0
$\begingroup$

Different universities around the globe offer DSP courses with different names

Especially some name the course "Digital signal processing" and some name "Real Time Digital signal processing"

I want to know,is there any difference between these two courses or they referring to exact same course?

Forexample,please check the below two links,both are from texas university

http://signal.ece.utexas.edu/~arslan/courses/dsp/index.html

http://signal.ece.utexas.edu/~arslan/courses/realtime/index.html

Both courses on above links have different course codes and websites,by which i am making assumption that both these courses are not exactly same?Is my understanding correct?

$\endgroup$
2
  • 2
    $\begingroup$ We can't tell you what a university puts in their course. That's why they have websites and syllabi, so that you can look into the contents. All universities are different. $\endgroup$
    – mmmm
    Jun 29 at 5:49
  • $\begingroup$ It feels like the websites are very explicit, so you can read and compare them, to see that they're very different. Could you pinpoint from where uncertainty about inequality comes? $\endgroup$ Jun 29 at 7:05
2
$\begingroup$

In this case these are actually very different courses.

The first one goes through the mathematical foundations of Digital Signal Processing: continuous vs discrete, sampling theorem, Z-transform, LTI systems, Fourier Transform, some light-weight filter design, etc.

You'll probably will use mostly pen & paper plus Matlab or Python for this one.

The second one appears focus and getting this to run on an actual piece of hardware. You'll most likely learn how to get the hardware up and running, getting development tools installed (C-compiler, board interface), figure out how to configure the processor and the peripherals so that you can actually receive and send data at the right format and correct speed and timing, build a data path through your processor, express an abstract algorithm in terms of actual working code, track performance & system metrics and test & debug (A LOT).

These are very different skill sets. IMO the first course is a pre-requisite to the second: Before you implement an algorithm in hardware, you need to understand how the algorithm actually works, what the expected behavior is and what you can or cannot tweak.

$\endgroup$
1
  • $\begingroup$ There are also some very important mathematical distinctions between something that is or can be real time and some other thing that cannot. there is an objective definition of real-time DSP. So something like time scaling (time compression or time stretching with or without pitch change) cannot be real time indefinitely. Overlap-add and Overlap-save are also a real-time concern. but you're right about the former. $\endgroup$ Jun 29 at 14:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.