Nafisa Nawrin Labonno / Physics / Faculty Mentor: Seunghyun Lee
Single-Photon Avalanche Diodes (SPADs) play a critical role in applications such as LiDAR, quantum cryptography, and biomedical imaging. However, crosstalk between adjacent SPAD pixels degrades detection accuracy and overall device performance. This project utilizes Silvaco TCAD simulations to analyze and optimize the impact of carrier extraction structures on crosstalk suppression in InGaAs/InP SPAD arrays. Prior research has demonstrated that careful engineering of device layers and electric field distributions can mitigate crosstalk, but further optimization is needed. By applying machine learning techniques such as Binary Classification model with Logistic Regression to simulation results, we identify key design parameters that minimize unwanted carrier propagation and enhance the collection of photo-generated carriers in each pixel. The study’s findings are projected to show that optimized carrier extraction structures significantly reduce crosstalk while maintaining detection efficiency. This research contributes to the development of high-performance SPAD arrays with improved noise reduction, which is essential for next-generation photodetection technologies.
Leave a Reply