원문정보
초록
영어
Advances in mobile computing, wireless communication, and positioning technology have flourished Location Based Service (LBS). This results in a large amount of trajectories accumulated in LBS applications. As a fundamental research spot, trajectory matching algorithm has drew much attention from image and transportation communities. While the various location update strategies in LBS make trajectory matching a more challenging task, since the different samples of one trajectory should not be falsely recognized as another trajectory. Besides this, LBS applications also require the trajectory matching algorithm to be sensitive to time sequence and tolerate to time scaling. A dynamic time warping based algorithm is proposed in this paper. Extensive experiments show the effectiveness, time sequence sensitiveness and time scaling toleration of the proposed algorithm.
목차
1. Introduction
2. Related work
2.1. Classification of Similarity Calculation
2.2. Existing Algorithms
2.3. Summary of Existing Algorithms
3. Formal Definition of the Problem
3.1. Trajectory Model
3.2. The Goals of Trajectory Matching Algorithms for LBS
4. Dynamic Time Warping based Algorithm
5. Experimental Evaluation
5.1. Experiments Setup
5.2. Similarities Calculations
5.3. The Effect of Dt
5.4 The Effect of Dd
6. Conclusions
Acknowledgements
References