Convergecast Tree Construction in Wireless Sensor Networks
Lenka Carr-Motyčkova and David Dryml
Data gathering is important research topic in the area of wireless sensor networks. The basic operation is collecting data from sensor nodes and their transfer to sink for further processing. Sensors are often powered by batteries with limited amount of energy. Therefore wireless sensor networks are able to work only for a limited amount of time. Network lifetime is defined as a time until the first sensor node depletes its energy. Main challenge in data gathering algorithms is to extend a network lifetime.
In this paper we study a scenario where the sink needs to collect data that is sensed in a distant disc target area (surroundings of a specified geographical position) of a network. The sink is located away from the target area. Main aim is to construct convergecast tree that connects sink with target area and allows data collection from it. Particularly we are interested in maximizing network lifetime through careful construction of this tree.
Nodes are selected to be tree nodes only if their residual energy after the previous round of data gathering is sufficient to perform the next data transfer. More precisely the selected neighbors are expected to have the maximal energy among considered neighbors. A new convergecast tree is built, when the current one is not able to continue a data transfer. This procedure of building convergecast tree is repeated until the first node in the transfer area runs out of battery power. According to simulations our approach doubles the lifetime of network compared to a tree constructed randomly and moreover it is able to use more than twice the amount of energy in nodes. Latter on we discuss the possibility of comparison with other state-of-the-art algorithms.
Keywords: Data gathering, wireless sensor networks, maximum lifetime, convergecast tree, energy consumption