4.2. Performance analysis
In the first scenario we consider an alarm originating at the concert area at 05:00 pm with an enabling radius equal to
200 m, and we analyze the results of the experiment as a function of the number of active nodes.
Fig. 4shows a qualitative comparison (result of a single simulation) of the alarm diffusion performance with 100 and 700
active nodes. In the figure, the enabling area is denoted with a blue circle, the 1 km target nodes with red circles and the 2 km
ones with orange circles, while the green circles represent the infected nodes. Obviously, the alarm spreading over time is
more evident when 700 nodes are active. Moreover, when 700 nodes are active, the alarm spreads in the larger (2 km) target
area, while the alarm is limited to the smaller (1 km) target area if only 100 nodes are active. Two factors concur in limiting
the alarm diffusion: the low device density (with respect to the transmission range) and the limited mobility.
Therefore, inFig. 5we conduct a quantitative comparison of four different metrics as a function of time: (i) the number of
enabling nodes (Fig. 5-a); (ii) the number of infected nodes (Fig. 5-b); (iii) the infection rate for the 1 km target area (Fig. 5-c);
(iv) the infection rate for the 2 km target area (Fig. 5-d).
Clearly, the number of enabling nodes (Fig. 5-a) depends on the number of active nodes. However, even if almost all the
nodes cooperate in the alarm diffusion, the average enabling node number is lower than eight. Moreover, the maximum
number of enabling nodes is reached at the beginning of the alarm spread and it remains steady, meaning that the alarm
spread is mainly due to broadcasting.