Hi everyone,

Maybe this is not an ordinary question but I would like to ask all of you, who has many years of experience as professionals engineers, and who knows market reality, if going for PhD in E.E. field like: radio communication, DSP etc. is really worth its effort. At the beginning I want to say that I don't have any intentions to offend anyone. I am just having a dilemma in my life and I am sitting on the fence. I am already at 2nd year of my PhD studies, without having even topic of my thesis. I've always been person who like more practicing.Theory without examples was difficult to comprehend for me. Nevertheless, I was always more than average in my group so I got the offer to try PhD studies. I still feel like I have lacks of fundamentals in DSP and stuff that I consider as basics. I am involved in few research projects which are very interesting for me but my role there is more being an engineer than scientist. I need to do something with my future and I am about to make a big choice. Before that, I want you to give your opinion about PhD studies and people with PhD in E.E. If owning PhD can really change something while you are not working at university. 

Sorry for that silly question.

Regards

Marcin       

That's not a silly question, Marcin.

First, a joke: When you think that you know everything, they give you a bachelors degree.  When you realize that you know nothing, they give you a masters degree.  ...and when you realize that nobody else knows anything, either, they give you a PhD.

I mention that because of your comment about lacking fundamentals.  In higher education, sometimes the hoopjumping that comes with the educational process causes us to move very quickly from the fundamentals to the advanced materials. You know more than you realize.  ...or you will be able to figure it all out quickly.  Sometimes, a person just needs a chance to breathe to have it fall into place.

Throughout my undergrad, I always had one foot out the door.  A cooperative education program with a steel mill, a summer job with Westinghouse Automation,... I even formally withdrew from my undergrad university once to work on a cool controls project.  At that time, I didn't see myself even doing an MS.  A goofy boss and some of the encouragement that you encountered caused me to go back.  

I got a PhD (Mechanical Engineering, but it doesn't really matter).  I spent my work career in active noise and vibration control - mostly in the automotive/transportation area. Now I am a EE professor.  I thought that I would answer because I've done both industry and - now - academics.

The PhD was useful in industry.  There is certainly the scientific aspect of it.  I spent much of my time translating/adapting/extending published algorithms into DSP-based products.  That would have been possible without a PhD. ...but difficult.  

If you stay for your PhD, you will also learn a lot about managing an amorphous project, managing yourself, and managing yourself doing an amorphous project.  That will help for industrial projects, as well.

There was some glamour value in the PhD.  ...and some level of professional insurance.  I could make the most stupid statement imaginable, but they still knew that I was at least kinda smart at something.  :)

Obviously, the PhD is valuable when in an academic setting.  Maybe - like me - you will end up going back to academe.  You may want to consider that in your decision.  

Your hardest part - IMHO - will be finding something that excites you sufficiently to do it.  You could grab some random topic and expand it sufficiently, sure.  ...but a dissertation and defense require no small effort and pushing yourself through that without a true interest in your research topic would be painful at best.

If I were you, I would focus on that topic search.  If you can find a topic that excites you, then you are in a place to make a good decision.  

A second option is to talk to your university about withdrawing for a year or two while you work.  Then, after that, you can decide whether continuing with work or returning to the university is the best for you.  If you do that, though, do not overspend.  Save the money for your decision time.  Too often people get themselves into a financial state where they can't stop working even if they want to.  I'm not saying that to be condescending.  It is just all too common.

Hope this helps.  You'll be fine whatever you choose.

All the above, plus, writing a thesis teaches you -- and shows that you can -- write a book-length work.

I've worked with PhD's who wouldn't get their hands dirty, or were disasters when they did -- basically the cliche of the useless academic.

I've also worked with PhD's who had a good grasp of how to do immediately practical work -- and they could do things that no one else could do.  So don't think that having a PhD prevents you from being practical (and, if it's not too late, consider finding a subject that lets you be practical).

I even worked with one guy who deliberately went to work as a plain old software engineer for a couple of years after his PhD -- when I asked him why he downplayed his degree, he told me he was doing a postdoc in practical application.  He's gone on to being a foundational member of somewhere around a dozen artificial intelligence startups, and is the only guy I knew in the 1990's who not only made money at neural nets, but who's customers made money on the neural nets that he designed.

Thank you very much for your answer. It really gave me better view of my current situation. Its really helpful to get good advises and guidance in your life by simple suggestions from someone who got trough all of the similar stuff.

Marcin

I began working after getting my Masters degree in 1961, with the idea that I would work for a few years and then get a PhD. I knew I wanted to do research. But during  my first few years of working, I was already doing good research and it seemed like a PhD was really unnecessary. 

Unfortunately, about fifteen years later I found that I was working on research projects that were too challenging. I wish that I had somehow gotten those two or three years more of education. For me, the degree would have been useless, but the education that that earned it would have been really valuable. 

I got a PhD in nuclear engineering.  I've spent the last 40 years doing E.E. work, all pretty much self taught.  I've also been told "you can't have a PhD, you actually know how to do stuff!"  Having the degree isn't very important for what you want to do, but in 20 or 30 years it will have a lot more usefulness.

Jumping through the hoops to get the degree is definitely disillusioning. I have a copy of my degree on a dart board and it's use is to show I have a tendency to throw diagonal - high right to low left.  The paper really doesn't mean much to me.  It means a lot to others.  

If you can stomach the hoops, then jumping through them will pay off in the long run.  As pointed out by others here, it is not essential because you can find ways to do what you like.  The most important thing to get out of the process is learning how to learn.  Finding new ways to do things never ends and it never gets boring.  You don't need a piece of paper to do that.  But it does open a few doors that would not otherwise appear.  

Mike

In the audio industry, a sorta consumer-oriented industry that is pretty close to the market (like nearly all of the time I was working on development of a product that was to be released in less than two years, very little "pure research"), I found both the management and fellow engineers to be utterly unimpressed with the PhD as a credential.  They, sometimes correctly sometimes incorrectly, thought that the PhD engineer would not be able to get down into the dirty details of product development (like writing code, designing interfaces including the so-called user interface).  There was one engineering manager that told me at an AES convention that he would never hire a PhD again.

Now there were some bigger companies that eventually became part of one of the big four: Bose, Avid, Harmon, Dolby, that did have a deliberate research effort that they could pay for.  And there were some PhDs with them, many of these PhDs are people that I know personally and respect.  But even if the product had a highly mathematical component, like audio DSP, if the company was small, like a dozen employees (and that is far more common than huge companies with a core research department), the PhD was not really valued.  They wanted engineers that could understand the problem and create solutions fast enough to get the product to market.

Of course, the other career direction is academic.  With a glut of PhDs out there, you will simply not get an academic job at a university teaching electrical engineering without a PhD.  Even if you're an amazing teacher and accomplished researcher, the product that the academe sells are academic degrees.  Even at non-prestigious colleges or universities.  They *must* maintain the perception that the degree is a valuable commodity and they don't do that if they don't value the degrees themselves and to do that, they must also devalue the lack of degrees.  Now that doesn't mean that you can't teach a single class now and then in some kinda adjunct capacity, but you'll get paid maybe $5000 to teach a class over an academic term (quarter or semester).  You'll find that you are sorta being taken advantage of.  Also you could teach at the community college level, but it will be nothing more sophisticated than sophomore-level physics or calculus.

Like John, I was also a PhD student that failed to get the degree.  This is 40 years ago.  I was ABD and I did teach at the University of Southern Maine (a non-prestigious startup EE department) briefly in 1989-91.  My contract was not renewed, and when I struck out to work in the only EE industries that interested me; audio and music instrument, I had the worst of both worlds, I was considered too Ivory Tower to be any good (as in the sentiment illustrated above) and I had no completed PhD so I was considered lacking for any research or academic position.  I just had to show people that my skill set was the best of both worlds.  I've had a spotty career and, at 64, I am semi-retired.  But even today, I am discussing possible collaboration with the president and owner of a company that makes sophisticated guitar effects processing.  I ended up giving people a lotta valuable advise for free that did something for my reputation, I try to control how much I give them (I want them to understand that I know a thing or two, but I also don't want to give away the store).  Such is the life in a competitive environment, not a collaborative environment.  It's the way it is.

Because the industry is constantly changing, your experience will be different than what people have experienced in the past.  My career included a lot of working in research labs with a lot of PhDs (I don't have one, though), and a lot of development work as well at smaller companies.  I don't think a PhD will hurt you, and I don't think it will keep you from doing hands-on work if that's what you're interested in.  The likelihood of a PhD opening doors for you is higher, I think, that it is of closing doors, which it may a bit if people think you are over-qualified for a job you might be interested in.

It's pretty rare that a person can cross effectively between a real research environment and implementation, so being able to do that is pretty valuable.  If you're on the road to getting the PhD, my advice would be to finish that, and then go pursue whatever it is that you really want to do.  Overall, the PhD should not be a barrier to that, and may open doors that wouldn't otherwise.

The bottom line is always to manage your own career the way you best see fit.  

Marcin:

I can only tell you my story, and that may help.  I went back for my PhD after working in industry because I was gravitating toward management, and I really wanted to stay on the technical track.  I finished all but the dissertation.  I knew more than my thesis advisor.  He wanted me to study a topic in which I had no interest.  My wife died.  In short, I just ran out of airspeed.  Fast forward 50 years.  I had a great career (but in all fairness I was in the golden age of research).  The transistor had just been invented and worked through the evolution of integrated circuits, etc.  The lack of a PhD did not inhibit my career except maybe holding a position requiring the degree.  I don't regret not having it for a minute.  There is plenty of room for research in the industrial world without having the degree.

I work in research in the automotive industry with several people who have PhDs and I have to say I'm some what envious of the fluency they have within their disciplines. To the point where I've actually considered getting a PhD myself. Although I'd have to say it's hard to separate that from the fact that they have about as much experience in industry than I do in life, lol. But there's one guy who can relate all the theory to exactly what we're doing and it's unbelievably useful in terms of guiding projects as well as awe inspiring. I can't imagine getting to those levels without going through some rigorous academic training. I've attempted to fill in the gaps myself and it's just hard to find the time juggling a full-time job w/ life. 

However, I do have to ask how surprised can you be if you find yourself more acting as an engineer than a scientist when you're enrolled in an electrical engineering program :p

Sorry, maybe I didn't make myself clear enough. By saying "being an engineer than scientist" I thought about my duties which are invloving mainly coding, simulating stuff that basically is well known but we only introduce some small changes and check its behaviour (genarally speaking). Maybe instead of saying"scientist" I should have said "researcher". 

P.S. Sorry for any blunder, english isn't my mother tongue. ;)

Marcin-

If you decide yes, my suggestion would be to focus on "EE aspects" of neural networks. By that I mean their architecture, timing, memory costs (energy per bit, both to store and to move bits), etc. Currently there is a stunning disconnect between the human brain handling complex recognition functions under 60 W and AI systems requiring 1000 times more to approximate even one task, such as chess or computer vision.  Guys like Kurzweil and Musk who think the "singularity is near" are over hyping.  It's 100 years at least.

Sadly the glory days of DSP and RF are behind us. You could still make a decent living, but if you want a 6.5 figure salary, it's all AI.

-Jeff

Hi Jeff,

Thank you for your answer. To be honest I thought about ML. During my free time I enjoy to develop myself into that field. I think I have strong background to deep into this seriously. After reading answers to my question above, I am wondering about taking at least one year off from my uni where I am enrolled and find internship or trainee in company related to ML/AI where I can test myself. I believe I need to try industry but in developing filed like AI, as you suggested. 

BTW: Is there any possibility on DSPrelated that people offer internships or jobs or give any recommendations? Or it is not the place.

Regards,

Marcin 

Marcin-

I'm not sure how to explore internship possibilities on the -related.com sites. I suggest to e-mail the owner/operator Stephane and ask, he's always very helpful and thoughtful.

-Jeff

Marcin,

This is a great question! 

If you decide to continue then pick the best adviser you can. Know your adviser and their research well and pick an area that your adviser is well versed. It may not be your favorite just as long as you don't hate it. You may find something a new area that you will come to like. I found the coursework, math courses, at least as valuable as doing the research. I’ve found the exact research topic isn’t that important unless you’re going into academia. Bottom line is get finished and having the right adviser will make or break your program. 

I got mine at the age of young age of 55 after 30 years of experience. I don’t recommend waiting nearly that long. But I worked for a large company that paid for most of it and I was able to work full time. Big companies and government labs tend to value the Ph.D. more than smaller ones but it depends on the industry. I’ve mostly worked on research tasks in a field where the degree is still valued.

Tim

I would like it if someone would check my math on this, but I have the impression that having an advanced degree makes it easier to get hired with gray hair (i.e., over 50).  I'm in my late 50's now, and still employed, but I know that the HR person at one of the jobs that I did not get was getting very ageist with my reference.

Contrast this with folks I know in my father's generation that ended up out of work and on an ever-descending salary scale after 55 -- and it's still happening today.

It doesn't matter to you now, but if you want to have the option to work as an engineer until you're 70, and get paid as if you're worth something, a PhD will help.

(Or save like mad when you're young, and plan on being a clerk at a hardware store, or a temp worker, 30 years from now).