[Date Index] [Thread Index] [Author Index]

Re: Simple Fourier help

• To: mathgroup at smc.vnet.net
• Subject: [mg53258] Re: [mg53237] Simple Fourier help
• From: yehuda ben-shimol <benshimo at bgu.ac.il>
• Date: Mon, 3 Jan 2005 04:29:34 -0500 (EST)
• References: <200501020912.EAA27755@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

This should be s rather simple task, but I need to know the source. Is 
it a composition of pure sine harmonic functions well separated in 
frequency, or a more complicated signal? And what is the sampling 
frequency? Are you sure that you are keeping the Nyquist condition for 
sampling (at least twice the maximal frequency) in order to avoid the 
aliasing effect?
If all these are met, you need to take the 
Abs[Fourier[signal-Mean[signal]]].
I order to see it nicely you need to RotateLeft [] (or RotateRight it) 
according to the number of samples you have.
Then you need to find the first five local maxima (that are, point that 
are larger then both their left and right neighbors), assuming that 
there are such five harmonic components.
If you send me your conditions I'll send you an exact code that 
generates it
yehuda



james wrote:

>Somebody knows how is possible to find frequency and amplitude for the
>most important waves of a primary wave with no complicated
>calculation? I need to find at max the first five subwaves. In anyway
>there is in the web a software or a site that do that? Thanks
>
>  
>

• References:
  • Simple Fourier help
    • From: attackack@yahoo.com (james)