Fall 2014 DSP Projects

The Digital Signal Processing class project links have now been posted! Click and prepare to be amazed. Great work students!

Productive December

December was chock full starting with a successful symposium at Global SIP on Information Processing in Big Data — thanks to my track co-organizers Yuejie Chi and Yao Xie, and thanks to Ryan Kennedy for presenting our paper on Matrix Completion for Ill-Conditioned Matrices. Then I attended FOCM and presented my work with Steve on the local convergence of GROUSE and its relationship to the ISVD. Besides all that I have almost wrapped up teaching Functional analysis and Digital Signal Processing, to two more outstanding groups of Michigan students.

Solving the Big Data Dilemma

MConnex has created a video highlighting my research in messy big data. We’re working on all kinds of applications to improve statistical methodologies with big data.

Robust Blind Calibration

My student John Lipor and I have had our work on robust blind calibration published in ICASSP. Previous work showed that it is possible to blindly calibrate sensor gains only knowing the signal subspace. This new algorithm allows you to calibrate sensor gains even when your knowledge of the signal subspace is inaccurate. John has shared his code for robust blind calibration here.

EECS 451 Projects from Fall 2013

I had a great time teaching Digital Signal Processing last semester. My favorite part was seeing the results of what the students did for their final projects. They really showed off what they learned in DSP!

Online Rigid Structure From Motion

We posted a paper on online algorithms for rigid structure from motion today on the arxiv. This is work spearheaded by Ryan Kennedy at Penn. Our results give good evidence that GROUSE-like algorithms could make a big impact in this area of computer vision.


I’m very excited that my proposal for blind calibration in environmental sensor networks has been given the NSF BRIGE award. Now we get to roll up our sleeves and do the work! You can read the original paper that inspired this proposal and the more detailed book chapter here.


Jun He, Dejiao Zhang, and I are happy to share our preprint and our code on t-GRASTA, a variant of GRASTA which includes estimation of geometric transformations of the data (such as translations and rotations of images, so that we can deal with jitter).

GROUSE local convergence

Steve Wright and I are pleased to share our work on the local convergence of GROUSE, an algorithm for estimating a subspace from partially observed vectors, which is now on the arxiv.

UW-Madison ECE Dissertation Award

Laura is very honored to receive the 2012 award for Best Dissertation from the University of Wisconsin, Madison ECE Department. The award even came with some cash, which Rob Nowak pointed out will buy a LOT of beer when we celebrate during the SILO meeting in June. Hope to see you there.