Reply by sauravrt June 20, 20092009-06-20
Tim:

Since I dont' have a calorimeter at my disposal at this point of time, I
just decided to digg thru Matlab help on wgn and this is what they have to
say:

 y = wgn(m,n,p) generates an m-by-n matrix of white Gaussian noise. p
specifies the power of y in decibels relative to a watt. The default load
impedance is 1 ohm.

Does that make some sense?





>On Tue, 16 Jun 2009 19:51:56 -0500, sauravrt wrote: > >> Matlab command y1 = wgn(100,1,10) generates a column vector of
length
>> 100 containing real white Gaussian noise of power 10 dBW. How can I >> verify that the power of the 100 samples is 10 dBW ? > >dBW? You mean "Decibels -- Watts"? > >If so, put the vector of numbers in a calorimeter for one second, and see
>if it registers one Joule of heat energy. > >-- >www.wescottdesign.com >
Reply by Tim Wescott June 18, 20092009-06-18
On Tue, 16 Jun 2009 19:51:56 -0500, sauravrt wrote:

> Matlab command y1 = wgn(100,1,10) generates a column vector of length > 100 containing real white Gaussian noise of power 10 dBW. How can I > verify that the power of the 100 samples is 10 dBW ?
dBW? You mean "Decibels -- Watts"? If so, put the vector of numbers in a calorimeter for one second, and see if it registers one Joule of heat energy. -- www.wescottdesign.com
Reply by sauravrt June 18, 20092009-06-18
@Julius: Thanks for the reminder. I know that

P_y = (1/N)*sum(y.*y)
That would give me the time average power of the bunch of numbers (from
wgn).

I'm want to get an analytic expression for P_y  for N samples of WGN.
Suggest me with that.
Thanks


>On Jun 16, 8:51=A0pm, "sauravrt" <sauravtulad...@hotmail.com> wrote: >> Matlab command =A0y1 =3D wgn(100,1,10) =A0generates a column vector of
le=
>ngth 100 >> containing real white Gaussian noise of power 10 dBW. =A0 >> How can I verify that the power of the 100 samples is 10 dBW ? > >You generated a bunch of numbers. I suppose that you can pretend that >they represent values of amplitude. Now you are asking about dBW, >which is a unit of power. You should review what is the relationship >between amplitude such as voltage, and power. Further, you are >working in discrete time. You should also review the representation >of signals in discrete time. > >Julius >
Reply by cpshah99 June 17, 20092009-06-17
>Matlab command y1 = wgn(100,1,10) generates a column vector of length
100
>containing real white Gaussian noise of power 10 dBW. >How can I verify that the power of the 100 samples is 10 dBW ? >
Try 10*log10(var(y1)) and see what u get. Chintan
Reply by julius June 17, 20092009-06-17
On Jun 16, 8:51&#4294967295;pm, "sauravrt" <sauravtulad...@hotmail.com> wrote:
> Matlab command &#4294967295;y1 = wgn(100,1,10) &#4294967295;generates a column vector of length 100 > containing real white Gaussian noise of power 10 dBW. &#4294967295; > How can I verify that the power of the 100 samples is 10 dBW ?
You generated a bunch of numbers. I suppose that you can pretend that they represent values of amplitude. Now you are asking about dBW, which is a unit of power. You should review what is the relationship between amplitude such as voltage, and power. Further, you are working in discrete time. You should also review the representation of signals in discrete time. Julius
Reply by sauravrt June 16, 20092009-06-16
Matlab command  y1 = wgn(100,1,10)  generates a column vector of length 100
containing real white Gaussian noise of power 10 dBW.  
How can I verify that the power of the 100 samples is 10 dBW ?