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
network as a cost function to form clusters. Mhatre and Rosenberg [7] take into account not only the battery of the nodes but also the manufacturing cost of hardware.
Krishnan and David Starobinski [14] used a message-efficient clustering, in which nodes allocate local “growth budgets” to neighbors. The algorithm produce clusters of bounded size and low diameter, using significantly fewer messages than the earlier, commonly used, expanding ring approach. They also presented a new randomized methodology for designing the timers of cluster initiators. This methodology provides a probabilistic guarantee that initiators will not interfere with each other.
Liu and Lin [15] introduce a re-clustering strategy and a redirection scheme for cluster-based wireless sensor networks in order to address the power-conserving issues in such networks, while maintaining the merits of a clustering approach. Based on a practical energy model, their simulation results show that the improved clustering method can obtain a longer lifetime when compared with the conventional clustering method.
When sensor nodes are organized in clusters, they could use either single hop or multi-hop mode of communication to send their data to their respective cluster heads. Mhatre and Rosenberg [16] presented a systematic cost-based analysis of both the modes, and provided guidelines to decide which mode should be used for given settings. They also proposed a hybrid communication mode which is a combination of single hop and multi-hop modes, and which is more cost-effective than either of the two modes.
Younis et al. [17] present a novel approach for energy-aware management of sensor networks that maximizes the lifetime of the sensors while achieving acceptable performance for sensed data delivery. The approach is to dynamically set routes and arbitrate medium access in order to minimize energy consumption and maximize sensor life. The approach calls for network clustering and assigns a less-energy-constrained gateway node that acts as a cluster manager. Based on energy usage at every sensor node and changes in the mission and the environment, the gateway sets routes for sensor data, monitors latency throughout the cluster, and arbitrates medium access among sensors.
Pan et al. [18] considered a generic two-tiered wireless sensor network (WSN) consisting of sensor clusters deployed around strategic locations, and base-stations (BSs) whose locations are relatively flexible. Within a sensor cluster, there are many small sensor nodes (SNs) that capture, encode, and transmit relevant information from a designated area, and there is at least one application node (AN) that receives raw data from these SNs, creates a comprehensive local-view, and forwards the composite bit-stream toward a BS. Their research focus on the topology control process for ANs and BSs, which constitute the upper tier of two-tiered WSNs. By proposing algorithmic approaches to locate BSs optimally, they maximized the topological network lifetime of WSNs deterministically, even when the initial energy provisioning for ANs is no longer always proportional to their average bit-stream rate. By studying intrinsic properties of WSNs, authors established the upper and lower bounds of maximal topological lifetime, which enable a quick assessment of energy provisioning feasibility and topology control necessity.
Clustering problems arise in many different applications too, such as data-mining, knowledge discovery and pattern recognition. These use approaches based on randomization such as CLARA [8] and CLARANS
[9], methods based on self organizing maps [10], and techniques designed to scale for large databases, including DBSCAN [11], BIRCH [12] and ScalKM [13].
In this paper, we propose a new clustering algorithm that does not require manually adjusted parameters which could also provide identical results in each test on the same sensor network to overcome the weakness of ACE. Rest of the paper is organized as follows. In Section 2, we present the clustering problem followed by Section 3 wherein the new algorithm is illustrated. Experiment results are presented in Section 4 and some conclusions are also provided towards the end.
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