Thread Subject: Narrow band Noise generation

I am trying to generate some narrow band noise. I have so far used
the rand object to generate white noise but am now unsure as to how
to go on from there. I eventually want to have a sample of sound to
use.

Thanks Matt

You might consider using the WGN or AWGN functions to generate your white
noise signals. For more info, type:

    doc wgn

    OR

    doc awgn

at the command prompt.

Once you have your white noise signal, try passing it through a low-pass
filter to create narrowband noise.




"Matt Joseph" <still_feel_alive@yahoo.co.uk> wrote in message
news:ef5ca3f.-1@webcrossing.raydaftYaTP...
>I am trying to generate some narrow band noise. I have so far used
> the rand object to generate white noise but am now unsure as to how
> to go on from there. I eventually want to have a sample of sound to
> use.
>
> Thanks Matt

I don't seem to get any info when typing
doc wgn or doc awgn

Do you have to have a specific version of Matlab for this?

Matt
 
Rick Rosson wrote:
>
>
> You might consider using the WGN or AWGN functions to generate your
> white
> noise signals. For more info, type:
>
> doc wgn
>
> OR
>
> doc awgn
>
> at the command prompt.
>
> Once you have your white noise signal, try passing it through a
> low-pass
> filter to create narrowband noise.
>
>
>
>
> "Matt Joseph" <still_feel_alive@yahoo.co.uk> wrote in message
> news:ef5ca3f.-1@webcrossing.raydaftYaTP...
>>I am trying to generate some narrow band noise. I have so far
used
>> the rand object to generate white noise but am now unsure as to
> how
>> to go on from there. I eventually want to have a sample of
sound
> to
>> use.
>>
>> Thanks Matt
>
>
>

Matt Joseph:
<SNIP missing functions...

> I don't seem to get any info when typing doc wgn or doc awgn...

they are part of the <communications tbx>...

us

I do not have this toolbox installed and as it is not my Matlab I
cannot install this...any one got any other ideas about how to tackle
this problem?

Matt

Us wrote:
 
> they are part of the <communications tbx>...
>
> us

The function RANDN will generate a sequence of normally-distributed random
numbers, with mean 0 and variance 1. Try the following code:

  % Parameters:

  N = 2048; % Number of data points
  mu = 0; % mean
  sigma = 3; % standard deviation


  % White gaussian noise sequence:

  x = mu + sigma*randn(1,N);


  % Autocorrelation:

  R = xcorr(x);
  plot(R);



This code will generate a WGN sequence with 2048 elements with mean 0 and
standard deviation 3, and then compute and plot the autocrrelation function.


Rick





"Matt Joseph" <still_feel_alive@yahoo.co.uk> wrote in message
news:ef5ca3f.3@webcrossing.raydaftYaTP...
>I do not have this toolbox installed and as it is not my Matlab I
> cannot install this...any one got any other ideas about how to tackle
> this problem?
>
> Matt
>
> Us wrote:
>
>> they are part of the <communications tbx>...
>>
>> us

Rick Rosson wrote:
> You might consider using the WGN or AWGN functions to generate your white
> noise signals. For more info, type:
>
> doc wgn
>
> OR
>
> doc awgn
>
> at the command prompt.
>
> Once you have your white noise signal, try passing it through a low-pass
> filter to create narrowband noise.

What's wrong with simply rand() passed through a band-pass filter?

That works too. The function RAND, however, generates a uniform
distribution, whereas RANDN generates a normal distrubution.


>
> <Steve.Amphlett@ricardo.com> wrote:
>
> What's wrong with simply rand() passed through a band-pass filter?
>
>>
>> Rick Rosson wrote:
>>
>> You might consider using the WGN or AWGN functions to generate your white
>> noise signals. For more info, type:
>>
>> doc wgn
>>
>> OR
>>
>> doc awgn
>>
>> at the command prompt.
>>
>> Once you have your white noise signal, try passing it through a low-pass
>> filter to create narrowband noise.

I use randn() + IIR fileter for narrow band generation.

However, I think it is not a so-good way, because:

you must re-scale the output to determine the nb noise power.

Anyone has some other good method?

Justin Yang

Steve.Amphlett@ricardo.com wrote in message <1183735615.504786.173400@57g2000hsv.googlegroups.com>...
> Rick Rosson wrote:
> > You might consider using the WGN or AWGN functions to generate your white
> > noise signals. For more info, type:
> >
> > doc wgn
> >
> > OR
> >
> > doc awgn
> >
> > at the command prompt.
> >
> > Once you have your white noise signal, try passing it through a low-pass
> > filter to create narrowband noise.
>
> What's wrong with simply rand() passed through a band-pass filter?
>

On Dec 1, 7:14 pm, "yuqing yang" <13564453...@163.com> wrote:
> I use randn() + IIR fileter for narrow band generation.
>
> However, I think it is not a so-good way, because:
>
> you must re-scale the output to determine the nb noise power.
>
> Anyone has some other good method?
>
> Justin Yang
>
>
>
>
>
>
>
> Steve.Amphl...@ricardo.com wrote in message <1183735615.504786.173...@57g2000hsv.googlegroups.com>...
> > Rick Rosson wrote:
> > > You might consider using the WGN or AWGN functions to generate your white
> > > noise signals.  For more info, type:
>
> > >     doc wgn
>
> > >     OR
>
> > >     doc awgn
>
> > > at the command prompt.
>
> > > Once you have your white noise signal, try passing it through a low-pass
> > > filter to create narrowband noise.
>
> > What's wrong with simply rand() passed through a band-pass filter?

Don't top post. It makes the thread hard to follow.
Put your reply UNDERNEATH.

You can use phase randomisation - search for that term in this
newsgroup.