Abstract: Preserving the energy of sensor nodes during data collection is always one of the most crucial problems in wireless sensor networks. In this project, the DCS scheme is proposed to exploit the appearing temporal-spatial correlation in most observable event for energy-efficient data collection of wireless sensor networks. Particularly, for temporal correlation between sensor nodes, to build lightweight Autoregressive model locally to capture data distribution at sensor node; for spatial correlation, by making use of similarity measure between sensor nodes and to perform centralized model clustering, where, this kind of clustering used to emphasizes data similarity between nodes but ignores the distance geographically, to make a group of sensor nodes with similar data distribution on both magnitude and trend into the same cluster. Through scheduling sensor nodes to report readings alternately and performing dual-prediction at both sensor nodes and Sink, Data Collection Scheme obtain sensing readings without compromising too much data accuracy loss. The wireless sensor networks are being deployed at escalating rate for various application fields. Wireless sensor networks (WSN) are mainly useful for obtaining data concerning events limited to a well-defined geographic region, like a disaster site or an agriculture dataset. The critical issue in wireless sensor networks is power saving since sensor nodes are battery-powered. Hence, developing secure and energy-efficient routing algorithm to guard WSNs against these attacks while efficiently utilizing the energy of the deployed nodes has become essential.
Keywords: Time Series Analysis, Energy Efficient technique, Temporal-Spatial, Data Collection, Wireless Sensor Network.