Research opportunities for MEng Students at UofT
The projects listed on this page are for current (or incoming) MEng students at UofT.
MEng students may complete a project under ECE2500Y in our team, or participate as a part-time reseaerch assistant. Note that the workload of ECE2500Y is equivalent to three ECE 1000-level courses. The requirements by the ECE department can be found in the following link: https://www.ece.utoronto.ca/graduates/degree-programs/meng/
If you are interested in applying for these positions, please email the following materials to Prof. Xilin Liu at xilinliu at ece dot utoronto dot ca:
1. A cover letter stating the research project that you are interested in and relevant experience
2. Latest CV
3. (Unofficial) transcripts of undergraduate and master’s studies
4. Sample of publications (if applicable)
Please include ECE2500Y or MEng in your email title
Project 1: Development of a Miniature, Wireless Brain-Computer Interface
Human brains are among the most complex and mysterious objects in the known universe. The emerging brain-computer interface (BCI) technology can create direct communication pathways between the brain and the external world, which creates unprecedented opportunities for truly understanding the brain and treating brain diseases [1]. Specifically, more than 385,000 Canadians are living with stroke and spinal cord injuries. BMI has shown promising potentials for the rehabilitation of these conditions and greatly improves the life quality of these patients [2].However, most existing BCI systems use rack-mount equipment for neural recording, stimulation, and signal processing, which significantly limits the experimental paradigms, especially during pre-clinical experiments in animal models. In this project, we aim to develop a miniature, wireless BCI device using a new generation microprocessor that is capable of running advanced signal processing algorithms (including deep learning). The microprocessor will interface with the brain bi-directionally via a state-of-the-art neural interface chip (for neural stimulation and recording) and communicate with computers via Bluetooth. The main outcome of this project is a battery-powered, miniature BCI prototype that can be used in a broad range of high-impact rehabilitation research and experiments.
The student will be working on advanced real-time embedded system programming, system integration, and miniature PCB prototyping (neural engineering knowledge is NOT required for this project). The student will work with a top interdisciplinary research team with expertise in electrical engineering, data science, neural engineering, and neuroscience. We are committed to promoting equity, diversity, and inclusion during our research and we strongly encourage students from underrepresented communities to join.
References:
[1] M. Zhang et al., “Electronic neural interfaces,” Nature Electronics, vol. 3, Apr. 2020.
[2] L. I. Jovanovic et al., “Brain–computer interface-triggered functional electrical stimulation therapy for rehabilitation of reaching and grasping after spinal cord injury: a feasibility study,” Spinal Cord Series and Cases, vol. 7, 2021.
Requirements:
Project 2: Design of Continuous-Time Noise Shaping SAR ADC
Low-power and high dynamic range analog to digital converters (ADCs) are the most critical building blocks in wireless and wireline communication [1]. Continuous-time (CT) delta-sigma ADC provides an inherent anti-alias filtering capability, thus is suitable for wideband applications. Conventional CT ADCs require a high oversampling ratio unless a high-order loop filter is used, at the cost of high complexity and power consumption. One solution to breaking this trade-off involves using a high-resolution quantizer. Flash quantizers can be used but are often limited to 4 bits because their hardware cost increases exponentially with the number of bits [2].In this work, we will explore the use of a noise-shaping successive approximation register (SAR) quantizer for CT ADC. Noise-shaping SAR can potentially achieve a high effective resolution with limited hardware resources while relaxing the order of the CT loop filters. The noise-shaping implemented in the discrete-time domain also has the advantages of robustness over process, voltage, and temperature (PVT).
The student will be working on modeling and designing a CT ADC with a 2nd order loop filter and a 2nd order SAR quantizer [3]. The student will fully understand the limitations of the architecture and will be able to perform design trade-offs. The circuit design and simulation will be performed in Cadence using a 22nm PDK from Global Foundries (or similar technologies). Students with experience in ADC design and using Cadence for simulation are especially encouraged to apply.
References:
[1] L. Jie et al., “An Overview of Noise-Shaping SAR ADC: From Fundamentals to the Frontier,” IEEE Open Journal of the Solid-State Circuits Society, vol. 1, pp. 149–161, 2021, doi: 10.1109/OJSSCS.2021.3119910.
[2] J. Liu, S. Li, W. Guo, G. Wen, and N. Sun, “A 0.029-mm2 17-fJ/Conversion-Step Third-Order CT $\Delta\Sigma$ ADC With a Single OTA and Second-Order Noise-Shaping SAR Quantizer,” IEEE Journal of Solid-State Circuits, vol. 54, no. 2, pp. 428–440, Feb. 2019, doi: 10.1109/JSSC.2018.2879955.
[3] W. Shi et al., “10.4 A 3.7mW 12.5MHz 81dB-SNDR 4th-Order CTDSM with Single-OTA and 2nd-Order NS-SAR,” in 2021 IEEE International Solid- State Circuits Conference (ISSCC), Feb. 2021, vol. 64, pp. 170–172. doi: 10.1109/ISSCC42613.2021.9366023.
Requirements:
Project 3: Low-cost Wireless EMG Sensor Network for Rehabilitation
More details to be addedStatement on Equity, diversity and inclusion (EDI):
Evidence clearly shows that increasing equity, diversity and inclusion (EDI) in research environments enhances excellence, innovation and creativity. I am committed to promote EDI in my research team and student training environment. I strongly encourage people with diverse backgrounds, especially those from underrepresented groups, to join my team.Back to other research opportunites.
Updated on 05/24/2022
