A Novel Mobility Similarity Measurement Method to Increase the Performance of Community-based Video Delivery in VANETs

The mobility of mobile nodes is a distinctly important influence factor for video sharing performance, user quality of experience and traffic load remission of core networks in vehicular ad hoc networks (VANETs). In this paper, we propose a novel mobility similarity measurement method to increase performance of community-based video delivery in VANETs (MSMM). In order to accurately represent movement trajectories of vehicles, MSMM calculates relative location between vehicles to refine the geographical location of vehicles. MSMM investigates continuous variation of refined vehicle location to estimate subjection relationship between vehicles and roads and designs a line-segment- based representation method for movement trajectories of vehicles according to the subjection relationship. By building an estimation model of traffic of roads in terms of the hydromechanics and the vehicle following model and by analysis for the historical movement trajectories of vehicles to calculate traffic of roads, MSMM extracts the movement patterns of vehicles. MSMM further respectively designs a recognition method of movement patterns of vehicles and a similarity estimation method of movement behaviors between vehicles, which enables the video requesters to select the video providers which have similar movement behaviors and implement high-efficiency video sharing. We use MSMM to replace the similarity estimation method of node mobility in our previous work PMCV and construct a new video sharing solution (called ＂MPMCV＂) in VANET. Simulation results show how M-PMCV achieves much better performance in comparison with the original solution PMCV.