Location Service Data Management And Application Research In Mobile Clouding Environment

With the development of Geographic Information System (GIS) and the development of mobile location and3G technology, the Location Based Service (LBS) applications grow rapidly, and the spatial data which a hand-held equipment to process is becoming more larger. In this paper, we develop a LBS application called Large-scale Car Ridesharing System in mobile cloud computing environment. Efficient processing of increasing massive data is crucial and challenging to the appilications in mobile cloud computing environment. Traditional spatial data index has limitations, only highly extended and distributed spatial index can achieve large-scale spatial data analysis and query tasks efficiently. There has been distributed parallel implementation methods using MapReduce for spatial query index, such as R-tree and Voronoi index. These methods have drawbacks:R-tree is not suitable for parallelization; Although Voronoi index can be used for parallel query, it should be reconstructed calculating local index when query with it.Compared to the above two methods, the grid index is more extensible and parallel. The inverted index can create index for unlimited data points only by limited i ndex entries. Combining the advantages of grid index and inverted index, we present the inverted grid index with MapReduce in this paper, Since inverted grid index is more simple, no sharing and loose coupling, so it is more suitable for parallelization. Based on Inverted grid index, we present the parallel KNN a-lgorithm using multiple threads, which can accelerate the search efficiency query of k nearest neighbors. Finally, on the basis of inverted grid index and parallel KNN method, we develop a Large-scale Car Ridesharing System, on one hand it verifys the feasibility of inverted grid index and parallel KNN method, on the other hand, it makes people travel conveniently. The cloud computing spatial index and query technique in this paper is suitable for developing applications of location based service, provid ing a new idea for LBS application development.