I tried to repeat the experiments
done by the CMU group to test the different protocols - DSR, DSDV, AODV and
TORA. When I tried to test the DSR
protocol using exactly the same set of parameters as in the eariler
experiment,
my Packet Delivery Rate (PDR) was
only about 50% compared to the over 95% result obtained by the CMU
group.
I have tried using different scenario
files with the same parameters but the results are still as bad.
To get the statistics as in the CMU's
experiment, I modified their total.pl perl script to analyse the trace
files.
Basically to get the PDR, I used the
script to search for 'AGT' in the trace files and then see whether the
opcode
is a 's' (send) or a 'r' (receive).
After counting the number of 's' (packet sent) and 'r' (packet received), I will
be
able to calculate the PDR.
Can anyone help me on this? Perhaps someone can
conduct the same experiment to check whether the protocol
is working. Thanks a
lot.
|