摘要：In Wireless Sensor Network (WSN), sensors are devices used to collect data from the environment related to the detection or measurement of physical phenomena. The original sensor network data are commonly large amounts of measurements and high-dimensional. Spatial data mining techniques used to retrieve high-dimensional spatial information from WSN is effectively. Due to the limited energy of the sensors in WSN, energy efficient query evaluation is critical to prolong the system lifetime. In this paper, we propose a local mapping based modeling sensor network data algorithm for query processing to render effective spatial data retrieval services in the resource constrained WSN. The algorithm can be associated with the present features in spatial data and project high-dimensional data to feature space by using local mapping linear embedding for reducing the computation of spatial data retrieval. As the experiment results shown, the algorithm can preserve important spatial feature information and provide effective preprocessing analysis results. Furthermore, the algorithm is the more effectual on feature preserving than local linear embedding (LLE) when the sample space is sparse source data.
关键词：wireless sensor network;query processing;local mapping;spatial data retrieval
