Fault and Calibration

Sensor data has been a focus of signal processing for decades. In modern sensor applications, engineers are deploying hundreds or thousands of sensors to monitorĀ  the environment, computer networks, vehicle traffic, and many other phenomena of interest. Signal processing for such large datasets can no longer rely on careful maintenance and calibration nor on expert-based outlier detection (i.e. a person looking at the data). We are developing machine learning algorithms to ensure that these data are still useful and enlightening, as well as the supporting theory to give guarantees that we may use data from large deployments with confidence.

Code for robust blind calibration is here and original code for our 2007 paper is here.