Path: news.mathworks.com!not-for-mail
From: <HIDDEN>
Newsgroups: comp.soft-sys.matlab
Subject: Re: Oversampling,window
Date: Mon, 20 Jul 2009 02:40:02 +0000 (UTC)
Organization: Uofc
Lines: 47
Message-ID: <h40le2$b95$1@fred.mathworks.com>
References: <h40iv9$2au$1@fred.mathworks.com>
Reply-To: <HIDDEN>
NNTP-Posting-Host: webapp-02-blr.mathworks.com
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Trace: fred.mathworks.com 1248057602 11557 172.30.248.37 (20 Jul 2009 02:40:02 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Mon, 20 Jul 2009 02:40:02 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 1662491
Xref: news.mathworks.com comp.soft-sys.matlab:556682


"guj " <gulatiakshay@gmail.com> wrote in message <h40iv9$2au$1@fred.mathworks.com>...
> I really need to clear this :),
> 
> In real world we sample a signal from A/D suppose with Fs=100 Hz..Now when we see that
> digital signal its non uniform sampled. 
> 
> Now, on non uniform signal Please correct me if  am wrong:)
> 
> i want to apply oversampling which means that i need to sample it again with Fs=400Hz....and later regularise it. But how will  i apply oversampling on it.....I mean i dont have amplitude value or i can say that i dont have signal to sample i only have some non uniform discrete values
> 
> How over sampling is apply in the real world. 
> 
> 1. We always do over sampling during A/D converter
> 2. Also how can i apply window function, I didnt get reply for this
> 
> 1.if my orginal signal s is between -30 to 30 Hz 40 by 1
> 2.And signal on whch filter need to apply is 127 by 1 w
> 3. Now how can i recover s by applying rectangular window function on w. I mean dimensions are different..its not working on matlab
> I am doing 
> 
> w.*RECTWIN(SIZE(s))
> 
> Any help will be appreciated 
> Thanks
> 
> Aki


I think, I need to put my doubt in right way :)

dft can be defined in a way

P(m) = sum{n=1:N}  p(n) exp(j * m * delk*x(n))
delk=2*pi/T

m=-M/2-----M/2-1

Now my question is i have N irregular spaced signals....and I am Transforming N samples to M points

oversampling=Nc/M;

now when oversampling is 2

Nc=2M,  so i am applying FFT on Nc samples, But where are my Nc samples i  only have 
N samples....how can i apply FFT on Nc signal when i dont have N values...

Am i able to clear what my doubt is