Minimization of the number of cluster heads in a wireless sensor network is a very important problem to reduce channel contention and to improve the efficiency of the algorithm when executed at the level of cluster-heads. This paper proposes a Self Organiz
2. The Clustering Problem
Clustering problem can be defined as following. Assume that nodes are randomly dispersed in a field. At the end of clustering process, each node belongs to only one cluster and be able to communicate with the cluster head directly via a single hop [3]. Each cluster consists of a single cluster head and a bunch of followers as illustrated in Figure 1. The purpose of the clustering algorithm is to form the smallest number of clusters that makes all nodes of network to belong to one cluster. Minimizing the number of cluster heads would not only provide an efficient cover of the whole network but also minimizes the cluster overlaps. This reduces the amount of channel contention between clusters, and also improves the efficiency of algorithms that executes at the level of the cluster-heads.

Figure 1. Clustering in a sensor network
3. Self Organizing Sensor (SOS) Networks by Minimization of Cluster Heads Using Intelligent Clustering
This Section presents the proposed clustering algorithm in a global scale, and the following section describes the algorithm at a node level.
3.1 Global level of clustering algorithm
The following steps illustrate an overview of the suggested algorithm.
1. Find the node (No), which has the maximum number of followers, and make a cluster with it.
2. Include clustered nodes into a clustered node set G.
3. Selects the next head node (Nf), which can communicate with a node in G and has the maximum
number of followers, and make a cluster with it.
4. If there exist an unclustered node or nodes then go to step 2
5. Else terminate the algorithm.
At first, it makes a cluster with the center node which has the maximum number of followers. We assume that there is a coordinator which controls globally in the entire network (for easy understanding). So it does not matter to locate the center node during step 1. In step 2, it includes the selected cluster head node and its followers to the clustered node set G. And in step 3, it selects the node, which can communicate with a node in G and has the maximum number of followers, and makes a cluster with it as a cluster head and include it and its follower to set G. Figure 2 illustrates step 3. A node 'a' is No node and node 'b' is the node which can communicate with the next head node (that is, node 'c'), which has the maximum number of followers. Then it elects node 'c' as a next cluster head node and makes a cluster with it. The process is then repeated until all the nodes are clustered.
built using piezoelectric quartz crystal sensors. ...feed-forward network theory and self-organizing ... v i ) (6) The minimization of the c-means...
built using piezoelectric quartz crystal sensors. ...feed-forward network theory and self-organizing ... v i ) (6) The minimization of the c-means...
热门文档