Skip to Main Content
This paper presents a quadtree-based domain decomposition method for matching Global Positioning System (GPS) data onto the digital map in parallel. The method uses two basic tools, quadtree and interval distance measurement. Quadtree is a structure which can facilitate to decompose massive GPS data in a domain into multiple data pieces in the sub-domains. These data pieces with the underlying boundary-extended maps create the computing tasks, which are assigned to Cyberinfrastructure (CI) resources for parallel map-matching. In the task, each GPS point measures the interval distances between itself and the projected points on the road segments surrounding in its sub-domain, and selects the shortest one to determine the mapped point. The experiments show that this method can achieve efficient speedup in computational time by load balancing, and keep high accuracy on the matching results.