Efficient Processing of Updates for Moving Objects with Varying Speed and Direction

Spatio-temporal databases aim at appropriately managing moving objects so as to support various types of queries. While much research has been conducted on developing query processing techniques, less effort has been made to address the issue of when and how to update location information of moving objects. Previous work shifts the workload of processing updates to each object which usually has limited CPU and battery capacities. This results in a tremendous processing overhead for each moving object. In this paper, we develop a novel update strategy, namely the possible region update strategy (PRUS), to explicitly indicate a time point at which object needs to update location information. As each object knows in advance when to update (meaning that it does not have to continuously check), the processing overhead can be greatly reduced. In addition, a possible region update procedure (PRUP) is designed to efficiently process the updates issued from moving objects. Extensive experiments are conducted to demonstrate the effectiveness and the efficiency of the proposed PRUS and PRUP.