[ns] Fw: DSR bad results using ns2

=?gb2312?B?utXR9A==?= heyanghappy at sina.com
Thu Apr 26 16:50:21 PDT 2007


Hi all,

I am working for dsr analyse and simulation. When I simulate DSR using ns2-2.29, sometime I get very bad results, but sometime I can alse get normal results.

I simulated dsr protocol in such situation:

In tcl: ifqlen set 50, 802.11 datarate set 11Mb, ratio range set 250m
dsr all opt enable, including ring 0 search, cache reply, salvage, shorten, etc.
cbr:50n-20mc-4rate
scene:50n-300p-20M-900t-1500x-300y

I generated six scene files using setdest.exe in the same scene situation as above show.

the results:
scene  overhead£¨route/data/ratio£©delay£¨s£©   pdf(send/recv/ratio) 
scene1£º304901/184097/1.656198  0.426943  64292/55843/0.868584
scene2£º7703/204136/0.037735  0.024476  64291/64208/0.998709
scene3£º440132/188394/2.336232  0.767900  64211/50903/0.792746
scene4£º 664971/232313/2.862393  0.679598  64479/30500/0.473022
scene5£º227687/179718/1.266913  0.499662  64277/61165/0.951585
scene6£º8151/157065/0.051896  0.012842  64377/64339/0.999410
I think only scene2 and scene6 are reasonable results, and others are bad results.

In .tr files, the main reason of the dropping packets is ifq and arp.

I have checked Maltz's article "The effects of  on-demand behavior in routing protocols for multihop wireless ad hoc networks".
Maltz did this simulation with datarate 2Mb, and all results seems OK.

I don't know what's wrong with my simulation. Can anyone help me! Thank you very much.





heyang
2007-04-27



More information about the Ns-users mailing list