Featured Engineer

Interview with Dr. Michael Wakin

Dr. Michael Wakin

Dr. Michael Wakin - Assistant Professor, Division of Engineering (Electrical Specialty); Colorado School of Mines

How did you get into electronics/ engineering and when did you start?

Up through high school, I always had an interest in math and science, but I really didn’t know what electrical engineering was until I got to college. Once I learned more about EE, it seemed like an exciting opportunity to apply mathematical thinking to challenging problems.

Toward the end of my undergraduate studies, I started to realize that I was personally more interested in the “informational” side of EE than the “electrical” side, and so I decided to go to graduate school to learn more about digital signal processing.

In graduate school, I really began to appreciate the experience of working on difficult research problems that would help shape the future of technology. When I completed my Ph.D., I decided to pursue an academic career so that I could not only continue to do research but also help train the next generation of electrical engineers. This has been a wonderful job, and by far the best part has been getting to meet and exchange ideas with so many amazing students and faculty.

What are you currently working on?

My research focuses generally on mathematical models and algorithms for signal and image processing. One of the classical problems that falls within this scope is that of data compression: once a digital camera records millions of numbers describing a scene, for example, how can a computer compress that information efficiently (e.g., into a JPEG file) so that it takes less space to store and less bandwidth to communicate? One of the very interesting research fields that has branched out of the data compression problem is known as Compressive Sensing (CS). In CS, we aim to design new sensors/imagers that directly acquire signals and data sets in compressed form, without the need for a wasteful two-stage “sense then compress” approach. CS has been rapidly evolving over the past several years, and it has drawn not only from researchers in EE, but also from those in mathematics, statistics, and computer science.

What are your favorite hardware tools that you use?

My work tends to focus on theory, algorithms, and simulation. However, one collaboration I was involved in led to the creation of a “single-pixel” camera that uses a digital micromirror (DMD) device to collect an image using many fewer measurements than there are pixels in the image. More information on the “single-pixel” camera is available here.

What are your favorite software tools that you use?

I mostly use Matlab for data analysis and for prototyping algorithms, but I work with colleagues who are skilled in Mathematica, Python, C, etc.

What is the hardest/trickiest bug you have ever fixed?

I once had a Matlab program that would produce a different output every time I started Matlab and ran the program. I could not figure out why this was happening—there was nothing that was random or that was changing in the code! Eventually, I learned from Mathworks that the Matlab FFTW library optimizes its choice of FFT function each time you start the program (presumably to find whichever one is fastest given the current state of your machine). However, each of the candidate FFT functions has slightly different rules for rounding numbers near machine precision. So technically, it is possible to run the same program over and over again and get slightly different answers each time!

What is on your bookshelf?

Textbooks on various signal processing and applied mathematics topics: digital signal and image processing, communications, probability, random processes, wavelets, linear algebra, and differential geometry.

Do you have any tricks up your sleeve?

Although it is not my own personal trick, if I could mention a concept that is extremely useful but not widely known, I would recommend the singular value decomposition (SVD). The SVD is arguably the “best kept secret” in linear algebra because most undergraduate linear algebra courses (widely regarded as mystifying!) don’t have time to cover it. However, the SVD beautifully ties together many fundamental linear algebra concepts, and it is critical to several advanced topics in signals and systems analysis.

What challenges do you foresee in our industry?

In the coming years, it will be essential to encourage more college students to major in the STEM disciplines (science, technology, engineering, and math). In the case of electrical engineering, young students need to know how broad the field of EE really is, how fundamental it is to so many areas, and how multidisciplinary, stimulating, rewarding, and meaningful EE work really can be. Just to give a couple of examples, we will continue to have a need in the foreseeable future for well-trained engineers to tackle challenging problems in developing renewable energy solutions, and in managing the vast quantities of data that arise in our increasingly information-hungry world.

Previous Spotlights

x
Like free stuff? Enter Here!
EEWeb Weekly Giveaway Sponsored by Mouser This Week: 4D Systems uLCD-32PTU 3.2" Intelligent LCD Module w/Touch
Enter Here
Login and enter if you're already a member.
Click Here