Greetings,
Using EMANE 0.9.1 I have a two NEMs configured to use raw transport on
two machines, one NEM per machine. One NEM is stationary and one is mobile,
both have one ieee80211abg interface. Traffic flows properly between two
external machines, that are my traffic generator and sink, through the NEMs
with the exception of the packet loss rate.
The scenario.eel file I'm using contains the following pair of lines:
2.6 nem:2 location gps 36.2419,-116.8258,120.0
2.6 nem:1 pathloss nem:2,50.3,50.3
Similar pairs of lines are included to modify location and pathloss every
200 milliseconds. These emit events that I am recording on the traffic
generator along with the congestion window of a single TCP stream to the
traffic sink. I compared this to the results of the live test that I used
to generate the scenario.eel file and discovered that EMANE is not changing
the rate of packet loss based on the changes to pathloss.
Example results: http://imgur.com/a/Liu5g
Notes: Although packet loss was not recorded the TCP congestion window
can stand in as a reasonably reliable proxy for packet loss.
The graphs both display RSSI rather than pathloss. The RSSI
value for the live graph was obtained directly from the WiFi card.
The RSSI value for the EMANE graph is calculated from the
pathloss and TX power.
Even manual changes to the TX power of the NEMs to excessively high
values (200dBm instead of 20dBm) had no effect on the packet loss rate. I
would appreciate any help debugging this.
John
_______________________________________________
emane-users mailing list
[email protected]
http://pf.itd.nrl.navy.mil/mailman/listinfo/emane-users
