Figure 1 presents experimental results about RDM network message overhead, normalized on the number of dense MANET nodes.
In other words, it shows the average number of REDMAN messages sent/received by each node for maintaining the required
resources replication degrees.
We measured the overhead in different simulation conditions. In particular, we varied the number of nodes belonging
to the network
(we tested 400 and 700 node scenarios) and the node speed (we tested the protocol for nodes moving at 1,2,3 and 4 m/s).
As we can see, the overhead slightly increases when speed grows; with regard to the number of nodes, we can argue
that the normalized overhead even decreases when passing from 400 to 700 node scenario.
The figure also shows that most traffic (between 91% and 94% on average) is represented by Hello packets exchanged
to maintain an updated view of the network topology
(DMC functionality).
The Hello period can be configured and affects the REDMAN promptness to detect node departures and to re-establish
required replication degrees. For all the simulation runs, we have chosen a 100s Hello period, which demonstrated small
enough for the 4m/s node speed. This means that in slower speed scenarios, the Hello period can be reduced, thus
carrying a drastic cut also in the number of messages exchanged by each node.
