General Question - Cultivating/Keeping afloat DSP skills

I'm a literally _new_ to digital signal processing.  Certainly have
enjoyed the college courses and as a result I'd like to seek additional
educational and training opportunities to cultivate and enhance skills.


I've been asked to identify some specific opportunities - seminars,
classes/distance learning courses,  etc.

For you seasoned (perhaps non-seasoned ) veterans.  What specific
opportunities would would you address that'll allow me to keep my
signal processing skills afloat.

I suspect I could subscribe to:
    XYZ (what are these magazines) magazine:
    Take XYZ course perhaps years  - recommendations?
    The obvious:  Read books.

Thanks in advance

forums_mp@hotmail.com wrote:
> I'm a literally _new_ to digital signal processing.  Certainly have
> enjoyed the college courses and as a result I'd like to seek additional
> educational and training opportunities to cultivate and enhance skills.
> 
> 
> I've been asked to identify some specific opportunities - seminars,
> classes/distance learning courses,  etc.
> 
> For you seasoned (perhaps non-seasoned ) veterans.  What specific
> opportunities would would you address that'll allow me to keep my
> signal processing skills afloat.
> 
> I suspect I could subscribe to:
>     XYZ (what are these magazines) magazine:
>     Take XYZ course perhaps years  - recommendations?
>     The obvious:  Read books.
> 
> Thanks in advance

Much of what you want to know is on http://www.dspguru.com/. Check it 
out and get back to us. I'm going to bed now.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

Hi,

I would also suggest besides those mentioned above:

a) working on small projects in your free time (ie, simple echo
cancellation; image restoration;source separation etc.But under very
basic assumptions so the project won't be too much overhead).

b) writing code of some of the dsp algorithms (ie, correlation,
convolution, adaptive filtersm, HMM). This will help you understand the
theory better as well as improve your coding skills...

hth

ikaro

If I were starting out again I would get into DSP implementations with
FPGAs. Seems to be more US jobs in that.

Dirk


forums_mp@hotmail.com wrote:
> I'm a literally _new_ to digital signal processing.  Certainly have
> enjoyed the college courses and as a result I'd like to seek additional
> educational and training opportunities to cultivate and enhance skills.
>
>
> I've been asked to identify some specific opportunities - seminars,
> classes/distance learning courses,  etc.
>
> For you seasoned (perhaps non-seasoned ) veterans.  What specific
> opportunities would would you address that'll allow me to keep my
> signal processing skills afloat.
>
> I suspect I could subscribe to:
>     XYZ (what are these magazines) magazine:
>     Take XYZ course perhaps years  - recommendations?
>     The obvious:  Read books.
> 
> Thanks in advance

Ikaro wrote:
> I would also suggest besides those mentioned above:
> 
> a) working on small projects in your free time (ie, simple echo
> cancellation; image restoration;source separation etc.But under very
> basic assumptions so the project won't be too much overhead).
> 
> b) writing code of some of the dsp algorithms (ie, correlation,
> convolution, adaptive filtersm, HMM). This will help you understand the
> theory better as well as improve your coding skills...

I've found that one of the best ways to learn something is by trying to 
teach it.  You can accomplish the same thing by contributing to an 
online DSP resource such as this one:

http://en.wikibooks.org/wiki/Digital_Signal_Processing

-- 
Jim Thomas            Principal Applications Engineer  Bittware, Inc
jthomas@bittware.com  http://www.bittware.com    (603) 226-0404 x536
The real fun of living wisely is that you get to be smug about it. - Hobbes

dbell wrote:
> If I were starting out again I would get into DSP implementations with
> FPGAs. Seems to be more US jobs in that.
>
> Dirk
Excellent point.  From what I'm seeing.  That seems to be the trend
lately.  I thought about this, expect now I have to learn all that
verliog mess.

Ikaro wrote:
> Hi,
>
> I would also suggest besides those mentioned above:
>
> a) working on small projects in your free time (ie, simple echo
> cancellation; image restoration;source separation etc.But under very
> basic assumptions so the project won't be too much overhead).
>
> b) writing code of some of the dsp algorithms (ie, correlation,
> convolution, adaptive filtersm, HMM). This will help you understand the
> theory better as well as improve your coding skills...
>
> hth
>
Thanks.   I've got the sofware skill set.   I'm fairly competent with
the C++ standard ( aka STL ), boost and the language itself.   I could
work my way around Java but C++ / Matlab is my strengths.

I'll peruse the web to see if I find projects for doing what you
alluded to.  cancellation, restoration etc.

> > Thanks in advance
>
> Much of what you want to know is on http://www.dspguru.com/. Check it
> out and get back to us. I'm going to bed now.
>
> Jerry

Thanks Jerry.

Not asking you to predict the future,  but with respect to signal
processing do you see any (not sure how to word this)  _new_ /earth
shattering ideas.  I've been researching signal processing for the past
few days and I get the feeling that there's good chance that anything I
need to do has already - literally -  been done  ( a beautiful thing of
course ) already.

> I've found that one of the best ways to learn something is by trying to
> teach it.  You can accomplish the same thing by contributing to an
> online DSP resource such as this one:
>
> http://en.wikibooks.org/wiki/Digital_Signal_Processing
I see.  Thanks

forums_mp@hotmail.com wrote:
>>>Thanks in advance
>>
>>Much of what you want to know is on http://www.dspguru.com/. Check it
>>out and get back to us. I'm going to bed now.
>>
>>Jerry
> 
> 
> Thanks Jerry.
> 
> Not asking you to predict the future,  but with respect to signal
> processing do you see any (not sure how to word this)  _new_ /earth
> shattering ideas.  I've been researching signal processing for the past
> few days and I get the feeling that there's good chance that anything I
> need to do has already - literally -  been done  ( a beautiful thing of
> course ) already.

DSP is a hobby that I took up to amuse myself in retirement and to learn 
the basis behind some of the things I did by the seat of my pants 
("inspired desperation" is an other term) while working. For as long as 
I remember, almost everything I could think of had been done already.* 
That's true for almost everybody here. Still, I have several patents and 
many more unpatented inventions. I used magnetic cores to build a 
character generator before ROMs were available. A lot more came later. 
There'll always be new things and new technologies to implement them in.

Whatever you do, be a generalist. One of my patents is for applying the 
idea behind a field flattener for a Schmidt sky camera (published about 
the time I was born in Scientific American's "The Amateur Telescope 
Maker" column) to underwater ultrasound imaging. Many new things are 
made from old parts, but the maker has to know those parts there.

Jerry
__________________________________________
* That clearly says more about what we can think of than about what's 
been done. With my knowledge and Napoleon's resources, one could easily 
have built ship-to-shore transatlantic radio. I even know where to mine 
and how to work tungsten and thorium for tube filaments. Did you know 
that the ancient Babylonians had batteries of galvanic cells?
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������