SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 556 | Next

Yingshu Li, My T. Thai, and Weili Wu

"Wireless Sensor Networks and Applications"

One
remedy for this is to split the cluster from the most recent merging edge and recluster
each fragment with other clusters in the neighborhood. Re-clustering
can be done recursively until there is no small fragment left. The two control
parameters are N and ?±.
4 Performance Comparison
4.1 Network Model
The overall goal of the simulation is to compare the three clustering schemes in
a sensor network environment.We simulate sensor networks on a 150m?—150m
region. Sensor nodes are randomly deployed with di?®erent levels of density.
The communication range of each sensor is set to 20 meters. Two nodes have
a wireless link between them if they are within the communication range of
each other.
Initially, each node was assigned a unique id, random x, y coordinates
within the region and 0.5J of energy. Then we run the three schemes separately.
For each simulation run, some important statistics were measured.
We assume the same radio model as in [10], where the radio dissipates
Eelec = 50nJ/bit to run the transmitter or receiver circuitry and ?«amp =
100pJ/bit/m2 for the transmit amplifier. We also assume an r2 energy loss
due to channel transmission, where r is the transmission distance. Thus, to
transmit a k-bit message over a distance of d meters using our radio model,
the radio expends
ETx (k, d) = ETx??’elec (k) + ETx??’amp (k, d)
= Eelec ?— k + ?«amp ?— k ?— d2
and to receive this message, the radio expends:
ERx (k) = ERx??’elec (k)
= Eelec ?— k.


Pages:
544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568