Query Processing Techniques in Spatio-Temporal Databases

• Main Authors: Yuan-Ko Huang, 黃淵科
• Other Authors: Chiang Lee
• Format: Others
• Language: en_US
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/38957535696782781933

博士 === 國立成功大學 === 資訊工程學系碩博士班 === 97 === Spatio-temporal databases aim at combining the spatial and temporal characteristics of data. Spatio-temporal queries have been used in many applications such as mobile communication systems, traffic control systems, geographical information systems, location-aware advertisement, and multimedia applications. Continuous Range (CR) query and Continuous K-Nearest Neighbor (CKNN) query are two important and widely used spatio-temporal queries.A CR query is to find the moving objects whose Euclidean distances to the moving query object are within a distance at each time instant within a time interval [ts, te]. As for the CKNN query, it can be utilized to retrieve the query object's K-Nearest Neighbors (KNNs) at each time instant within [ts, te]. The context of this dissertation is categorized into two parts according to whether objects are moving in Euclidean space or road networks. In the first part, we investigate how to efficiently process these spatio-temporal queries over moving objects in Euclidean space (that is, the results are determined based on the Euclidean distance between each moving object and the query object). Existing methods for processing spatio-temporal queries, however, assume that each object moves with a fixed velocity (speed and direction). Different from the existing methods, we relieve this assumption by allowing the velocity of each object to be uncertain. This uncertainty on the velocity of object inevitably results in high complexity in processing spatio-temporal queries. We will discuss the complications incurred by this uncertainty and propose several efficient methods to answer the spatio-temporal queries over moving objects with uncertainty. Our performance results demonstrate the effectiveness and the efficiency of the proposed approaches. In the second part of this dissertation, we focus our attention on processing the spatio-temporal queries over moving objects in road networks. As the movements of objects are constrained to a road network, the distance between two objects should be computed based on the connectivity of the road network rather than the two objects' locations. We first highlight the limitations of the existing apporaches for processing the spatio-temporal queries in road networks, and then propose a cost-effective method to overcome these limitations. By using the proposed method, we can determine the result set of spatio-temporal queries under the following three conditions: (1) All objects (including the query object) move continuously in a road network. (2) The distance between two objects is defined as the distance along the shortest path between them in the network. (3) The result set of the query object at each timestamp should be completely determined. Comprehensive experiments are performed to investigate the efficiency of this method.