Voice only analysis

From TD-er's Wiki
Revision as of 13:08, 23 February 2007 by Rubzzzz (talk | contribs) (→‎Experimental Setup)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

Experimental Setup

Because it is impossible to measure data traffic between two VM ware's on the same machine, the following setup is used. The SIP-phones all use the g711u codec.

Schematic overview of all PC's running which software.

Results

With WireShark data has been collected and the following filters are used in the in I/O-graph:

Filter-Black: ip.src == 192.168.1.101 && ip.dst == 192.168.1.106
Filter-Red  : ip.src == 192.168.1.3   && ip.dst == 192.168.1.106
Filter-Green: ip.src == 192.168.1.189 && ip.dst == 192.168.1.101
Filter-Blue : ip.src == 192.168.1.106 && ip.dst == 192.168.1.101
Filter-Pink : ip.src == 192.168.1.100 && ip.dst == 192.168.1.101

Wireshark-graph G711 G711 traffic Bps oneway-traffic.png

Legenda:

  • Blue: Traffic between the Asterisk-boxes in the other direction.
  • Black: Traffic between the Asterisk-boxes in one direction.
  • Red: Traffic to and from SIP phone 301.
  • Pink: Traffic to and from SIP phone 202.
  • Green: Traffic to and from SIP phone 201.

Conclusions

  • From the graph shown above there can be concluded that the IAX2 traffic (black and blue line) uses less bandwidth than the SIP-traffic (red and green line). This is due to the IAX trunking which reduced the overhead.
  • We also filtered for traffic between all three SIP-telephones and no direct traffic was detected between them.
  • The audio traffic is measured to be approximately 10500*8/1000=84kbps, which corresponds very well to the expected NEB value of 87.2 kbps.