Análisis multiobjetivo de la selección de líderes en redes inalámbricas de sensores

A Wireless Sensor Network (WSN) is composed of a set of energy and processing-constrained devices that gather data about a set of phenomena. An efficient way to enlarge the lifetime of a wireless sensor network is clustering organization, which structures hierarchically the sensors in groups and assigns one of them as a cluster-head. Such a head is in charge of specific tasks as data gathering from other cluster sensors and resending that data through the network to the base station. The cluster-head selection may be random or based on well defined criteria such as residual energy, node distance, signal strength or connectivity; hence, this problem is NP-hard for WSN. In the literature we may find proposals based on heuristics algorithms that consider at the same time different objectives and purposes. Therefore, in this paper we perform an analysis from the many-objective perspective about the pertinence of the most commonly used objectives.