1 VACaMobil: VANET Car Mobility Manager for OMNeT++ Miguel Báguena, Sergio M. Tornell, Álvaro Torres, Carlos T. Calafate, Juan-Carlos Cano, Pietro Manzoni Department of Computer Engineering, Universitat Politècnica de València Camino de Vera S/N 46022, Valencia, Spain. mibaal@upvnet.upv.es, sermarto@upv.es, atcortes@batousay.com, {calafate, jucano, pmanzoni}@disca.upv.es Abstract—Since the performance of communication protocols in vehicular networks highly depends of the mobility pattern, one of the most important issues in the simulation of this kind of protocols, is how to properly model the mobility of vehicles. In this paper we present VACaMobil, a VANET Car Mobility Manager for the OMNeT++ simulator which allows researchers to completely define vehicular mobility by setting the desired average number of vehicles and its upper and lower bounds . We compare VACaMobil against other common methods employed to generate vehicular mobility. Results clearly show the advantages of the VACaMobil tool when distributing vehicles in a real road map scenario, becoming the best simulation framework to evaluate the performance of different communication protocols and algorithms in VANET environments. Index Terms—Vehicular Networks, Mobility patterns, Simula- tion Tool, SUMO, TraCI. I. I NTRODUCTION The reproducibility of experiments is a major issue when evaluating smart communication protocols and algorithms, especially over Vehicular Ad-hoc NETworks (VANETs). In [6] Sommer et al. provide a complete review of the minimum set of parameters that should be identified in publications in order to allow other researchers to reproduce simulation exper- iments. They pointed out several key parameters, such as the simulated hardware, the network simulator, the scenario, and the road traffic simulator. However, regarding node mobility, there is another parameter that has been mostly ignored by the research community: the amount of traffic.. As other authors pointed out before, mobility models [10] and the chosen scenario [4], as well as the node density, heavily influence the final network performance. However, since mobility generators and road traffic simulators are often tough to configure, the simulated node density and distribution may depend on complex data that is usually not included in the published academic results, which compromises reproducibil- ity. In this paper we present VACaMobil (VANET Car Mobility manager), a mobility manager module for the OMNeT++ simulator which is the first, to the best of our knowledge, able to generate SUMO [1] driven nodes in a vehicular network while ensuring that certain user-defined parameters, such as the average, maximum, and minimum number of vehicles, are correctly achieved. This goal is useful for mid-length simulations, typically one hour, where vehicle density should remain stable. At the same time, since our solution is tightly coupled with SUMO through the TraCI interface, it is able to mimic real vehicle behavior. By running in parallel with SUMO, VACaMobil executes the following tasks: (i) manages when a new vehicle must be introduced in the network, (ii) assigns a random route from a predefined set to each vehicle, and (iii) determines which type of vehicle should be added. Given a specific road map, when using VACaMobil, researchers will be able to completely define the network mobility merely by defining the desired average number of vehicles and its standard deviation value (upper and lower bounds). Going a step further, our tool also aids researchers at selecting among the different types of vehicles defined in SUMO. This allows researchers to easily define road traffic simulations with heterogeneous vehicles, such as trucks, cars, or buses. The remaining sections of this paper are organized as fol- lows: In section II, we shortly introduce the different methods for generating VANET mobility patterns that the research community has been employing. In section III, VACaMobil is fully described. In section IV, we compare our proposal with other methods available in SUMO. Finally, in section V, we expose our conclusions and some future plans to improve VACaMobil. II. A REVIEW OF EXISTING MOBILITY GENERATORS FOR VANETS Before presenting the details of our proposal, we analyze some of the methods commonly used to obtain suitable mo- bility patterns in urban vehicular scenarios. We have analyzed several papers published during the last few years, most of them published in the Vehicular Networking Conference (VNC) and the Vehicular Technology Conference (VTC). Early approaches relied on too simple mobility models based merely on random mobility. Since these simple models do not represent vehicle mobility properly, other mobility models have been recently developed based on real world traces and artificial mobility models from the field of transportation and traffic science. In this section, we briefly describe the most relevant works.