Vol.62 (Sensor 2014), pp.21-24

Cluster-based structure is not a new research topic but it still challenges the scientists. In wireless sensor network, cluster-based is one of important approaches to achieve effective network performance such as energy saving, throughput maximizing, end-to-end delay and maintenance cost decreasing, etc. In this paper, we propose a clustering scheme which utilizes the geographical information in order to construct the clusters, and is used in underwater wireless sensor networks. First, we use the geographical information and skewness measurement to classify the underwater wireless sensor network into even or skew distributed; Secondly, we construct the clusters using also the geographical information and network type to elect a group of cluster head candidates so that the networks can be maintained for achieving the high performance. Simulation results show that: the proposed clustering scheme can obtain higher throughput with less end-to-end delay as well as decrease energy consumption of the network.