The present study was conducted with a focus on methods of maximizing the life time of entire networks in ubiquitous sensor networks (USN). To maximize life time, the present study concentrates on energy consumption algorithms, which are applied to USN with cluster tree topology. In general, the operation of sensor networks with a cluster tree topology can largely be classified into two stages: the initialization stage, where clusters are constructed, and the operation stage, where, after initialization, the data collected are transmitted. In sensor networks with a cluster tree topology, the initialization and operation stages continue repeatedly. As for energy consumption in sensor nodes, it occurs at both the initialization and operation stages. The present study focuses on algorithms that minimize the energy consumed at the operation stage.

The process from the construction of clusters in cluster tree-based routing to the selection of cluster heads, exchange of routing tables among clusters, and transmission of messages to sink nodes requires the transmission of packets, which are very important components that affect not only the batteries of individual sensor nodes but also the life time of entire networks. The selection of which nodes in clusters play the roles of cluster heads can be an important factor affecting the life time of entire clusters, because nodes that play the roles of cluster heads consume more energy than do other nodes in the clusters. Consequently, it is possible to extend the life time of entire networks simply by selecting appropriate cluster heads. The present study proposes routing algorithms that, in selecting cluster heads, delay the head replacement period when a certain level of energy remains after the amounts of residual energy in various nodes have been compared.