The Design and Application of Navigation and Location Services Data Index

This paper defines the existing navigation and location services data and six kinds of retrieval patterns. By contrasting to the existing trajectory index methods, we propose several index algorithms of different dimensions in the MapReduce mode, particularly, a spatiotemporal index algorithm named Layer-by-layer Index Based on Grid Partition (LIBGP), which efficiently avoids the problem of low trajectory recognition rate. In order to achieve real-time query to a large number of location services data, GPS data as an example, we present a multidimensional retrieval framework, combining the index mechanism of Hbase with our proposed algorithms. An analysis of our experimental results indicates that our methods build a flexible retrieval with high efficiency.