Um algoritmo de amostragem multivariada para redes de sensores sem fio.

Abstract

A wireless sensor network (WSN) is energy constrained, and the ex-tension of its lifetime is an important issue in its design. A WSN can be used to collect and process data of environment that can be, in some cases, multivariate. In this way, to help the data processing, this work proposes a multivariate sampling algorithm which uses component analysis techniques to rank the data and then to select only the data more relevant to the application. Simulation results show that our technique reduces the data keeping its representativeness. In addition, the energy consumption and delay on the network are reduced.