Hi Ravi, Shouldn't it? As pjmedia rtcp calculates mean of jitter that way. The code of calculating mean seems to be different as the mean needs to be updated each time a packet come. Inputs or comments are greatly appreciated. Regards, nanang 2008/6/16 Ravi Inder Singh <raviinder at gmail.com>: > Hi nanang, > Thanks alot. > I dont have any issues in calculating Jitter values accoding to RFC 3550 > A.8, > now only question arises of "common statistical definition", if we see > according this then it should be like :- > ( sum of all jitter values ) / no of packets . > > Regards, > Ravi Inder Singh > > > On Mon, Jun 16, 2008 at 6:09 PM, Nanang Izzuddin <nanang at pjsip.org> wrote: >> >> Hi Ravi, >> >> Not sure how Wireshark do the calculation, however, the standard >> reference of calculating jitter is RFC 3550 A.8. And mean, max, min,s >> std dev of jitter are calculated as in common statistical definition. >> That's the calculation method used in pjmedia rtcp. >> >> Regards, >> nanang >> >> >> 2008/6/16 Ravi Inder Singh <raviinder at gmail.com>: >> > Hi Nanang, >> > Thanks alot. >> > This seems to be different method from which wireshark is using , >> > wireshark uses (sum of all deltas) / no of packets. >> > >> > In this file aaa.cap attached , in Statitics ---> RTP--> show all >> > streams, >> > it shows means Jitter 18.09. >> > Your method defined above doesn't give same values . >> > >> > And in another file mean_calc.pcap ,from (172.26.1.101 --> 172.26.1.104 >> > ) it >> > calculates mean with different method.(Mean value is 1.28 ). I dont have >> > any >> > idea over that ? >> > >> > Does this wireshark uses some different calculation methodology for >> > calculating mean for different no of packets or I am missing something >> > ? >> > >> > Coz all are using different method for calculation so i am confused over >> > it >> > . >> > >> > >> > >> > Thanks & Regards, >> > Ravi >> > >> > On Mon, Jun 16, 2008 at 1:43 PM, Nanang Izzuddin <nanang at pjsip.org> >> > wrote: >> >> >> >> Hi Ravi, >> >> >> >> Yes, that is mean jitter, but it seems you are using 0.8.0, no >> >> standard deviation jitter calculation in that version. If you work >> >> with the latest SVN source, you will get both calculations: mean >> >> jitter & standard deviation, even some report blocks of RTCP XR are >> >> already implemented (stats summary is one of them). >> >> >> >> Cheers, >> >> nanang >> >> >> >> >> >> 2008/6/16 Ravi Inder Singh <raviinder at gmail.com>: >> >> > Hi Benny, >> >> > I need to calculate Mean Jitter & Deviation Jitter , as per RFC >> >> > 3611 >> >> > section 4.6 Stats Summary Block >> >> > Is this calculation below is mean jitter >> >> > sess->avg_jitter = >> >> > (jitter + sess->avg_jitter * sess->stat.rx.jitter.count) / >> >> > (sess->stat.rx.jitter.count + 1); >> >> > >> >> > And also Standard deviation jitter for the same. >> >> > >> >> > >> >> > -- >> >> > Thanks & Regards, >> >> > Ravi Inder Singh >> >> > _______________________________________________ >> >> > Visit our blog: http://blog.pjsip.org >> >> > >> >> > pjsip mailing list >> >> > pjsip at lists.pjsip.org >> >> > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >> >> > >> >> > >> >> >> >> _______________________________________________ >> >> Visit our blog: http://blog.pjsip.org >> >> >> >> pjsip mailing list >> >> pjsip at lists.pjsip.org >> >> http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >> > >> > >> > >> > -- >> > Thanks & Regards, >> > Ravi Inder Singh >> > _______________________________________________ >> > Visit our blog: http://blog.pjsip.org >> > >> > pjsip mailing list >> > pjsip at lists.pjsip.org >> > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >> > >> > >> >> _______________________________________________ >> Visit our blog: http://blog.pjsip.org >> >> pjsip mailing list >> pjsip at lists.pjsip.org >> http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org > > > > -- > Thanks & Regards, > Ravi Inder Singh > _______________________________________________ > Visit our blog: http://blog.pjsip.org > > pjsip mailing list > pjsip at lists.pjsip.org > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org > >
