Interview with Ehsan Afshari

Featured Engineer

Ehsan Afshari

Ehsan Afshari - Assistant Professor of ECE at Cornell University

image: My group at Cornell, we work on two general areas: high frequency (terahertz) and low noise

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

When I was in high school, I found calculus fascinating. In particular, I was amazed by all different things that one can do with integration. So I decided to make a simple machine that takes the integral of a curve. The idea was to move a pen on the curve and the system finds the integral. Initially, I designed a mechanical version using gears. It was expensive and not very precise. Then I decided to make an electrical version. This was a lot faster and more precise. After doing this project, I wanted to continue in electronics.

Why did you choose electronics?

Electronics is a fascinating field for research. One major aspect that has always attracted me is that the theory behind electronics is very methodical. There are a lot of problems that can be modeled and solved mathematically, yet electronics is not an abstract field. Theoretical findings can be tested in the lab, and they most often agree with the experimental results. And when theory and practice do not agree, and there is no error in either the theory or the applications, there is some phenomenon to uncover, analyze, understand, and utilize.

Another reason I am very interested in circuit design is that it could provide a great platform to demonstrate and verify mathematical and physical theories. For example, a two dimensional LC lattice is a perfect medium to observe nonlinear wave propagation phenomena such as formation of Solitons, Kinks, Kelvin-Helmholtz instability, and nonlinear Schrödinger equation. It is also possible to develop new phenomena in electrical circuits like electrical funnel and lens which have many applications such as power combining and ultra fast specialized computations.

What are your favorite hardware tools that you use?

My group at Cornell works on two general areas: high frequency (terahertz) and low noise.

We design novel circuits that can operate beyond the conventional limits of frequency, power, or noise. To do so, we need to test devices circuits from DC to 500GHz and above. This requires sophisticated equipment that can accurately measure different parameters at very high frequencies with high bandwidth. Recently, we measured an oscillator that generates 0.16mW of power at 482GHz in a standard 65nm process. This was a very exciting and expensive testing that set the world record of generating power using integrated circuits at this frequency range.

What are your favorite software tools that you use?

We use different software for different aspects of our projects. The initial ideas are simulated in MATLAB. We then use SONNET and Ansoft HFSS to simulate the passive structures such as inductors and transmission lines. The circuit simulation and layout is done in Cadence. Finally, the papers and reports are made with LaTex!

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

A few years ago, I was working on a SiGe power amplifier at 85GHz which was not working properly. I discovered that the problem was the wiring inside the chip. Using a laser trimmer, I found the best place to fix the connections. Looking back, the process of finding the error and fixing it was very intriguing, even though at the time it was frustrating.

What is on your bookshelf?

Beside the technical books, for my personal reading, I enjoy philosophy. Particularly, I am interested in epistemology and have many books in this area.

Beside these two areas, if I have time, I like to read books such as the “calculus of friendship” by Steve Strogatz (another colleague at Cornell).

Do you have any tricks up your sleeve? (special way to analyze circuits, special process you use to make something, etc.)

The best trick is to know how to do research. Research is the art of looking at what everyone else has looked at, and seeing something new. By learning how to see a problem, and developing a process for resolving it, one holds the key to solving almost any problem; this makes the way one tackles a problem as important as the problem itself. In my opinion three elements are necessary for successful research. After having gained general knowledge in an area, one must begin to explore various sciences to have a broader view of the problem at hand. The second element of success is the ability to step beyond what is conventional and to dream of new possibilities; this is a necessary quality for those who would lead in their field. And lastly, when an insight is gained, one must delve as deeply as possible into the problem and examine new ideas by relevant experiments to achieve a discovery. It is always useful to go back and forth between these phases, or step aside and look at the big picture before going back to details.