期刊名称：International Journal of Computer Science and Network Solutions
印刷版ISSN：2345-3397
出版年度：2016
卷号：4
期号：1
页码：1-9
出版社：International Journal of Computer Science and Network Solutions
摘要：One of the most important methods for energy consumption reduction in Wireless Sensor Networks(WSNs) is clustering. In this respect, cluster head selection method can highly impact on energy savingin the overall network. In mobile scenarios, for gathering environmental information as an example,due to natural barriers in operational environment, topology isn’t fixed and changes continuously.Therefore, cluster head selection can be adapted based on physical location of sensors. This means thatthe nodes with fewer barriers in front of them are better to be selected for cluster head. The traditionaland statistical methods can’t consider the topology of network dynamically. In this paper a newintelligent method based on Artificial Neural Network (ANN) is proposed in order to reduce energyconsumption and increase the network performance by choosing the cluster head based on residualenergy of nodes and physical positions of them (topology). In this method, at first the positions ofnodes mapped to a 2D pattern; then this pattern feeds into ANN that has previously trained and as theoutput of ANN the best CH position is suggested. The proposed method is compared with the LowEnergy Adaptive Clustering Hierarchy (LEACH) algorithm and the results show that it outperformsLEACH method especially when the nodes are mobile
关键词：Wireless sensor networks; Artificial Neural Networks; Clustering; Energy management;LEACH.
