Deep learning enhanced optimization of a broadband and wide-angle reflective linear terahertz polarization converter

Simon Abdani; Rajour Tanyi Ako; Madhu Bhaskaran; Sharath Sriram

doi:10.1117/12.3028074

2 October 2024 Deep learning enhanced optimization of a broadband and wide-angle reflective linear terahertz polarization converter

Simon Abdani, Rajour Tanyi Ako, Madhu Bhaskaran, Sharath Sriram

Author Affiliations +

Proceedings Volume 13109, Metamaterials, Metadevices, and Metasystems 2024; 131090M (2024) https://doi.org/10.1117/12.3028074
Event: Nanoscience + Engineering, 2024, San Diego, California, United States

Conference Poster

Abstract

Artificial neural networks (ANNs) are known to be a versatile tool for device optimization. This work proposes a method to optimize a polarization converter composed of T-shaped periodic resonators, inclined at 45 deg using an ANN. The result is compared with previous work conducted using CST simulation, demonstrating broadband and wide-angle reflective linear polarization conversion. Employing an ANN resulted to improved performance metrics, leading to increased fractional bandwidth of 7.6% for normal incident and 9.8% for 45° incident angle. The neural network achieved a mean square error (MSE) as low as 5.78 × 10⁻⁵, indicating high accuracy. This approach demonstrates the efficiency of ANNs in designing metasurfaces for a wide range of applications.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Simon Abdani, Rajour Tanyi Ako, Madhu Bhaskaran, and Sharath Sriram "Deep learning enhanced optimization of a broadband and wide-angle reflective linear terahertz polarization converter", Proc. SPIE 13109, Metamaterials, Metadevices, and Metasystems 2024, 131090M (2 October 2024); https://doi.org/10.1117/12.3028074

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available