Electrical and Computer Engineering Newsblog

Happy New Year! Today marks the end of Fiscal Year 2017 at Michigan Tech, as it does for many other universities and businesses. This is the year boundary that really matters for anyone doing accounting or record-keeping at the university. For the past couple of weeks, a lot of staff members have been hard at work, making sure our financial house is in order. In July we will start the process of looking back at the past year, preparing year-end reports, and seeing how we did relative to a lot of different metrics. Of course, at the same time we are starting all over again with FY 2018. It’s the circle of life.

Earlier this month I offered some views on our fundamental motivation for being in this line of work – why we teach and why we do research. Today I thought it would be good to take a look at the interplay between teaching and research in the university setting.

In one of those earlier posts I made the observation that we are not a business, rather we are an institution that serves the public good and as such we have multiple stakeholders that we try to keep happy. That same notion about multiple stakeholders holds true at the individual faculty member as well, and if not managed properly it can lead to a lot of stress. In one sense the faculty members are accountable to only one person – me, the department chair – but in reality their performance depends in large part on keeping a lot of other people happy. On the teaching side, there are the students of course, one of our most important constituencies, and sometimes the parents, who generally only surface when things are not going well. There are also department colleagues, as we depend on each other to teach all the necessary prerequisite material for the next course or courses, to make sure a course plays its proper role in the overall curriculum, and to provide documentation needed for accreditation. In research, faculty are held accountable by their external program managers, who often do not understand that we have multiple obligations, by their national and international professional colleagues who provide anonymous peer review of the work, and by journal editors and conference organizers who expect timely compliance with paper reviews and other research-related activity. It’s a lot to juggle!

Even if we simply divide our activity into two broad areas, teaching and research, it can be a struggle to find the right balance between the two. They are often seen to be in conflict, two polar opposites competing for our attention. Students wonder why the faculty are wasting time doing research when they should be available 24/7 for questions and concerns. Research sponsors wonder why faculty are putting so much into teaching when they should be setting the world on fire with their latest scholarly achievements. Faculty members themselves are conflicted, feeling that they enjoy one activity while getting messages that they should spend more time on the other. Sometimes those are mixed messages, since at a place like Michigan Tech there are multiple gatekeepers for promotion and tenure, and there is the concern that different people have different opinions about what is important.

I believe that the answer to this conflict is not to see this as a conflict at all. Even though this is hard to pull off all the time, I still believe in the old-fashioned notion of the teacher-scholar, the person who has a high-level research or scholarly program in his or her own right, and is passionate about educating the next generation of students to make their own contributions to the field. This is really where the magic happens at a university. People who are brilliant scientists, engineers, mathematicians, or thinkers in any discipline, and are not jerks about it but instead really care about students and their education, are like gold at a place like Michigan Tech. The trick to making this work is to see that teaching and research are not pulling in opposite directions but are actually two sides of the same coin – the quest for new knowledge.

A strong research program can have a beneficial impact on one’s teaching. Sometimes we think that the teacher brings the results of his or her latest cutting-edge research into the classroom, keeping students current and motivated, but actually I do not think that is completely correct. Especially if we are talking about undergraduates, most cutting edge technology is beyond them – they are just not ready. After all, the faculty member has a head start on them by at least five years and probably more like 20 or 30. I think the real value of the research program, as it applies to teaching, is that it allows the teacher to know what is important and what is not in the fundamentals. In fact, it gives the faculty member the certainty that the fundamentals really are important, and that certainty will lead to clarity and passion. Sometimes we have to say “trust me, you really need to know this and you will thank me someday”. I get that that does not always work without some taste of good things to come. Here the faculty need to lead by example, demonstrating the kinds of things that can be accomplished if you follow their lead, without overwhelming students with details beyond their comprehension (the so-called “fire hose of knowledge.”)

The mutual benefits of teaching and research go the other direction too. Some of the key attributes of good teaching are, one has to be prepared, one has to be organized, and one has to communicate effectively. The discipline that comes with doing those three things on a regular weekly schedule can pay huge dividends in research programs, where often there is not the same pressure to break one big task down into lots of little tasks. The best teachers are the ones who know how to explain difficult concepts clearly, and clear communication goes hand-in-hand with clear thinking. I have often had the experience where just talking about some problem I am wrestling with leads to new and better ways of thinking about it. Put another way, in order to understand a problem one needs to be able to explain it well, and if you do not understand the problem chances are you are not going to understand the solution. Again, it all comes back to the fact that seeking knowledge and communicating knowledge are really not all that far apart; quite the opposite, they are complementary.

The way we learn things and the way we explain them are often quite different. When I think I know something pretty well, I can lay it out in a linear fashion: I say “here is concept A, which leads to concept B, which in turn implies concept C.” Mathematical proofs usually work this way. If I have been working on something for a while, and having some success, it is very satisfying to put things down in a neat set of notes with the proper flow of one idea into the next. The problem is, with 99% certainty that is not the way I learned the material. Usually I learn things in a more circular fashion, going forward and back and sometimes in random directions, figuring out bits and pieces and eventually figuring out how they are linked together. When the pieces are in place, and I want to convince someone of my results, then my explanation will be nice and linear. This is precisely how most of us organize our lectures and our courses: a nice logical flow from the beginning to the end. Actually I think this is perfectly acceptable. We just have to understand that our students, just like us, are not going to learn the material that way. Instead, they will get part of the lecture, then through homework, labs, and studying for tests they will go around and around in circles until it starts to make sense. Perhaps the goal should not be to have students comprehend a topic in that nice linear fashion from the very beginning, but rather to come to a linear understanding of that topic in the end. My point here is that this circular or random nature of discovery/learning, and the linear nature of understanding/explaining, are quite complementary and are mirrored in the way we do teaching and research.

Bringing this closer to home, I thought I would brag a little bit about one of our own. Tim Havens, an associate professor with a joint appointment in ECE and Computer Science, has found the sweet spot when it comes to balancing teaching and research. He is one of our most active researchers in the ECE Department, with a portfolio of funded projects in computational intelligence and signal processing totaling about $250k per year in research expenditures. He is the Director of the Center for Data Sciences, within the Institute for Computing and Cybersystems, and is also the Director of the non-departmental MS in Data Sciences professional degree program. As a teacher, he can cover just about anything in the computer engineering curriculum, from sophomore-level digital logic design to graduate-level machine learning, and he always gets outstanding student course evaluations. Having been to several of his recent graduate student thesis and dissertation defenses, I am impressed by the quality of his students’ work, and by the level of enthusiasm and camaraderie among the students in his research group. He is an outstanding example for all of us. Tim, feel free to use the comment feature of this blog if you want to tell us how you do it.

Next week is Fourth of July – have a safe and happy holiday!

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University

A very happy midsummer to all from the northern reaches of Michigan! This is the season for long days in the Keweenaw, and I thought it would be fun this week to explore some of the basic mathematical facts about sunrise, sunset, and the length of days, and throw in a little signal processing to boot.

First off, while the days are long this time of year, what makes them seem longer here is the timing of sunrise and sunset. Yes, we are pretty far north compared to most of the 48 contiguous states, but we are not really that far north. At 47 degrees N latitude, we are at about the same latitude as the northern tip of Maine, we are slightly south of Seattle, south of most of Canada and all of Alaska, and well south of all of Great Britain and Scandinavia. Our longest days in the summer are about 16 hours, and the shortest days in the winter are about 8 hours. There are plenty of places on the globe with greater variation in the length of day than that. The reason we think the days are so long right now is because of a quirk in the time zone map. Like all but three counties in the Upper Peninsula, Houghton County is in the Eastern time zone, despite the fact that we are slight west of Chicago, which is in the Central time zone. The story goes that we are on Eastern time so that we would be in the same time zone as the bankers and mine owners on the East Coast, 100 years ago. As a result, this time of year the sunset occurs around 9:50pm, and twilight extends for another hour or so after that. For those of us working at Tech and leaving around 4 or 5pm, it’s like another whole day to play outside.

This year the summer solstice occurred on Wednesday, June 21. While we often think of the solstice as a day, in fact it is a particular moment in time when the Earth’s axis of rotation is most tilted toward the Sun. At that instant, the axis of rotation is co-planar with the axis of revolution of the Earth around the Sun, and the Sun shines directly down on the Tropic of Cancer. This year the solstice occurred at 12:24am EDT, on Wednesday, July 21. The time of the summer solstice moves forward about 6 hours, or one quarter of a day, each year, as the period of revolution of the Earth around the Sun is about 365-1/4 days. The 1/4 day is why we have a leap year ever four years, and on those years the time of the summer solstice moves back 18 hours from the previous year. Oddly enough, part of the reason we say the solstice occurred on June 21 this year has to do with Daylight Saving Time; if we were on Standard Time the solstice would have occurred on Tuesday, June 20, at 11:24pm. As it turns out the longest day of the year, measured from sunrise to sunset, was actually June 20.

Here is a little-known fact which has fascinated me ever since I discovered it. The longest day of the year does not coincide with either the earliest sunrise or the latest sunset. At our latitude, the earliest sunrise occurs about 5 days before the solstice, and the latest sunset occurs about 5 days after. That means that, at the time of this writing, we have not even seen the latest sunset this year; that will occur on Sunday, July 25, at 9:54:06 p.m. The sunset time is not changing quickly, though: on both June 24 and June 26, sunset is at 9:54:05 p.m. Those who understand the basic concept from Calculus 101, that the slope of a function is zero at its maximum, will appreciate that.

The length of the day is defined as the time between sunrise and sunset, or if we want to do an arithmetic calculation, it is the sunset time minus the sunrise time. The addition or subtraction of two periodic functions that are synchronized in time is an important concept from the course I teach, EE1110, Essential Mathematics for Electrical Engineering. There we consider a particular class of functions, called sinusoids, and show that as long as two sinusoids have exactly the same frequency, then the sum or difference will also be a sinusoid, and furthermore there is a straightforward algorithm to figure out where the peaks and valleys of the sum (or difference) will be relative to the peaks and valleys of the signals being added or subtracted. In the case of the sunrise and sunset times, we already see that the earliest sunrise and the latest sunset are offset by about 10 days at our latitude, and that the longest day will occur somewhere in the middle.

Thinking there might be an interesting connection between electrical engineering and astronomy, I figured I would just go ahead and look at the numerical data in MATLAB and see if I could use it to illustrate EE1110 principles. There are lots of places on the Internet to find sunrise and sunset data times; here is one operated by the U.S. Navy: http://aa.usno.navy.mil/data/docs/RS_OneYear.php. What is nice about this site is that it provides the data for an entire year, in a format that is easy to cut and paste into an Excel spreadsheet. So, that is exactly what I did: I put the 2017 data into Excel, then imported it into MATLAB, then reformatted it so that times are expressed in minutes (from midnight) and kept everything in Eastern Standard Time. I also got rid of the months and dates, simply numbering the days sequentially starting with Day 0 being January 1, 2017. All of that took longer than it should have, but now I have the data conveniently in a .mat file.

The upper panel in Figure 1 below shows the time of the sunrise (in green) and sunset (in red), measured in minutes from midnight, Eastern Standard Time, as a function of the day, for the entire year 2017. In the lower panel I show the length of the day (in blue), in minutes, which is simply the sunset function minus the sunrise function. For point of reference, one full day is 1440 minutes.

Here is where I got the first of three surprises in this little exercise. The sunrise and sunset functions are quite asymmetric, in the sense that they do not look the same when you flip them upside down. The latest sunset occurs after the summer solstice, whereas the earliest sunset occurs before the winter solstice, which means that the time from a peak to valley is considerably shorter, like 20 days, then the time from a valley to a peak. We see the same behavior in the sunrise data. Now the symmetry of sinusoids is important to a lot of the EE1110 theory, and because of the asymmetry issue we cannot use sinusoids to model sunrise and sunset data. Consequently, the idea of using sunrise and sunset times as an illustrative example of EE1110 concepts is out the window. Dang!

We are not done yet, however. As can be observed in the lower panel, the length of day function does exhibit symmetry, in fact it looks downright sinusoidal. So, I thought maybe we could throw some of our signal processing tools (well beyond the scope of EE1110) at this data and see if we can determine the period, or time for one complete cycle. To make this a little more accurate, I decided to look at four consecutive year’s worth of data, from 2017 to 2020. This data is shown in the Figure 2 below, which is essentially the same as Figure 1 except it goes for four years. To compute the period, or more precisely the frequency (the inverse of the period, in cycles per day), I used a common technique from signal processing of computing the Discrete Fourier Transform (DFT) of the data, using an algorithm called the Fast Fourier Transform (FFT), and looking for the point at which the DFT reaches its maximum. For those following along at this point, I subtracted off the mean of the data, and zero-padded it out to 65536 data points before computing the DFT. Doing these kinds of calculations in MATLAB comes very easily to me after many years of signal processing research; it’s the kind of stuff I can sit at my desk and bang away and have it work right the first time.

Except…I made a crucial mistake, and got the second surprise. The absolute value of the DFT of the length-of-day data is shown in Figure 3 below. The horizontal axis has units of frequency, in cycles/day. I was able to zoom in and find the frequency at which the DFT reaches a peak, and that value is 0.002762 cycles/day. 1 over this should be the correct period for one revolution, right? Wrong. 1/0.002762 = 362.06 days. I knew that can’t possibly be right – the period should 365.25 days. Where did I go wrong? It turns out I fell into a common trap (that I often rail against) of using the DFT without thinking carefully about the interpretation of the results. I had “known” forever that the best way to determine the frequency of a single sinusoid is to the compute the Fourier Transform and look for a maximum. That result is part of the collective wisdom of everyone in signal processing, and goes back at least to the often cited paper by D. Rife and R. Boorstyn, “Single Tone Parameter Estimation from Discrete-Time Observations,” IEEE Trans. Information Theory, September 1974. Well, I went back to that paper and found my error. Rife and Boorstyn consider the frequency estimation for a function called a complex exponential, sometimes called a complex sinusoid. (EE1110 students know all about complex exponentials, right?) For complex exponentials, computing the Fourier Transform and looking for a peak is exactly the right thing to do. However, a real sinusoid, like our length-of-day data, is actually the sum of two complex exponentials, one at a positive frequency and one at a negative frequency. The Fourier Transforms of those two complex exponentials can interfere with one another in such a way that the peaks can be shifted from what we would consider the correct location, in this case some 3.2 days (or the equivalent error in frequency). After some reflection I realized that the only way to really get the frequency right – that I could think of, anyway – is to do what is called nonlinear least-squares estimation, which essentially means looking exhaustively across all sinusoids for one that comes closest to matching the given data. Without going into too much more detail, I did exactly that for my length-of-day data and came up with a frequency of 0.002738 cycles/day, which corresponds to the period I expected, 365.25 days.

Last observation, and last surprise. I mentioned above that, before taking the Discrete Fourier Transform, I subtracted off the mean value. Out of curiosity, I went back and looked at that mean value; it was 734 minutes, or 12 hours and 14 minutes. Hold on, I thought – how can the average length of day be anything other than 12 hours? Every spot on the Earth enjoys equal amounts of light and darkness over one entire year, so the average has to be 12 hours, right? Again, wrong. Thanks goodness for the Internet. I Googled “average day length greater than 12 hours?” and hit on this beautiful little explanation: http://rickbradford.co.uk/DayLength.pdf. The author identifies three separate effects, but the largest and easiest to explain has to do with the non-zero diameter of the disk of the Sun, as seen from the Earth. We define sunrise and sunset as the moments when the Sun just appears or disappears over the horizon, but in fact it might be more accurate to define it as the moment when the center of the Sun disk crosses the horizon. That would bring more symmetry to the definitions of day and night, and shave a few minutes off the time we associate with day. Because of the nonzero diameter of the Sun, more than 50% of the Earth can see at least a portion of the Sun at any given moment, thus making the average length of day greater than 12 hours.

Make the most of these long days and the beautiful weather! The days are already getting shorter.

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University

I am taking a break this week from recent discussions of relatively important topics like why we do what we do in academics, so that I can vent about one of my pet peeves. In the big scheme of things, today’s topic is totally unimportant, but sometimes things like this occupy more of our attention than they rightfully should so it’s good just to put it out there. I am referring to a particular issue of grammar that comes up quite often in academic circles.

I am amazed by the number of people who are willfully unfamiliar with the correct use of the word alumni. I am talking about people who should know better: students, faculty, our local daily newspaper, and yes, sometimes, professionals at Michigan Tech who work in communications, development, and alumni relations.

The word alumnus is a noun that comes from Latin, and like all Latin nouns it has what is known as a declension, that is, it has various forms depending on number and gender. This is sort of like conjugation, except that conjugation is applied to verbs, with various forms that depend on who, when, and whether or not the action is hypothetical.

Here is the correct (nominative) declension for the noun alumnus:

Alumnus  Singular, male or unknown
Alumna    Singular, female
Alumni     Plural, male, mixed or unknown
Alumnae  Plural, female

We have all heard or read it: an individual person is described as “an alumni” of some institution. For me this is like fingernails on the chalkboard. I must have said this a hundred times, and I guess I will keep saying it as long as I see the error being made, which is probably forever: ALUMNI IS PLURAL. ALUMNUS IS SINGULAR.

[Aside 1: The full declension for a noun includes a lot of other variations, like whether the noun is a subject, an object, or turned into a modifier. Most of those variations apply to the original Latin, not the way the derived word is used in English. In the above I am sticking just to the nominative case; I don’t want to get too far distracted here.]

[Aside 2: Have you ever actually scraped your fingernails on a chalkboard? I did it once, in junior high, and it is really awful. Now go back to what you were doing and try not to think about it.]

I recognize full well that there are two schools of thought regarding grammar. There is the constructivist school, which posits that grammar should follow carefully prescribed rules of usage that are written in stone forever, and the pragmatist school, which says that languages are alive and evolving, and that grammar is whatever the users say it is. I am kind of in the middle; I can tolerate some changes to the language if they make sense, like split infinitives. On the issue of alumni, however, I am a strict constructivist: anything other than the use of the word as given in the table above is flat-out wrong, I don’t care who you are or how many times you have used it incorrectly. Repeating a lie a hundred times does not make it true.

Some people skirt the issue of gender by using the more informal term alum. I think of this as a kind of slang used by people in development, and that’s fine. I am pretty certain that the term is singular and that the correct plural is alums. I have sometimes heard people use alum in the plural, like deer or moose, but every time I do, I find it jarring and think that it has to be wrong. Since we are talking about slang, I doubt if there are any strict rules.

Another one of those things that will never, ever be right is “for you and I” or any other use of the subject pronoun I instead of the object pronoun me in a prepositional clause. Somewhere along the line, people began to think that I sounds more sophisticated and correct than me, in all uses, and this error crept into the language. Again, we hear it all the time.

Here is another common error, still wrong but leading more to raising of eyebrows rather than gnashing of teeth. I often see the word itinerary when people are putting together a schedule for visitors on campus. An itinerary is a schedule for travel, like airline flights and hotels. It comes from Latin roots meaning a journey or a road. A schedule for activity in one place is simply that, a schedule. A schedule for a meeting is an agenda, although that word can have other more subtle meanings depending on the context.

Having gotten all that off my chest, I will close with a couple of items that are more matters of legitimate debate. The first is the pronunciation of the word alumnae, which comes up fairly often at Michigan Tech since we have an organization called the Presidential Council of Alumnae. There are two accepted pronunciations, alum-NEE and alum-NAY. There is not really a whole lot to debate here; it is more of an “agree to disagree” situation between the advocates of the respective positions. I am good either way.

My last item has to do with what we call someone who has served with distinction on the faculty, has moved on to retirement, is recognized by the institution, and happens to be a woman – is that person a Professor Emeritus, or Professor Emerita? The strict constructivists would say that Emeritus is correct since the choice has to do with grammatical gender, not biological sex, and since Professor is a masculine noun (in Latin) it should take the corresponding masculine modifier. The pragmatist can legitimately say yes, but how did Professor become masculine in the first place? When our very own Martha Sloan retired a few years, we came down on the pragmatist side and are proud to call her the very first (and only, so far) Professor Emerita in the ECE Department. I’m good with that one too.

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University

In my last post, I raised the question of why we do research in a university setting.  My main point, which I hope I made clear enough, is that we should have research programs because we are passionate about the work itself, and not for some other reason such as supporting a PhD program.  While PhD programs are an important part of what we do, I believe strongly that we cannot be effective advisors and mentors for PhD students—training our replacements, essentially—unless we are good at research in the first place, in our own right.

Having addressed that question, an obvious companion question came to mind—why do we teach?  At first, my motivation for addressing this was only out of symmetry.  Normally, when I realize I am about to do something for that reason, I remember the old adage “foolish consistency is the hobgoblin of little minds” and then go ahead and do it anyway.  In this case, however, raising the question itself was an opportunity for me to explore some of my own life choices.  This was one of those cases where, having decided on a topic, I was not entirely clear on the points I was going to make until I was done writing! Hopefully you will find a little more clarity below, and if not I hope this might stir any readers who are teachers to ask the same question of himself or herself.

By way of context, I should mention that I come from a long line of teachers, including several math teachers.  My grandfather on my father’s side was a farmer in Oklahoma and Kansas, and in the 1910s and 1920s his primary income came from teaching in a one-room schoolhouse in western Oklahoma.  The handbell he used to call students to class after lunch and recess still sits on the piano in my home.  My grandmother on my mother’s side taught grade school in a white community adjacent to an Indian reservation in eastern Washington, during World War I, before she married my grandfather and moved to Oklahoma.  My mother took up teaching as a second career in the 1970s, and was a math teacher in the Tulsa public school system and later at Tulsa Junior College.  I have one cousin who teaches high-school math in a suburban Tulsa school district, and another who recently retired from teaching high-school math in Fremont, California.  I have a nephew who is a faculty member in math at Glendale Community College, in Los Angeles, a niece who taught high-school math in Niwot, Colorado, a niece who teaches middle-school English in Norman, Oklahoma, and is married to the high school debate coach, and the wife of a nephew who taught elementary school in another Tulsa suburb.  I guess it’s in my DNA.

This is as good a place as any to mention that my father was an electrical engineer and my mother was a math teacher, and for the past five years I have put a lot of effort into teaching a large freshman class called “Essential Mathematics for Electrical Engineering.”  Sometimes the universe does offer up beautiful symmetries, if we are paying attention.

But back to the question of why we teach in the first place. Here is my top 10 list.

1. We are good at it.  Some people just have a knack for communication and connecting with others, for organizing and presenting their ideas clearly and concisely, for showing that what they are trying to do really matters.  I have heard it said that “before someone cares what you know, they have to know that you care.”  Teachers who can pull this off have found the right line of work.

2. Someone pays us to do it.  This often goes along with #1 above.  If there is something that we do reasonably well, and others are willing to pay us for it, chances are it is something that we find enjoyable and rewarding.  At the very least it gives us a reason to get out of bed and put one foot in front of the other.  A paycheck might not be world’s best motivation, but it is not the worst one either.

3. It keeps us young.  This was something I noticed when I first entered graduate school—all of my professors seemed younger than their counterparts at the same chronological age in industry.  There has to be something about the constant contact with those from an earlier generation that allows us to continue seeing the world with fresh eyes.  This doesn’t really address the problem of the ever-widening gap between my generation and that of our students, but I don’t think anyone is going to solve that one.  Time’s arrow moves in one direction.

4. Benefits to the individual students. At a place like Michigan Tech, our teaching programs are built around improving the lives of our individual students, giving them the skills and the knowledge that they need to be successful, personally, professionally, and financially.  Of course we have to hold our students to a high standard; otherwise our efforts are meaningless.  At the end of the day, however, our placement rates and starting salaries speak for themselves.  We aim as well to provide students with the means for lifelong learning and an ability for critical thinking that will serve them well in all aspects of life.

5.Workforce training and economic development.  Going hand-in-hand with the benefits to our own graduates, as individuals, is the benefit that those graduates bring as newly trained engineers to our state, our region, and the nation as a whole.  As I have pointed out before, Michigan Tech is unique in Michigan in that our obligation to the welfare of the various industries in the state is called out in our founding legislation. Placing properly educated engineers in positions of responsibility is a big part of what we do. The economic situation in Michigan has had its ups and downs, but right now things are looking up, and we need to do our part to continue moving in the right direction.

6. Benefits to society at large.  The benefits of a college education to society have long been recognized. Thomas Jefferson noted that democracy and self-government could not function without an educated populace capable of making well-reasoned decisions, and he used that as an argument for education reform in our new nation.  Benefits that go beyond economics in modern society are well-documented, see e.g. the recent report by the College Board entitled “Education Pays 2016.” A college degree is associated with a healthier lifestyles, regular exercise, reduced healthcare costs, greater engagement with family members, community volunteer activity, and higher voting rates. It’s hard not to want to be a part of that.

7. It’s the best way to learn a new topic. This is one of those academic truisms that gets played out all the time.  If a faculty member wants to steer his or her research or scholarship in a new direction, the first thing we try to do is teach a new course in that area.  This forces us to learn the subject thoroughly, in an organized way, and we are put on a schedule with a weekly deadline where we have to stand up in front of a classroom and defend what have learned.  One does not understand a subject until one can explain it clearly to someone else.

8. Fundamentals are important.  Just like teaching a new course is good way to learn a new topic, it is also true that teaching a familiar topic is a good way to maintain a healthy intellectual discipline.  This is an important message to carry to those outside of academia with whom we work, as well as our graduates about to enter the workforce.  We sometimes hear this narrative that material taught in the classroom is irrelevant, and that the “real” technical knowledge is what one learns on the job.  I emphatically reject that notion.  While an industrial perspective can be quite valuable for our educational programs, I have also seen situations in which technical projects go awry when engineers lose sight of the very foundations of their field.  In engineering, just like in baseball, the fundamentals never go out of style, and the most successful companies and organizations know that.

9. We have nothing better to do.  It goes without saying that this is the absolute worst reason to teach. Unfortunately, it can happen that faculty members who are unproductive in other areas find themselves with increased teaching responsibilities. In the worst-case scenario these responsibilities are assigned as a form of punishment. (I do not subscribe to this management philosophy!)  It is as if teaching is the fallback position, our “day job” as it were, like waiting tables or driving for Uber. (I hasten to add that here is absolutely nothing wrong with waiting tables or driver for Uber if that is what brings satisfaction or what one needs to do to make ends meet.)  In a future post I hope to take up the relationship between teaching and research, but for now I will just say that this assumption—that we can always teach—does not help to counter the myth that research is more important than teaching.  Our academic workload models tend to be built around the notion that we do as much research as we can find support for, and whatever time is left over is then devoted to teaching.  Part of me wishes that faculty had to compete for the right to teach courses (and get paid for it), the same way we have to compete to do research through writing grant proposals and the like.  The good news, at least in the ECE department at Michigan Tech, is that this sort of default teaching is pretty rare.  It is true that we have a few faculty members who are not research active, but most are outstanding teachers and would win hands-down any competition that would earn them the right to teach as much as they do.

10. We want to make the world a better place.  This is my half serious, half glib, reason that encompasses a lot of the other good reasons I give above, and I add it here because I really did not want to end with #9.  It is the same reason that I gave in my last post about why we do research.  In teaching we make the world a better place through leverage, by giving large numbers of young people the tools that they need to go out and improve life for themselves and for everyone around them.  I will admit, this is the viewpoint of the eternal optimist, but that is a criticism I can live with.

Although not utilized very often, this blog post does have a comment feature.  Teachers out there, I would love to hear from you.  Why do you teach?

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University

A very happy Memorial Day Weekend to everyone. The weather here has warmed up nicely, all of a sudden, just in time for long holiday weekend and the unofficial beginning of summer. Over the years this has become one of my favorite holidays. For academics it is a time when we can look forward to the summer in anticipation of all of our great progress on projects we have been putting off, before reality and humility hit home in August. Back when I lived in Missouri, it was the start of the summer boating season at the Lake of the Ozark, and here in Houghton I am happy just to see everything green again after a long winter. I guess the main reason I enjoy this weekend so much is that my wife and I were married on Sunday of Memorial Day weekend, 22 years ago this year, and we celebrate our anniversary on Sunday independent of the date – although this year the 28th does fall on Sunday and when that happens it’s even nicer. Sunday is the day of the Indianapolis 500, by the way, so even though I am not a big racing fan it serves as a pleasant reminder of that happy day.

Last week I decided to raise the “big question” – asking why we do what we do. What is it that gets our juices flowing, gets us excited about coming to work, gives us some purpose in life? Obviously I cannot answer that for all the individuals in the ECE Department, but I certainly hope that everyone would have a good answer, and be able to come up with it pretty quickly too. A big part of my job is to connect those individual passions with the group goals, whether we are talking about the strategic plan for the ECE Department or the mission and vision of Michigan Tech as a whole.

The question I want to address today is, why do we do research? One would think that would not be too hard to answer, but actually over the past few years I have heard a lot of conflicting opinions about this. I have also come to my own conclusion about this, and now hope to take advantage of my little bully pulpit to make my case. If I am right, then I think that ultimately we may be able to put policies and procedures in place that serve to strengthen our research programs, and at the same time strengthen everything else that we do.

Here is my answer: we do research to make the world a better place. Now that may seem a little trite and simplistic, but I mean that in the broadest possible sense. We make the world a better place in a lot of different ways, and what I really mean by that statement is, the research itself is primary. We do research because we believe that the results of our research will have some tangible benefit to humankind, whether it is advancing our understanding of how the world works, or it solves some intellectual puzzle in mathematics, or it leads to technological advances that improve the human condition. Everyone who does research properly knows the reason behind the research, and the range of possible applications. Whatever problem we are trying to solve, the solution to that problem should be our primary motivation, and if we are lucky enough to find it we should go home happy.

The point above will be made clearer if I state some reasons that I believe are the wrong reasons for doing research. Here is my list of three misguided reasons; there may be others: 1) we do research so we can have a PhD program, 2) we do research for faculty development, 3) we do research because someone tells us we have to.

Let me take the PhD issue first, because this is where I find the greatest confusion and misunderstanding, and the greatest difference of opinion. Virtually all great research institutions have strong PhD programs, and the two go hand-in-hand. Since a lot of research is carried out by graduate students, it is easy to see why one might be led to believe that the research activity exists to serve the PhD program. I maintain that nothing could be further from the truth. A successful research program should be led by experienced faculty, who are doing the work out of their love for the field itself, as described above. In the process, if they so chose, they can bring PhD students into that research activity, and serve as an example for them. As one of my former colleagues at Washington University, Marcel Muller, used to put it, a PhD program is an apprenticeship in research. A graduate student comes to the university, lines up with a research advisor, and learns all about how research is done from that personal one-on-one relationship. We must remember, however, that the PhD is still an educational program, and it exists separate from the research. That is why we have specific milestones along the way to the degree, such as qualifying exams and dissertation proposals, and why we have committees to make sure that those milestones are being met. At an educational institution like Michigan Tech we may view training PhD students as an important part of our mission, which it is, but we are making a mistake if we take that attitude too far and do all of our research vicariously through the students.

A strong research program can accomplish a great deal even without PhD students. Many successful programs include personnel at all different levels, including post-doctoral research fellows and full-time research engineers or research scientists. I would argue that a tenured or tenure-track faculty member, who spends a lot of time as a research supervisor, should always have some project that they consider theirs alone, that they can work on without the collaboration of students. A research organization should be like a music conservatory, where the leaders are performers as well as teachers. Perhaps another good model is the teaching hospital, where the mentors for the next generation of doctors are all practicing clinicians and surgeons in their own right. The PhD program is just one component in the research mix, and prospective PhD students should be looking for opportunities to learn the craft of research from the masters of the craft.

The second flawed reason I give for doing research is faculty development. Here I mean that doing research so that faculty members have something stimulating to do, that they stay current in the field, have something to put on their CVs, or so that they have a reason to go to conferences and interact with their peers. All of those are perfectly valid things to want to do. However, all of those activities serve to further the research agenda, and should not be seen as a reason for the research activity in the first place. Just as in the case of the PhD students, the latter is putting the cart before the horse. One might argue that doing research makes us better teachers, and while that may very well be true, it is still not the the right reason to do research. The ideal scholarly situation occurs when research and teaching co-exist in a sort of symbiotic balance. It’s not always easy to pull off, but when it’s working that’s when the magic happens.

The absolute worst reason to do research is to do it because someone else wants you to do it. Any such research program is doomed to mediocrity. Anyone who joins a research university as a teacher, and complains of a “publish or perish” culture, or thinks of teaching as the real work while research is just something we do to entertain ourselves, probably ought to find another place to work. If we are doing research because we want to make the world a better place, then we are motivated to publish so that we can tell the rest of the world exactly how we have done that. Research is a “get to”, not a “have to.” I am not saying that research is more important than teaching, nor am I saying that everyone at a successful university has to focus primarily on research. We need to recognize that we are a diversity community, and strive to make the most of everyone’s talent and passion. The last thing we need to be doing is force people to do things they are not good at or for which they have no motivation. The trick of course is to identify those individuals, through the processes of hiring, promotion, and tenure, whose goals and aspirations line up with those of the institution.

As we work to build our research activity in the ECE Department at Michigan Tech, we may need to go through a period of contraction in our PhD program until the two are properly aligned. Over the past several years we went through a period of intentional growth in our PhD program, but I cannot honestly say that our research program has experienced a corresponding similar growth. Let me be more precise, and also fair: our research program, as measured by research expenditures, has not grown substantially. However, our research program, as measured by the number of faculty members with external research funding has grown by quite a bit, and I take that as a very good sign. I intend to encourage all our research-active faculty to be clear in their own minds what really drives them and what they hope to accomplish through their research. Only then does it makes sense to invite PhD students into that activity, and perhaps to excite in them that same sense of purpose. Ultimately, as I stated at the outset, I believe firmly that this will strengthen both the research activity and the PhD program, and secure for the Department the visibility and the recognition it seeks.

That’s my story and I’m sticking to it. Have great Memorial Day everybody!

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University

It’s been a quiet week in Houghton, just like in Lake Wobegon I suppose. It seems like hardly anyone is around except for the few instructors we have teaching summer classes. The weather has been pretty lousy – cold, rainy, and windy – and even though the lawns around town are greening up, the leaves on the trees are still struggling to come out. The academic year is over but it is too early in the season to enjoy any summertime outdoor activities in the Keweenaw. It’s a perfect time to travel.

This is also a good time to take a breather to step back and think about the bigger picture at Michigan Tech. We have the search for a new president coming up next academic year, along with searches for three deans, in the College of Engineering, the College of Sciences and Arts, and the School of Technology. (I hasten to add here, as does our current Dean of Engineering Wayne Pennington: there is no crisis. Everyone just reached retirement age at the same time.) A lot of people are going to be taking a hard look at the kind of university we want to be as we move forward, and I count myself among them.

Thinking about strategic issues and traveling at the same time provides the opportunity to get in some extra reading, in airports, on planes, and by the hotel pool. As luck would have it my wife was reading the book Start with Why, by Simon Sinek, and she loaned it to me for my recent travels to Houston, Seattle, and Tulsa. It is the perfect catalyst to get one thinking about the larger, more important issues in any organization.

Pretty much everything you need to know about Sinek’s book you can get from the title. Essentially, he makes the case that every successful business, organization, or movement knows at its core its reason for existence – the WHY. The HOW and the WHAT will follow naturally from the WHY. If the leaders of the business, organization, or movement can articulate and communicate the WHY to both the members (e.g. employees) and the stakeholders (e.g. customers) then everyone is motivated for the right reasons, and the organization will flourish. He cites Apple, Southwest Airlines, and the civil rights movement under the leadership of Dr. Martin Luther King Jr., as examples of this principle in action. Best line in the book: Dr. King gave the “I Have A Dream” speech, not the “I Have a Plan” speech. If you sit back and think about it, this is not rocket science, but it is an idea that is critically important, and easily forgotten in the day-to-day operations of HOW and WHAT (and yes, Sinek puts those three words in ALL CAPS throughout the book.)

So why does Michigan Tech exist? Good question. There is actually one very good answer, spelled out in the opening section our founding legislation. Here, according to the State of Michigan in 1885, and amended in 1963 and 1964 to change the name, is our raison d’etre:

The institution established in the Upper Peninsula known as the Michigan College of Mining and Technology, referred to in the constitution of 1963 as the Michigan College of Science and Technology, is continued after January 1, 1964, under the name of Michigan Technological University, and shall be maintained for the purpose and under the regulations contained in this act. The institution shall provide the inhabitants of this state with the means of acquiring a thorough knowledge of the mineral industry in its various phases, and of the application of science to industry, as exemplified by the various engineering courses offered at technological institutions, and shall seek to promote the welfare of the industries of the state, insofar as the funds provided shall permit and the Board of Control shall deem advisable.

This is pretty unambiguous: we exist to provide a means for the inhabitants of Michigan to acquire knowledge in the application of science to industry (which I would argue means STEM) and to promote the welfare of industries in the state. [OK, there is that part about the mineral industry which seems a bit dated, although I am certain my friends over in Geological and Mining Engineering and Sciences love it.] In essence, the founding legislation speaks to education and research, and specifically STEM education and industrial research. Close inspection reveals that this paragraph does not say anything about educating students from other states or other countries, nor does it say anything about doing government-sponsored basic research, nor does it say we will promote the welfare of industries in California.

Don’t worry, I am not going to be a strict constructionist here. I realize that our founding legislation is a living document, much like the U.S. Constitution, and that the very changes in society, technology, and industry that we have helped to bring about force us to reconsider exactly what it means to be useful to the State of Michigan. I am happy that we have students from all over the U.S. and from abroad, I am happy that our research portfolio includes a lot of basic science as well as applied science, and I am happy that our graduates have good job opportunities all across the country. One can easily argue that all this activity is good for Michigan citizens and Michigan industry, and besides, the world is much smaller now than it was in 1885 and we need to have a global perspective. Thankfully, we have a Board of Trustees who acts as our “supreme court” and which can interpret our founding legislation in a way that keeps us relevant for the 21st century.

That being said, I am not shy about asserting that Michigan Tech is and always has been a technological university at its core. We need to embrace that identity and not try to run away from it; it’s who we are, it’s what we do, it’s in our DNA. I am also not shy about saying that Michigan Tech has a responsibility to the State of Michigan in some way or another, whether that means providing a pipeline of well-prepared talent in STEM fields or supporting industry through basic and applied research. Lately I have been throwing in the phrase “and the larger Great Lakes region” when I speak or write about our role in the state, because I think we all interconnected now, and what is good for Wisconsin, Illinois, and Ohio is by and large good for Michigan too – and vice versa.

An issue related to our purpose in life occasionally comes up in conversation around the department, when someone throws out the question “Who are our customers?” It took me a while but I now have my stock answer to this question, which is: we are not a business, therefore we do not have customers. We are an institution that serves the public good, and we have many stakeholders. These include our students, our students’ families, our alumni, our research sponsors, our industrial recruiters, our other industries in the state, and the State of Michigan as a whole. There is a whole ecosystem surrounding discovery, innovation, education, and workforce training, and when we are operating at our best these parts are all working together for the betterment of society as a whole. Now it is tempting to say that “students are our customers, they are the ones paying the bills” and it is very easy to see why many students and their parents would adopt this stand. However, this is an unfortunate consequence of the drop in state funding and the subsequent increase in tuition which shifts the financial burden to the students and their families, and I certainly agree that it is substantial. Please don’t misunderstand: we take our responsibilities to our students very seriously. I do want to point out that there was a time when students paid a nominal fraction of the cost of their education, and the rest was borne by the state because the higher education of students who would contribute to economic and social development of the state was a benefit to all citizens, not just those attending college. [This is going off on a tangent, but I recommend reading the editorial in the New York Times Magazine on February 21, 2017, lamenting the loss of the “public” in public schools.]

If we fast-forward from 1885 we can find a more modern version of Michigan Tech’s WHY in our strategic plan, easily found on the website https://www.banweb.mtu.edu/pls/owa/strategic_plan.p_display. There you will find our Mission, our Vision, and our Goals, as developed over several years recently by the administration and the Board of Trustees with lots of input from the entire university community. At first I thought it would be straightforward to map WHY, HOW and WHAT onto Mission, Vision, and Goals, but that didn’t quite work out. In fact, in doing some background reading on mission and vision statements, I found conflicting guidance on what belongs in a mission statement, with different authors claiming it should be WHY, HOW, or WHAT. The one consistent guidance I found was that the mission speaks to the present, while the vision speaks to the future. So, with that little admission of my own state of confusion, I am going to take the university’s Vision as the definitive statement of why we believe we exist now. I am going to make one little modification, and change the future tense to the present tense:

Michigan Tech leads as a global technological university that inspires students, advances knowledge, and innovates to create a sustainable, just, and prosperous world.

I’m good with this. Obviously this statement has much broader reach than the opening paragraph of our founding legislation, but there is nothing in this statement that outright contradicts that original document. If we are successful in all our global aspirations that in all likelihood we will fulfill all our local responsibilities.

There is another little phrase that has been used by the university for many years. It is not our mission or our vision, nor is it an official motto or slogan of any kind; some people simply call it our “tagline.” It pops up on a lot of Michigan Tech promotional material, and it goes like this:

We prepare students to create the future.

This is very catchy and I acknowledge the author, unknown to me, for succinctly capturing a nice idea. Unfortunately, I am not good with this as a statement of the Michigan Tech WHY because it does short shrift to our aspirations in research and our responsibility to support industry. I know, everybody’s a critic.

My whole point in this exploration of the Michigan Tech WHY, beyond just pontificating on someone else’s wordsmithing, is that I think we all need to keep the big picture in front of us at this critical juncture in the life of the university. It is my hope that our new leadership will not only have a compelling vision for the future of the university, but will also work to communicate that vision regularly to the university community. We all need a reason to get out of bed in the morning, and we look to our leaders to give us a better reason than a paycheck. I can get behind inspiring students and advancing knowledge, but so can a lot of universities (all of them, actually) so I want us to do it in a way that is a reflection of Michigan Tech’s special place in the world. We have a lot to be proud of, and a lot to offer. As long as the university community and the rest of the world know WHY that is true then we will be in good shape.

Coming up: I will get further into the weeds of WHY we do certain things in the ECE Department. In the meantime, enjoy the last few days of May.

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University

Greetings one and all from beautiful Seattle, Washington, where I have been attending the 2017 IEEE Radar Conference. It has been a nice change of pace to immerse myself in a technical environment, catch up with some old friends, and think about some problems that I have not visited in some time.

Radar is an interesting field. In many ways it is the perfect field for EEs, since it covers just about everything that is electrical engineering and includes almost nothing that is not. To understand radar systems one needs to understand electromagnetic wave propagation, electronics, antennas, amplifiers, signal processing (a lot of signal processing) and computing hardware. As in almost all technical fields the computing piece is becoming more and more important, since the advances in speed and reduced size and power of the electronics, combined with advances in computational intelligence, are making possible applications that no one would have thought possible 10 or 20 years ago. Radar is also a pretty weird field to study from the academic side, since most of the applications to date have been in the military and defense world, and not being in that environment all the time one is never sure if the theoretical work is fully relevant to real applications. That hasn’t stopped me from moving ahead (in fits and starts, admittedly) and it hasn’t stopped some of my colleagues outside of academia from showing interest in my work over the years.

Conferences like this are a nice mix of the technical and the personal, and often drive home the point that it is the people that get the work done. On the personal side there were two very nice moments for me. One was seeing Dr. Marco La Manna, my first PhD student at Michigan Tech, present our joint paper on hybrid-MIMO radar signal processing. Marco is just starting out his career as a post-doc at the University of Wisconsin-Madison. He did a fine job with the presentation and fielded questions well; I was happy and proud of him. On the other end of the career arc, one my first PhD students back at Washington University, Dr. Frank Robey, was recognized for being elevated to IEEE Fellow status. Frank is my first PhD student to reach this milestone, so again I was happy and proud. I made IEEE Fellow myself back in 2010, and ironically our joint work that was part of Frank’s PhD dissertation played a big role in that. Our paper “An Adaptive Matched Filter Detector” published in the IEEE Transactions on Aerospace and Electronic Systems in 1990 with two other authors, is my most-cited paper, and Frank is the first author. Frank has gone on to a very distinguished career as a radar engineer for MIT Lincoln Laboratory, and he knows a lot more about how radar really works than I ever will. I was Frank’s advisor for four years, but he has been my advisor ever since. It is great to see him get the recognition he deserves.

Coming to a conference like this is like putting on an old shoe. There is a familiar cast of characters: the curmudgeon who stands up and says that your work was done 50 years ago, the young engineer who is nervous and shy, the older seasoned engineer with too many slides, and the guy who gets unnecessarily positioned about the superiority of one technology over another. There are an awful lot of talks that appear to be a new mix of a lot of old concepts and buzzwords, and one is never quite sure if the speaker is really moving the state of the art forward or just reinventing the wheel. What is clear is that technological progress never moves in a straight line. There is a lot of going around and around in circles as the level of understanding in a technical community reaches critical mass to actually make something new happen.

I couldn’t help but notice one change in the conference dynamic which is a result of the ubiquitous smart phone. There is a lot of good conversation in the hallways as there always is, but there were also a lot of people off to the side checking their texts and e-mail. A lot of people were doing the same during the talks, usually at the back of the room when the talk got a little boring. I caught myself doing it too! Staying fully present in this environment is actually quite difficult, for me anyway, and cell phone addiction does not help.

A conference like this can be considered a success if one comes away with at least one new idea or the recognition that the field has changed in some significant way. The most striking thing for me was a presentation on the Google Soli project, which is putting micro-radars into small personal devices like smart watches, to track finger and hand gestures as part of a user interface. Just do a search on “Google Soli” and you can see all about it; it is very cool. This project demonstrated for me the potential that exists in the commercial world for moving technology forward. Even though I did not see much in this conference about automotive radar, it did make me think that there could be a lot of advances coming to support autonomous vehicles also. Given our level of interest in robotics, control, and automation in the Michigan Tech ECE Department it would probably be worth my while to find out as much as I can. The presentation itself on the Google Soli project totally raised the bar in terms of speaker polish and audio-visual aids. It put the rest of us PowerPoint hackers on notice that we need to raise our game if we are going to stay competitive.

Back in the office next week, and with luck it will be spring in the Keweenaw. Hope springs eternal, as it does at the beginning of every summer, that some of the inspirations from this week will turn into concrete results before the start of a new school year.

– Dan

Daniel R. Fuhrmann
Dave House Professor and Chair
Department of Electrical and Computer Engineering
Michigan Technological University