All these drawbacks occur because this approach does not
have any restriction on the number of nodes in a cluster. In this chapter, we
present a new topology-based heuristic, Forest scheme, in which clusters are
formed as disjointed trees and the clusterheads are roots of trees. To solve
these problems, we add restrictions on the cluster size in the Forest scheme.
2.3 Energy-based Heuristic
With the rapid development of sensor networks, many energy-based heuristics
have been proposed ([10], [13], [17], [9], [14] and [1]). LEACH([10]) is
a well-known representative in this category and will be discussed in detail
in Section 3. HEED (Hybrid Energy-E?±cient Distributed clustering) ([14])
periodically selects clusterheads based on the hybrid of their residual energy
and a secondary parameter, such as node proximity to its neighbors or node
degree. In general, the energy-based heuristics have lower energy dissipation
and longer system lifetime than other schemes.
3 Overview of Algorithms
In this section, we will first provide an overview of the schemes to be compared:
LEACH, Max-Min D-cluster and a new scheme, Forest. Each of the
three schemes is a representative of energy-based, ID-based and topologybased
clustering schemes respectively. We briefly describe the key features of
each scheme and the key control parameters in the implementation.
3.1 LEACH: Low-Energy Adaptive Clustering Hierarchy
In conventional clustering algorithms, clusterheads are chosen a priori and
fixed throughout the system lifetime.
Pages:
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562