Hi,
see http://wiki.wireshark.org/RTP_statistics -> "How jitter is calculated" (bottom of page).
Basically, jitter is a smoothed derivative of the interarrival delta.
So it will not get nearly as high as the deltas itself, unless fluctuations of deltas are very frequent and of high amplitude over a longer period of time.
Lars
Jaakko Hakalahti wrote:
Hello,
I have been troubled with understanding the connection between the Max
Delta (ms) and the Max Jitter (ms) column in the RTP stream analysis.
If I use a G.711 codec (PCM) for the VoIP-call, there should be one
packet sent every 20 milliseconds and therefore the Max Delta (ms)
value should be pretty close to that. (In an ideal conditions). In this
case the Jitter value also should be close to nothing.
I made some tests with a lot of heavy traffic over the same network
where I made the VoIP-call, and I got some wild Max Delta values up to
360 milliseconds. I expected that the Jitter value would also follow
the Max Delta value and go really high but that is not the case. The
highest Jitter that Wireshark shows for me in that same test is 42
milliseconds. I made several other tests as well and for example if I
got the Max Delta value up to 160 ms, the Max Jitter was 16 ms.
Jitter is supposed to be the variation of the time between packets
arriving to the receiver, right? Then why, if there is such a huge cap
between packets (as that 360 ms is), the Jitter value is only 42 ms?
How is it really calculated then?
I searched the web for this and I found out that Interarrival Jitter is
supposed to be: "The interarrival jitter J is defined to be the
mean deviation (smoothed absolute value) of the difference D in
packet spacing at the receiver compared to the sender for a pair
of packets." @ RFC 3550.
Though that Jitter means the RTCP-packet field "Interarrival Jitter",
not the Jitter measured from the RTP-packets.
Help me understand the connection between the Max Delay and the Max
Jitter, because I don't really get it.
Thanks.
_______________________________________________
Wireshark-users mailing list
Wireshark-users@xxxxxxxxxxxxx
http://www.wireshark.org/mailman/listinfo/wireshark-users