VERGILIUS: A Scenario Generator for VANET Eugenio Giordano 1,2 , Enzo De Sena 3 , Giovanni Pau 1 , and Mario Gerla 1 1 University of California at Los Angeles, California, USA. 2 DEIS, WiLab - University of Bologna, Bologna, Italy. 3 King’s College London, CDSPR, London UK * ABSTRACT Vehicular networks are on the fast track to become a reality either through a car manufacturer that introduces a commu- nication device in the car electronics or through an aftermar- ket vendor such a GPS navigator or a in-vehicle entertain- ment system. This paper introduces VERGILIUS a nouvelle urban mo- bility and propagation toolbox designed to streamline the mobility trace generation and path loss computation in ve- hicular network studies. The aim of VERGILIUS is to en- able a whole new level of simulation through the introduc- tion of Urban Maps, finely tunable motion patterns, and de- tailed trace analysis. 1. INTRODUCTION Vehicular networks are clearly emerging and they will be- come a reality in the next few years either through after mar- ket devices such GPS navigators or a car manufacturer that integrates 802.11p or similar technology in the vehicle elec- tronic heart. A key factor for vehicular networks to succeed is the de- velopment of applications and protocols that fit the user needs. Studying the characteristics of the network is fundamental to understand the constraints and feasibility of protocols and services that can be run on, and even exploit, the networks formed. For example, are these networks frequently and dy- namically partitioning? What is the role of the urban prop- agation? What are the key factors that impact applications and protocols performance? Being able to study a multitude of detailed and realistic models is essential to explore the de- sign space and develop systems that are able to cope with a large number of very different scenarios in terms of network density, network load, and urban environment. We here argue for a streamlined ability to systemically study urban vehicular scenarios and we propose VERGILIUS 1 * This work was partially supported by Microsoft Research through its PhD Scholarship Program and by the National Science Foun- dation under Grant # 0520332-NSF-SWARMS and UC Discovery Grant # SEA private Sponsor STMICROLECTRONICS. Any opin- ions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily re- flect the views of the research sponsors 1 Publius Vergilius Maro was Dante’s guide to Hell and the greater a urban mobility toolbox for vehicular networks. VERGILIUS, features highly configurable mobility parameters and a num- ber of known and new scenario metrics to enable an in depth systematic study of the causal relationship between the dif- ferent scenario parameters and network performances [1]. VERGILIUS brings two important contributions to ve- hicular communications in urban environments. The first contribution is the macroscopic-level ”tunable” vehicle mo- tion pattern generator. It is well known that the urban traf- fic can be considered as the superposition of several ele- mentary components, say, commuting pattern (to and from work), ”shopping mall” pattern (say, from one mall to an- other in a random pattern, with pause time at each), lunch hour restaurant pattern (converging to a few popular malls), etc. The VERGILIUS motion pattern generator module can mimic any combination of traffic patterns, say commuting and ”shopping mall” traffic, etc. More precisely, by chang- ing a few parameters (e.g. number of paths used) we can increase traffic on local streets in a controlled manner. This tunable generator enables us to evaluate performance trends of various protocols (e.g. routing) under controlled traf- fic conditions. The output of the motion pattern generator is suitable to run microscopic-level traffic simulators such as SUMO[2] and CORSIM[3] thus allowing a effortless re- production of a large number of mobility traces for vehicu- lar networks. VERGILIUS automates the process of road map extraction and mobility trace creation, in contrast to todays best practices in vehicular network studies that re- quire a constant human attention. As a second contribu- tion, VERGILIUS integrates a trace analyzer that extends the IMPORTANT framework [1] to account for urban realis- tic proagation in the metric computation and further extends the set of network and connectivity metrics. The reminder of this paper is organized as follows: in sec- tion 2 we introduce VERGILIUS, our urban mobility and propagation toolbox, in section 3 we show how the different metrics and parameters affect the network performance in an urban scenario, and finally in section 4 we conclude our part of Purgatory in The “Divine Comedy.” We offer our tool to the research community as aid to explore the design space in vehicular ad hoc networks. The software can be downloaded at http:// vehicular.cs.ucla.edu/ 1