Dense MANET Identification Protocol Simulation

We tested the Dense MANET Identification Protocol in different deployment scenarios, with a number of nodes ranging from 50 to 550 (in steps of 20). The dimensions of corresponding ES/IS were scaled so as to maintain the same density in all simulations. They were set as follows:
   

Number of nodes IS Side ES Side
50 500 1000
70 592 1184
90 671 1342
110 721 1442
130 806 1612
150 866 1732
170 922 1844
190 975 1950
210 1025 2050
Number of nodes IS Side ES Side
230 1077 2144
250 1118 2236
270 1162 2324
290 1204 2208
310 1245 2490
330 1285 2570
350 1323 2646
370 1360 2720
390 1381 2792
Number of nodes IS Side ES Side
410 1432 2864
430 1466 2932
450 1500 3000
470 1533 3066
490 1565 3130
510 1599 3198
530 1628 3256
550 1658 3316



   

Participants were distributed so that the IS density was 8 time greater than the ES density. This means that 65% ca. of participants was placed within the IS. The number of neighbors required for belonging to the dense MANET was set to 11. This value was experimentally obtained to make the dense MANET bounds almost coincide with IS bounds. Other parameters were set as follows:

   
Paramter Value
Neighbor Limit 11
Discovery Delay 5 s
Participant Advertise Interval 20 s
Participant Advertise Delay 5 s
   
   

Next figure shows the average number of packets sent/received by each participant, by considering all the different node densities.