Computational Advances in Multi-Sensor Adaptive Processing

I just returned from a great CAMSAP meeting, where I learned about a probabilistic theory of deep learning, new work in tensor completion, and an importance sampling strategy for non-convex block coordinate descent. My student John and I also presented our new work on active labeling in the subspace clustering problem, where we showed that a small amount of label information helps tremendously in clustering noisy data when each cluster lies near a low-dimensional subspace. Thanks to the organizers for putting together such a great meeting!

Intel Early Career Faculty Honor Program

I am honored to have received the Intel Early Career Faculty Honor Program Award for my research in big data. The purpose of the program is to help Intel connect with early career faculty members who show promise as future academic leaders in disruptive computing technologies.

DSP undergraduate projects highlighted in EECS news

With every new fall semester comes the excitement of teaching EECS 351, Digital Signal Processing. This week the EECS news highlighted my initiative to incorporate data collection and analysis with DSP into 351 with a course project. It’s an initiative I began two years ago and will continue for as long as I teach DSP. They mentioned a few of the cool projects students have done, and there are many more at the links. I’m looking forward to seeing what the students will do in my class this fall!

Perry has hit the water

Check out this great video featuring my collaborator Branko Kerkez and his students Brandon Wong and Rachel Menge. My student John Lipor and I joined them last week out at Sister Lake to test our autonomous lake sampling watercraft.

Perry on the Sister Lake

Perry is a small watercraft whose main control hardware consists of an Arduino and an Android phone. These work together to navigate the boat using GPS to take samples of water temperature and oxygen levels. With the phone, the boat can be controlled from anywhere in the world. We are implementing adaptive sampling algorithms so that Perry will know where best to take a next sample based on previous measurements.

SFM code

Ryan Kennedy put our SAGE GROUSE code for Structure from Motion online. We’ve also updated the preprint of our paper on this topic. Please let us know if you have any questions!

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!