基于DR-tree的室内移动对象索引研究

Indoor Moving Objects Index Research Based on DR-tree

对于移动对象历史轨迹索引,现有的方案绝大多数都基于室外空间,难以直接应用于室内空间中;同时,未将对象本身作为一个独立的维度加以索引,无法提供高效的对象轨迹查询方式。对此,提出了一个室内环境下的移动对象索引结构DR-tree来对移动数据的位置、时间、对象三个维度进行索引,并将位置维与对象维解耦,将三维索引转换为两个二维索引,同时给出查询优化方案。实验结果表明,与现有的室内环境下的索引方案RTR-tree相比,该结构不仅能够提供高效的时空查询,而且还能提供高效的对象轨迹查询。

For the index of historical trajectories of moving objects,most of the schemes are based on outdoor space,which are hard to be directly applied to indoor space.Moreover,the object itself is not indexed as an independent dimension and the efficiency of the queries based on objects is quite low.Thus,this paper proposed an index structure DR-tree（Dual R-tree） which can index three dimensions,such as the localization,the object and the time.This scheme can convert the three-dimension index into two two-dimension index by decoupling the location and object dimension,and provide query optimization method.The experimental results show that compared with RTR-tree,DR-tree,the scheme can not only support the efficient spatiotemporal query,but also provide the trajectory query based on objects.