Approximate path searching for supporting shortest path queries on road networks

Although various efficient indexing techniques have been proposed to improve the performance in distributing road data to mobile clients to perform shortest path searches on road networks, how to minimize the tune-in and path searching costs are still important performance goals since the clients usually have limited processing capability and energy supply. In this paper, based on the Next Region (NR) approach, we propose a new method called Approximate Path Searching (APS) for constructing the broadcast index at mobile clients with soft arrival times to destinations. APS adopts an approximation technique in calculating the sets of required regions for each 'start, destination' region pair in a road network to reduce the index generation cost at the road server and the tune-in and path searching costs at the clients. Different from NR, which only provides one set of required regions containing the shortest path from a client's start region to its destination region, APS generates n-sets of required regions for a client to choose. Each set of required regions contains different numbers of regions. Although choosing a smaller region set may give a slightly longer delay in arrival, the tune-in and path searching costs could be significantly reduced. Theoretical analysis and extensive simulation experiments have been performed to demonstrate the effectiveness of APS in improving the system performance with just a few percentages delay in arrival time on average comparing with NR using data sets from real road networks.