This project implements a real‑time engine knocking detection system on a resource‑constrained embedded platform. On‑board microphones capture combustion noise, a compact ML model classifies “knock” events in‑situ, and summarized diagnostics are pushed securely to a cloud backend for visualization, alerts and analytics. By enabling predictive maintenance and continuous performance monitoring, it helps prevent engine damage and optimize operational uptime.

This project involves the design, simulation, and layout of a CMOS image sensor using Cadence tools. Students will build and validate a 3-transistor active pixel sensor (3T APS), column and row C2MOS shift registers, and a column read-out circuit. The final design integrates a 4×4 pixel array with scanning logic and fits within a 2.5mm × 2.5mm layout. The imager will be tested through transient simulations showing pixel readouts. Deliverables include DRC/LVS-verified schematics, layouts, presentations, and a four-page IEEE-style final report.

Heavy Menstrual Bleeding (HMB), or Menorrhagia, is a menstrual condition defined as greater than 80 mL of blood loss per cycle. HMB can be an indicator for many serious conditions, such as endometriosis, reproductive cancer, and anemia. The most used methods to measure menstrual output volume are inaccurate and subjective. Furthermore, the volume threshold for HMB does not consider the unique characteristics between people who menstruate. This means clinicians need more insightful information about a person who menstruates menstrual output to assist in patient care. We are developing an imaging-based app solution that will utilize a machine learning model to offer an accurate and objective way to measure the menstrual blood output during a period, while also using a personalized approach to establish individual baseline blood loss values in order to predict abnormal menstrual bleeding.

Recent literature has discovered a fixed Silver step size schedule with occasional long steps for gradient descent which achieves an accelerated overall convergence guarantee on smooth convex functions. The problem with the Silver steps is that acceleration is only guaranteed at designated stopping points in the schedule. Our project seeks to address the open question of whether there exists a fixed schedule that achieves acceleration when stopping anywhere during the schedule. We first hypothesized that the improved step size schedule does not exist, based on evidence from numerical experiments using Performance Estimation Problems (PEP) software. Next, we employed proof techniques, including using Huber function and half quadratic function, as well as induction to try to prove the null result that a schedule with an accelerated anytime guarantee does not exist.