2
$\begingroup$

I've been dabbling with Signal processing for a little over a year and a half. Despite having a BS and a MS in Computer Science, I have a shit math background.

Right now I'm going through Richard Lyons's book. And I've tried a Coursera Course before.

A lot of times I feel like I'm hearing or reading things that don't make sense to me and so I keep reading/playing them again and again, interspersing the repetitions with internet searches and sometime I also end up posting specific questions on Stackexchange (and I'm really thankful to you amazing people for taking the time to answer my questions! I'd probably have given up studying this without you guys)

But I was wondering if any of you had any pro-tips around this process? Because my current method is very time consuming and I was wondering if there was something I could do to better my position.

$\endgroup$
2
$\begingroup$

Well, I believe the ease of learning comes once you experience the need of knowing something. When maths is the only obstacle between you and a goal that you want to achieve, it becomes natural to just tackle down the problem to go past through it.

If schools or books are not stimulating enough for you, I'd say, start a project; that's how I learned pretty much everything. Most books are written to be a reference, not to teach, even if they pretends so (that's my opinion).

Build something. And if you can afford it, don't hesitate to spend money on that. You'll probably spend less than a semester of school (depends on where you live) and you'll get an excellent practical background

Few application I could think of

  • Build a speaker tower, you'll have to design the crossovers and you can compare your results with existent systems.
  • Make a synthesizer
  • Program a guitar tuner
  • If you can put your hand on a SDR (some are affordable), buy an antenna, some RFID tags and try to read them.
  • Try to decode some air signal. Standard radio or more unusual signal like airplane ADS-B
  • Try GNU Radio (free software) and try to make a transmission system.

Well, you get the idea.

$\endgroup$
  • $\begingroup$ have you ever designed a speaker tower ? $\endgroup$ – Fat32 Jun 11 '17 at 23:39
  • $\begingroup$ No, I don't pretend that I did all of the above. I know few peope who actually did their own tower though and they shared their experience with me. I believe it is a great project for learning $\endgroup$ – Pier-Yves Lessard Jun 12 '17 at 0:20
  • $\begingroup$ Yes it's a great thing! For many years I wanted to design one such. Attempts at acoustical theory miserably failed however, the remnants of which is the question: what is the interaction between the volume of the tower and the bass response of the box? Rather than a formula, a physical explanation is preferred... $\endgroup$ – Fat32 Jun 12 '17 at 1:32
  • $\begingroup$ I believe this kind of question would be more appropriate for a real SE question. I would let someone with a stronger background in acoustic waves answer this. $\endgroup$ – Pier-Yves Lessard Jun 12 '17 at 1:40
  • $\begingroup$ Eventhough there r certainly users who would know the answer, dsp.se wouldn't be the proper place I assume, audio.se or acoustics.se would be better ;-) . Anyway thanks. $\endgroup$ – Fat32 Jun 12 '17 at 1:50
0
$\begingroup$

Work, work, and work. And fun, fun and fun. Fun can lever a lot of difficulties. The need for some math skills is a common repellant in signal/image processing. Yet, it is fun, because you can see results easily.

Find a DSP topic you are interested in enough to spend time on it. And dig.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .