In this paper we present a new mobility model for mobile network simulations and evaluate user distributions in a real topology using different parameters.
SUMMARYThe Random Direction (RD) and the Random Waypoint (RWP) mobility models have been discussed extensively in the literature. The focus of this discussion was the stationary distribution of users and speeds of users. In this paper, we extend the investigation to street-based variants of the two mobility models. We compare statistical properties including user distribution, sojourn times in hotspots and link durations in ad hoc networks. In order to motivate the use of street-based mobility models, we also perform a comparison with the basic models. Finally, we present our approach to achieve identical user distributions in both streetbased models and investigate the remaining differences, which prove to be negligible for wireless network simulations.
In this study we evaluate the ability of isolated Wireless LAN access points to serve fast moving mobile users under a suitable mobility model for vehicular movement. For this purpose the Random Waypoint City Model is used with five different city maps. The focus lies on the difference in performance resulting from the distinct stationary user distributions of four real city maps and an artificial Manhattan grid. Detailed packet-based simulations using the network simulator ns-2 show the importance of assessing the stationary user distribution before fixing the locations of the wireless access points. The compared factors that are influencing the system's performance include mean data rates, number of users in hotspots as well as mean sojourn-times in hotspots and times in-between coverage. Although the resulting performance of the scenarios is quite diverse, the use of Wireless LAN as an access technology for sporadic vehicular Internet usage is possible in all of them.
In order to be able to simulate wireless networks in a non-uniform environment resulting in specific user distributions, it is essential to predict and control the stationary user distribution of the employed mobility model. We propose a framework to calculate turning probabilities at intersections of a graph-based mobility model to realize an arbitrary stationary user distribution on a street network. In this way, the user distribution can be fitted to a measured distribution. Our framework is able to cope with different average velocities on streets as well as different pausing probabilities at intersections. We evaluate its scope with the help of an example distribution on a European city center map featuring different street types and a variety of different crossings.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.