Tao Wu
James Watt Building South
Glasgow, UK, G12 8QQ
Hi, I’m a third year PhD student in Electronic Engineering with James Watt School of Engineering, University of Glasgow, Scotland, UK, advised by Prof. Bo Liu. I obtained a BEng degree in Electronic Engineering from University of Glasgow, a BEng degree in Electronic Engineering from UESTC, and an MS degree in Computer Science from Georgia Institute of Technology, Atlanta, GA. I have also worked at MathWorks as a research intern.
My research interests lie at the intersection of artificial intelligence (AI) and radio frequency (RF) designs. I leaverage few-shot learning enhanced optimization and large language models (LLMs) to help RF engineers engage with efficient design. I have developed and upgraded optimization methods for RF designs, including proposing a new optimization method for pixelized antennas with a resolution of 2000 or higher (IEEE TAP), and refining optimization methods for microwave filters (also used for photonic filters ECOC2025) and power amplifers (IEEE TMTT). Recently, my work looks at digitalizing and streamlining RF design workflow by using LLMs (arXiv & GitHub).
news
| Dec 12, 2025 | Our paper, Large Language Model-Based Intelligent Antenna Design System (an updated version for LEAM), has been accepted by EuCAP 2026! Let’s goooo Dublin! |
|---|---|
| Oct 14, 2025 | Our paper, An Efficient Method for Complex Digitally Coded Antenna Design Based on Evolutionary Computation and Machine Learning Techniques, has been online. This paper demonstrates an advancement of pixelated antenna design in resolution (i.e., more than 2000) and specifications (i.e., more than 10). |
| Oct 01, 2025 | For ECOC 2025, I went Copenhagen and our paper, Integrated Multi-Band Photonic Filter Based on MRR–SSG for Tunable Frequency Hopping, was presented. |
selected publications
-
An efficient and general automated power amplifier design method based on surrogate model assisted hybrid optimization techniqueIEEE Transactions on Microwave Theory and Techniques, 2024 -
Leam: A prompt-only large language model-enabled antenna modeling methodarXiv preprint arXiv:2504.18271, 2025 -
An efficient method for complex digitally coded antenna design based on evolutionary computation and machine learning techniquesIEEE Transactions on Antennas and Propagation, 2025