Efficient Similarity Query in RFID Trajectory Databases

摘要：

Similarity query is one of the most important operations in trajectory databases and this paper addresses this problem in RFID tra jectory databases. Due to the special sensing manner, the model of RFID trajectory is different from the traditional trajectory, leading to the in efficiency of existing techniques. This paper proposes a novel distance function-EDEU(Edit Distance combined with Euclidean Distance), which supports the local time shifting and takes the distance between adja cent logic areas into consideration. We also develop two filter-refinement mechanisms based on Co-occurrence Degree and Length Dispersion Ra tio to improve the efficiency of the similarity analysis. Furthermore, we extend our solution to detcrmine the local similarity from the global dissimilarity trajectory pairs. The extensive experiments verify the effi ciency of our proposed algorithms.

作者: Yanqiu Wang Ge Yu Yu Gu Dejun Yue Tiancheng Zhang

作者单位: Northeastern University, China

会议类型: 国际会议

会议名称: 11th International Conference,WAIM 2010(第十一届网络时代管理国际会议)

会议地点: 九寨沟

会议语种:英文

页码: 620-631

在线出版日期: 2010-07-14（万方平台首次上网日期，不代表论文的发表时间）