I'm an engineer, do I really need to take this math subject?

Linu

Hey everyone, I'm an engineering/science student at ANU majoring in signal processing or photonics for eng and Physics for science.
I'm in a predicament because I cannot decide whether to do this subject:

http://studyat.anu.edu.au/courses/MATH2320;details.html

I have done two first year subjects which basically introduce you to proof although the assessment in exam situations was more or less technique based (I did well in the subjects.)
I am a little hesitant to do the subject as I can write solid proofs when given time to think about and articulate my thoughts, but in exam situations I have a lot of difficulty. It will be even harder to do well since exams will be completely proof based in this subject.

I am wondering exactly how useful this subject will be in my fields? Is this something I can pick up later on? What exactly will it be used for?

Thanks in advance. Hopefully I've done this right, I just joined up :)

homeomorphic

I studied engineering, but switched to math, so I can't say exactly what it is useful in engineering, but as a math grad student, I can make some comments on the general usefulness of analysis. It should be noted that one of the big inspirations for the development of analysis in the 19th century was the theory of Fourier series, and signal processing involves a large amount of Fourier stuff (physics, too). If you are like me, even if you don't use it explicitly, you will find comfort in understand better why the theory works. Granted, you probably won't see the applications to Fourier series in the class. You may have to read about that on your own. A nice book which covers this stuff (particularly the role of Fourier series in the development of analysis) is Discourse on Fourier Series by Lanczos. Another book that talks about this is A Radical Approach to Real Analysis.

An additional benefit is that you will gain a better mastery of calculus and generally better math skills. It will open the door to any other math you might want to use, like maybe complex analysis. After studying so much math, the easy (and admittedly, more useful math) is now mind-numbingly trivial and easy for me. That's a PhD's worth. But if you take one class, you take one step towards that skill level.

jasonRF

My best advice is to ask the professors in your intended field of study. Chances are they will give you the most appropriate direction for your situation and for the expectations in your part of the world.

Anyway, I work with a lot of folks who received PhD's in signal processing. Most, if not all, took at least one real analysis course. Some took such classes during undergraduate studies while others waited until grad school. When i was in grad school all of the signal processing PhD candidates did a minor in math, so had to take 4 upper division (undergrad level) "honors" classes: 2 in real analysis, one in linear algebra, and one in abstract algebra. So yes, it can be useful, but on the other hand it may not be necessary for you to go too far down this road as an undergrad.

On the other hand, for photonics more standard math methods for physics/engineering courses (covering pdes, complex analysis, etc) are likely more useful.

good luck,

jason

bcrowell

Depends completely on what type of engineering you do. The vast majority of engineers wouldn't need this course.