Cloud computing-based map-matching for transportation data center

• Propose a leapfrog method to improve the efficiency of map-matching algorithm.
• Use MapReduce to adapt the serial map-matching algorithm for cloud computing environment.
• Propose a privacy-aware map-matching model over hybrid clouds.

Transportation data center has recently become a common practice of modern integrated transportation management in major cities of China. Being the convergence center of large-scale multi-source vehicle tracking data, it caused great challenge on GPS map-matching efficiency and privacy protection. In this paper, we propose a secure parallel map-matching system based on Cloud Computing technology to meet the demand of transportation data center. The main contributions are as follows: (1) we propose a leapfrog method to improve the efficiency of traditional serial map-matching algorithm on the increasingly common high sampling rate GPS data; (2) we adapt the serial leapfrog map-matching algorithm for cloud computing environment by reforming it in the MapReduce paradigm; (3) we propose a privacy-aware map-matching model over hybrid clouds to realize the sensitive GPS data protection. We implemented the proposed map-matching system in the hadoop platform and tested its performance with a large-scale vehicle tracking dataset, which exceeds 100 billion records. The experimental results show that our approach is highly efficient and effective on massive vehicle tracking data processing.