Energy-efficient Adaptive Distributed Data Collection method for Periodic Sensor Networks

Abstract

This article suggests a method, called energy-efficient adaptive distributed data collection method (EADiDaC), which collects periodically sensor readings and prolong the lifetime of a periodic sensor network (PSN). The lifetime of EADiDaC method is divided into cycles. Each cycle is composed of four stages. First, data collection. Second, dimensionality reduction using adaptive piecewise constant approximation (APCA) technique. Third, frequency reduction using symbolic aggregate approximation (SAX) approach. Fourth, sampling rate adaptation based dynamic time warping (DTW) similarity. EADiDaC allows each sensor to remove the redundant collected data and adapts its sampling rate in accordance with the monitored environment conditions. The simulation experiments on real sensor data by applying OMNeT++ simulator explains the effectiveness of the EADiDaC method in comparison with two other existing methods.