunet, beam-control codes correction, deep learning, phased arrays",
author = "Tianrang Wang and Yuedong Liu and Yongkang Shi and Shengming Gu and Qiang Ren and Donglin Su",
note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 2026 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2026 ; Conference date: 01-07-2026 Through 03-07-2026",
year = "2026",
doi = "10.1109/NEMO66765.2026.11622301",
language = "英语",
series = "IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2026 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2026 - Proceedings",
address = "美国",
}