| Proj No. | A2266-251 |
| Title | AI-Driven Optimization of High-Frequency CNT-Based Filters Using Conditional Generative Adversarial Networks (cGANs) |
| Summary | The project will focus on the AI-assisted design optimization of high-frequency filters for millimeter-wave and sub-millimeter-wave applications (150 GHz – 300 GHz). This project explores the use of Conditional Generative Adversarial Networks (cGANs) to optimize filter geometries based on vertically aligned carbon nanotubes (VACNTs). VACNT-based high-frequency filters offer advantages such as low loss and CMOS compatibility, making them attractive for applications in telecommunications and next-generation wireless systems. However, optimizing their design to achieve specific performance targets, such as minimal insertion loss and precise frequency selectivity, is computationally demanding. In this project, you will develop or modify a cGAN framework to automate the optimization of filter structures based on simulated electromagnetic performance data. cGANs [Isola et al., 2017] have been successfully applied in antenna and metasurface design [Liu et al., 2020; Li et al., 2021], and this project aims to extend their application to high-frequency filter design. Objectives of the project: Implementing and adapting a cGAN model to generate optimized filter geometries. Training the model on simulated performance data from HFSS (high-frequency simulation software). Evaluating the generated designs based on key performance metrics, such as insertion loss and frequency response. Reducing computational costs by generating high-performance designs with fewer simulation cycles. This project will provide hands-on experience with AI model development, electromagnetic simulation, and high-frequency component optimization. It will equip you with skills applicable to research and industry roles in telecommunications, semiconductor technology, and AI-driven RF design. This project is part of an ongoing research effort and offers opportunities for collaboration with graduate students working on related topics. |
| Supervisor | Prof Tay Beng Kang (Loc:S1 > S1 B1A > S1 B1A 29, Ext: +65 67904533) |
| Co-Supervisor | - |
| RI Co-Supervisor | - |
| Lab | Nanoelectronics Lab. I (Loc: S1-B3a-01) |
| Single/Group: | Single |
| Area: | Microelectronics and Biomedical Electronics |
| ISP/RI/SMP/SCP?: |