NSF REU 2026 Summer Research: Edge Devices for Microplastics Detection

This REU project reflects Dr. Wang's current interest in using ECE knowledge to guide undergraduate research toward meaningful real-world applications. Students are introduced to sensing, embedded systems, model deployment, and hardware-aware design through edge-device-based microplastics detection.

In this project, AI is not treated as an isolated software model. Instead, it is framed as a complete engineering pipeline that connects sensing, inference, deployment constraints, and system validation—an extension of Dr. Wang's long-standing hardware-software co-design perspective.

Project site: NSF REU 2026 Homepage

AI-Driven Clothing Recognition and Pricing System for Goodwill

Related award: Faculty Development Council Grant Program FY2026 →

This project applies AI and computer vision to a practical recognition-and-pricing system for clothing items in a circular-economy setting. It demonstrates Dr. Wang's continuing emphasis on taking intelligent models beyond theory and adapting them to real application environments.

The project also reflects her broader research direction in efficient AI deployment: building systems where model performance, application constraints, and implementation practicality are considered together rather than separately.

Fast Convolutional Algorithm Optimization and FPGA Acceleration

This research direction is a central bridge between algorithm design and hardware realization. Dr. Wang's doctoral work on fast convolution algorithms focused on improving computational efficiency for AI acceleration, while also moving those ideas toward FPGA-oriented implementation and validation.

The project clearly represents a software-algorithm + hardware-platform co-design approach: signal processing theory, algorithm optimization, architecture-level thinking, and FPGA verification all interact as part of a unified development path.

Grapevine Bud / Inflorescence Detection and Yield Prediction

This project studies grapevine bud burst detection, inflorescence detection, and yield-related analysis using field-acquired RGB images. It represents Dr. Wang's extension of AI model development toward practical edge-oriented vision tasks in challenging natural environments.

In the larger context of her work, this project highlights the transition from model design toward efficient deployment thinking: how AI models can be improved, adapted, and eventually moved closer to resource-constrained edge devices and real operational scenarios.

Video Decoder Chip Development at Montage Technology

As a Firmware Engineer at Montage Technology, Dr. Wang worked on video decoder chip related development for embedded multimedia and set-top-box systems. This experience gave her direct exposure to digital IC development, verification processes, firmware-hardware interaction, and the complete workflow required to move from design objectives to validated system behavior.

That industry foundation remains an important starting point for her current academic work. It explains the continuity in her research—from system validation and implementation constraints, to FPGA-based acceleration, and now toward broader VLSI and lower-level hardware design topics in both research and teaching.