🚀 go-pugleaf

On 10 Aug 1995, Walter Gray wrote:

> In article <40b367$bjo@netnews.upenn.edu>, george@mech.seas.upenn.edu ( George Jefferson ) writes:
> >
> >
> >:: >	______         _____	   ___
> >:: >	|    |	       |   |        ^
> >:: >	|    |	       |   |        C	
> >:: >________|    |_________|   |_____  ___
> >:: >	
> >:: >	|<A >|<-- B--->|<A>|     
> >:: >
> >:
> >:: mean = C*A/(A+B)
> >:: rms = C*sqr(A*B)/(A+B)
> >:
> >:Sorry to say, but according to your formula, when B approaches 0, rms
> >:also approaches to 0. Well - rms of fixed voltage is not 0.
> >
> >
> >rms = sqrt[ 1/t int[v^2 dt] ] = sqrt[ 1/(a+b) * ac^2 ] = c sqrt[ a/a+b ]
> >
> >rms = c sqrt[ a/a+b ] = c / sqrt[ 1+ b/a ]
> >
> >limit b->0, rms -> c
> >limit a->0, rms -> 0
> >
> >
> 
> limit b->0, mean ->c, therefore rms ->0, no problem
> 
> Obviously I subtract the mean before squaring, don't you?

So the rms value of a DC signal is zero? Sorry, but I'm too ignorant
to fall for that one ;) If you can provide a reference which defines
rms this way, would you post it please? All the texts and references
I have define rms as the "root of the mean of the square of the
function". (But these texts all deal with the EE application of rms,
and not any mathematical subtleties of the function definition - if
there are any).

> But seriously though, why would anybody want the rms of a *digital*
> signal anyway? What use is it? I think we have just done the guy's
> homework for him. I hope he is suitably confused.

I can only think of one application immediately: rms of a digital signal
will provide information on power dissipation/consumption. The peak
values obviously give max values, but a high peak value with a real low
rep rate (PRR) doesn't give you the whole picture for devices like
transmitters. It is often much more important to set power limits based
on rms values as well as the peak value. 

> 
> walter
> ------
> 
> 
> 

--
............................................................
Steve White, Lab Technician @ Human Interface Technology Lab
e-mail: stevew@hitl.washington.edu       voice: 206.616.1411
http://www.hitl.washington.edu/  (My page under consruction)
............................................................