, inclusion of the energy
consumption from clusterheads to the base station will make it hard to
illustrate the subtle di?®erence in the clustering schemes. Therefore we only
consider the energy dissipation of data transmission from children to their
clusterheads and do not count the energy consumption from clusterheads to
the base station.
We ran LEACH, Max-Min and Forest for 100 rounds over the randomly
deployed 100, 200, 300 and 400-node networks (assume each node has su?±-
cient energy and no node will die during this period) and measure the energy
dissipation for data gathering from sensor nodes to clusterheads. We assume
the size of a data packet from a sensor node to its cluster head is 2000 bits and
compute the transmission energy using the formula in Section 4.1. LEACH
has di?®erent energy consumption each round, therefore we take the average
value. For the other two schemes, with a fixed cluster structure, the energy
consumption is the same in each round.
Fig. 2. Average energy dissipation in data transmission per round.
Figure 2 shows that as the number of nodes increases, the energy dissipation
is increased in all three schemes. However, the di?®erences among them are
significant. Max-Min has a sharp increase as the number of nodes increases.
Forest is the most energy e?±cient among all three schemes. LEACH consumes
twice as much energy as Forest does when the network size is 100 nodes, and
gradually gets close to Forest as the network size increases.
Pages:
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574