Oregon State University

Can’t find an event? We’re busy migrating to a new event calendar. Try looking new calendar



Event Details

PhD Oral Preliminary Examination – Vahid Behravan

Wednesday, September 14, 2016 2:00 PM - 4:00 PM

Compressive Sensing for Low Power Sensor Design
Recent sensor system-on-chips have enabled significant advances in energy-efficiency by incorporating various micro-powered building blocks. Unfortunately, most of these sensor systems do not address the high power cost associated with data storage and transmission that in some cases vastly exceeds the power consumed by the rest of the SoC. In recent years, Compressive Sensing (CS) has been proposed as a method to achieve significant sensor data compression, achieving compression rates up to 10x depending on the signal sparsity.

In this work we address conventional compressive-sensing issues including non-adaptive compression rate and suggest solutions for these issues. We conduct a feasibility study on adaptive compressed sensing, investigating the sparsity variance of some biomedical signals and propose an adaptive CS framework, adjusting the compression rate based upon the input signal’s sparsity on-the-fly.

We also propose a compressive-sensing framework whose reconstruction is aided by statistics collection. We demonstrate how to fuse sensor data and statistics information together to improve the signal-to-error ratio (SER) of reconstruction. We fabricated a test chip in TSMC 65-nm technology to implement compressive sensing in a sensor on chip incorporating statistics collection block in order to improve performance of the compressive sensing algorithm.

Major Advisor: Patrick Chiang
Committee: Jinsub Kim
Committee: Matthew Johnston
Committee: Huaping Liu
GCR: Michelle Dolgos

Kelley Engineering Center (campus map)
Nicole Thompson
1 541 737 3617
Nicole.Thompson at oregonstate.edu
Sch Elect Engr/Comp Sci
This event appears on the following calendars: