I received B.S. from University of Science and Technology of China (USTC) in 2012. Between 2012 and 2014, I worked at Hong Kong University of Science and Technology (HKUST) as a postgraduate student and achieved Ph.D. candidacy. Since joining Rice in 2014, I have been working with the remarkable fellows in Rice Efficient Computing Group led by my advisor Dr. Lin Zhong.
With my background in circuits and systems, my latest research focuses on radical mixed-signal computing architectures with operating system and programming language support.
Despite recent efforts to tackle energy inefficiency in mobile sensing systems by shifting signal processing into analog domain, research community still faces challenges in generality, design complexity and signal fidelity of analog computing systems.
This work retrofits general-purpose digital systems by proposing a dual-chip programmable computing system with analog datapath, including design of the analog instruction set architecture, and mechanisms to mitigate signal fidelity concerns.
Continuous mobile vision is limited by the inability to efficiently capture image frames and process vision features. This is largely due to the energy burden of analog readout circuitry, data traffic, and intensive computation.
To promote efficiency, we shift early vision processing into the analog domain. This results in RedEye, an analog convolutional image sensor that performs layers of a convolutional neural network in the analog domain before quantization. We design RedEye to mitigate analog design complexity, using a modular column-parallel design to promote physical design reuse and algorithmic cyclic reuse.
More information on my past projects can be found on my LinkedIn page.
Last updated: Sep 2016