Machine Learning for Design and Optimization of Coding Metasurfaces - PhD Thesis

Aug 4, 2026

Uploaded by Asher

Metasurfaces, ultra-thin two-dimensional arrays of subwavelength nanostructures, enable precise control of electromagnetic behavior including phase, polarization, and propagation direction. Coding metasurfaces provide digital programmability for innovative manipulation of EM waves, but traditional electromagnetic simulations are computationally intensive and restrict real-time large-scale design. This thesis applies machine learning to build fast electromagnetic simulation, optimization, and analytical expression correction methods. It uses supervised learning for reflection-phase prediction, proposes active learning for efficient optimization with major speedups, and employs deep learning to improve analytical expressions by accounting for mutual coupling effects. Results demonstrate scalable reductions in optimization time while achieving high-accuracy metasurface designs and performance improvements beyond prior work.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages154

Reading Time6 h 28 min

AccessFree to download

Declaration

Dedication

Acknowledgements

Abstract

List of Publications

Machine Learning for Design and Optimization of Coding Metasurfaces Parvathy Chittur Subramanianprasad PhD Thesis School of Electronic Engineering and Computer Science Queen Mary University of London May 2024 Declaration I, Parvathy Chittur Subramanianprasad, hereby declare that this thesis is my own work or that where it has been carried out in collaboration with, or supported by others, that this is duly acknowledged below and my contribution indicated. Previously published material is also acknowledged below. I attest that I have exercised reasonable care to ensure that the work is original, and does not to the best of my knowledge break any UK law, infringe any third party’s copyright or other Intellectual Property Right, or contain any confidential material. I accept that the College has the right to use plagiarism detection software to check the electronic version of the thesis. I confirm that this thesis has not been previously submitted for the award of a degree by this or any other university. The copyright of this thesis rests with the author and no quotation from it or information derived from it may be published without the prior written consent of the author. Signature: Parvathy C S Date: 20th May 2024 Dedication This thesis is dedicated to my grandparents Thathanna, Maniammal, Paatiammai and Paatappa. Acknowledgements First and foremost, I would like to express my gratitude to my supervisor, Prof. Yang Hao, for being a steadfast support throughout my PhD journey. Prof. Hao not only provided invaluable research guidance but also offered unwavering emotional and financial support, instilling me with the confidence to progress forward. I would also like to thank my second supervisor Prof. Ioannis Patras and independent assessor Prof. Akram Alomainy for their valuable suggestions on my study. I am also obliged to express a deep sense of gratitude to all the group members of the Antenna & Electromagnetics Research Group. A special gratitude to antenna lab managers, Dr. Max Munoz Torrico, Dr. Andre Sarker Andy, and my fellow PhD student Mr. Jonas Kolb. I sincerely thank my colleagues including Dr. Henry Giddens, Dr. Benjamin Vial, Dr. Hangfeng Zhang, Dr. Young Ok-Cha, Dr. Yihan Ma, Dr. Orestis Christogeorgos, Dr. Valentina Cicchetti, Dr. Achintha Avin Ihalage, Miss Mojan Omidvar, Mr Shalitha Pathiranage, Mr Ishan Fernando and Mr Rahul Dutta. It is their kind help and support that have made my study and life in the UK a wonderful time. I thank Westfield Nursery staff for taking care of my toddler son for the last one year. Their help is incredible without which I could not have completed my research work. Leaving my son with them was helpful for me to peacefully focus on my research and helpful for my son to learn lot of new things and make many friends. I thank my university for all support and opportunities provided. Thanks to Melissa Yeo for helping to keep track of finishing every milestone in the PhD journey. I am grateful to my son, Vasanth Chittur Krishnakumar, for being an incredible source of inspiration to me. I express gratitude to my parents, Lakshmy R Nand Subramaniaprasad C K, as well as my entire family, for their unwavering support. They serve as a constant source of inspiration, encouragement, and joy in my life. Their genuine care extends to both me and my son, and I am deeply appreciative of their presence in our lives. I thank Radhika Chechi, Kala Akka, Vellai Amma and Kamalam Chechi for taking care of my son in his first year of birth and allowing me to concentrate on my research work. I thank my dear most friends Revathy and Barbara to be of strong help and support to me always. Abstract Metasurfaces, ultra-thin two-dimensional arrays of subwavelength-sized nanostructures, offer precise control over light properties such as phase, polarization, and direction. The digital representation of metasurfaces through coding metasurfaces enables innovative manipulat

FAQs

What problem does the thesis address in coding metasurface design?

Traditional electromagnetic simulations are computationally intensive for large-scale designs, preventing practical real-time optimization.

How does the thesis improve metasurface design efficiency?

It integrates machine learning for electromagnetic simulation and optimization, including supervised learning for reflection-phase prediction and active learning for optimization speedups.

What approaches are used to enhance analytical accuracy?

Deep learning is used to refine analytical expressions, specifically correcting for mutual coupling effects to improve accuracy.

Aug 4, 2026

Uploaded by Asher