PDA

View Full Version : Use of FFT to recover parameters of waves


Bacle
Jan3-11, 07:56 PM
Hi, everyone:

As I am sure will be clear from my post, I am not--nor have I ever been-- an EE :).
(For the sake of giving some context to help taylor the answer,
I am a mathematician-in-training. I only know the most basic ideas of signal-processing
but I do know --intro-to-mid-level-- statistics: CLT, hypothesis-testing, etc.)

I am just curious on how one can use FFT to recover some parameters from
given waves. Would someone please explaina bit , or suggest a ref? I have read
just a bit on using the mean and likelihood methods to minimize noise.

I was also hoping someone would suggest some good sources at intro
level dealing with signal processing.

Thanks in Advance.

Phrak
Jan3-11, 09:14 PM
This comes under the topic Numerical Methods in computer science. It might help to know Fourier transforms in advance.

UR_Correct
Jan4-11, 01:03 AM
Don't assume Fourier is a beast that can be easily tamed! There is some intense mathematics involved in signal processing!

Phrak
Jan4-11, 04:19 AM
Don't assume Fourier is a beast that can be easily tamed! There is some intense mathematics involved in signal processing!

Care to share?

otto9K9otto
Jan4-11, 11:05 AM
I found this site to be enormously helpful in all aspects of digital signal processing, including the FFT: http://www.dspguide.com/

rbj
Jan4-11, 02:38 PM
i might suggest to go to the USENET group comp.dsp and as them about questions regarding signal processing. it has a pretty high S/N ratio and, if you don't have a newsserver with your ISP, you can always use Google Groups.

Bacle
Jan5-11, 06:09 PM
Thanks to All for your Replies.
rbj:

I recently read in my handbook a definition of the signal-to-noise ratio
in a sample, as the ratio of the mean to the standard deviation. Is this
correct?