|
I am an Electrical Engineering student, working toward a PhD
with Professor Robert Nowak at the University of Wisconsin, Madison. My
research interests include estimation, signal processing, graphs, and sensor
networks.
During my Masters work, I focused on the following projects.
- Calibration
Sensors are notoriously prone to calibration errors, and arguably these
errors are one of the major obstacles to the practical use of sensor
networks. Calibrating every sensor by hand is infeasible if sensor
networks are to scale even into the tens of devices; yet it may be that
applications need more accurate measurements than uncalibrated, lowcost
sensors provide.
A calibration function is one that takes voltage measurements from a
sensor and translates that to what the actual measurement value should be.
Often that function is a function of both the true phenomenon and a set of
parameters. I am applying several estimation techniques in order to find
these parameters without the aid of a known stimulus or a high-quality
reference sensor.
Data coming from sensors is the confluence of
many environmental and technological factors merging into a sensor
measurement.
Often with varying battery voltages and ambient conditions, it
isn't clear if our calibration translations are enough to provide a
measurement we have confidence in. This project seeks to gather data on a
set of applications that can provide some useful insight into the accuracy
of calibration and the most important factors in the calibration problem.
- Fault and Outlier detection + Robust and Resilient data fusion
If we indeed hope sensor networks to become vast in numbers,
providing lots of data without much human intervention, we must build them
from the start to be resilient to faulty sensors. We also must keep in
mind that sometimes unusual data is what we find most interesting. This
project seeks to formulate models for fault along with statistical tools
for recognizing both data following a faulty model and unusual data which
does not fit into a faulty model.
|
