Trajbase

Abstract

The recent years witnessed popular use of smart phones and wearable devices. Such mobile devices are widely equipped with GPS sensors to record their geographical positions, generating massive spatial trajectory data. Many real applications, such as urban computing and intelligent transportation systems, search the trajectory data. To process such queries, it is important to meet two objectives including fast running time and low space cost. Previous work frequently only meets one objective but compromises another. In this work, we design a compact data structure TrajBase to index trajectory data. The novelty of such an indexing structure is a carefully designed encoding approach. The approach can achieve comparable compression ratio compared with the state of the art approach and fast building time. Extensive evaluation with two real trajectory datasets can successfully verify that TrajBase outperforms previous works in terms of query processing time and indexing space cost.