[81453] in North American Network Operators' Group
Re: Calculating Jitter
daemon@ATHENA.MIT.EDU (Fred Baker)
Fri Jun 10 12:58:10 2005
In-Reply-To: <42A96305.6050008@nessoft.com>
Cc: nanog@nanog.org
From: Fred Baker <fred@cisco.com>
Date: Fri, 10 Jun 2005 09:56:38 -0700
To: Jeff Murri <jeff@nessoft.com>
Errors-To: owner-nanog@merit.edu
you saw marshall's comment. If you're interested in a moving average,
he's pretty close.
If I understood your question, though, you simply wanted to quantify
the jitter in a set of samples. I should think there are two obvious
definitions there.
A statistician would look, I should think, at the variance of the set.
Reaching for my CRC book of standard math formulae and tables, it
defines the variance as the square of the standard deviation of the
set, which is to say
sum of ((x(i) - xmean)^2)
------------------------
n - 1
where the n values x(i) are the members of the set, xmean is the mean
of those values, and n is the number of x(i).
A sample set with a larger standard deviation or variance than another
set has contains more jitter.
In this context, the other thought that comes to mind is the variation
from nominal. If the speed-of-light delay between here and there is M,
the jitter might be defined as the root-mean-square difference from M,
which would be something like
sum of ((x(i) - xmin)^2)
-----------------------
n - 1
with the same variables except that xmin is the least value in the set.
