Counter-based Traffic Management Scheme for Vehicular Networks

Transcription

1 Counter-based Traffic Management Scheme for Vehicular Networks Tarun Prakash, Ritu Tiwari Department of Information and Communication Technology ABV-Indian Institute of Information Technology and Management Gwalior, India ABSTRACT Vehicles traffic congestion on the road is reflected as delays while traveling. This congestion has a number of negative effects such as energy consumption, wastage of time and increased tailpipes emission of idling vehicles probably bad for our health. Vehicular congestion has become the serious problem and it is getting worse day by day as the growth of the vehicles significantly increased. In this paper, we proposed a novel counter approach to avoid such vehicular congestion on the road. We have also proposed a path selection algorithm that ensures best path suggestion to vehicles in terms of reduction in trip time and less fuel consumption during whole trip. The whole traffic management solution is combination of "stochastic turn" (i.e. vehicles choose a new direction at each intersection or any other way point) and path planning (i.e. origin and destination of the vehicle required in advance) that ensured by suggested path selection algorithm. In the later part of this paper simulation results prove the effectiveness of our traffic management scheme in terms of reducing traffic congestion on the road. In addition, this scheme utilizes best of the resources and characteristics of vehicular networks to provide less congested path prediction and also smoothed flow of traffic for vehicles in high density vehicular traffic conditions. Keywords Vehicular networks, Road congestion, Vehicular traffic, Vehicles-to-vehicles communication (V2V), Counter approach, Road Side Unit (RSU). I. INTRODUCTION Vehicular networks are a novel class of wireless networks that have emerged recently. These networks are attracting considerable attention from the researcher as well as the automotive industry. One of the main reasons for this is its potential contribution to informative applications. In this context, dedicated short-range communications (DSRC) system has emerged in North America, where 75 MHz of spectrum was approved by the U.S. FCC (Federal Communication Commission) in 2003 for such type of communication that mainly targets vehicular networks. On the other hand, the Car-to-Car Communication Consortium (C2C-CC) has been initiated in Europe by car manufacturers and automotive OEMs (original equipment manufacturers), with the main objective of increasing road traffic safety, efficiency and reducing road congestion by means of inter-vehicle communication [12]. With the recent advances in wireless technology and advancing trends in ad hoc network scenarios allow a number of deployment architectures for vehicular networks [2], in highway, rural, and city environments. Such architectures should allow communication among nearby vehicles and between vehicles and nearby fixed roadside equipment. Three alternatives include (i) a pure wireless vehicle-to-vehicle ad hoc network (V2V) allowing stand alone vehicular communication with no infrastructure support, (ii) a wired backbone with wireless last hops that can be seen as a WLAN-like vehicular networks, (iii) and a hybrid vehicle-to-road (V2R) architecture that does not rely on a fixed infrastructure in a constant manner, but can exploit it for improved performance and service access when it is available [8]. The basic architecture of vehicular network shown in fig. 1 including two primary components: Onboard Unit (OBU) and Road Side Unit (RSU). Figure 1 Architecture of vehicular network Vehicles traffic congestion is reflected as delays while traveling. This congestion has a number of negative effects such as energy consumption, wastage of time and increased tailpipes emission of idling vehicles probably bad for our health. This is one of the biggest problems faced especially in the developed cities around the countries. On top of the existing problems there is a worsening trend [4]. The growth in the number of vehicles on the road outpaces growth in road capacity worldwide. From 1982 to 2002, the total number of vehicles in the US 251

2 grew by 36% and vehicle miles travelled - by 72%,while road capacity increased by less than 5%. Between 1990 and 2004 the number of cars in the 25 EU member states rose by over 40% and continues to rise. Meanwhile the total length of motor ways in the EU grew by 28% between 1990 and 1998 and has remained roughly stagnant since then [13], [14]. This paper draws its inspiration from the observation of unique characteristics of vehicular networks and aim to utilize them to reduce road congestion and improve the performances of existing algorithms. In this context, We have proposed an counterbased solution to reduce road congestion in inter-vehicle communication scenarios and evaluate the requirements of corresponding mechanisms. We have also proposed an route selection algorithm that improves the quality of driving experience by optimizing the requested route of the vehicle. Goals of this paper- To propose an effective decentralize traffic management scheme so that small failures can be handled. To reduce overhead involved in existing traffic management approaches. To provide an effective solution to address journey time and fuel consumption issues. II. RELATED WORK With the recent progress on vehicular networks, it has become clear that the study of vehicular traffic management and its requirements is currently the hotspot in vehicular ad-hoc networks. In this context, many research groups exploring this area of research so that they can improve Quality of Driving Experience (QDE) by influencing vehicles routes. A number of Traffic Information Systems have been developed i.e. systems which gather traffic data and disseminate traffic information to users, so they can make better informed decisions regarding their route [5], [6], [16]. While these systems do keep drivers better informed about traffic conditions, there is no such information how the driver will interpret the information given. In [11], Abderrahmane Lekas and Moumena Chaqfeh present a vehicular communication system for road congestion detection and avoidance by disseminating and exploiting road information. In Japan, as one of traffic information dissemination systems, VICS (Vehicle Information and Communication System) are widely used [15]. VICS provides latest traffic information to cars via FM broadcast and various types of beacons. T. Shinkawa and et al present a system for discovering and disseminating traffic congestion information is proposed where vehicles build their own local traffic maps of speeds experienced on visited roads, and share them with other vehicles. In this context TraffCon, a novel Traffic Management System for WAVE [3] is introduced. In this TMS Vehicles act as client nodes and communicate with the server in two asynchronous modes i.e. information gathering and traffic management. Traffcon employs the Best Route Selection Algorithm in its route decision making process. Decision making process starts when a vehicle begins a journey by sending its origin and desired destination to the server. However, this algorithm lacks to address the overhead issue as server needs to calculate fitness function for each of the vehicles in the Vehicular Network. Our traffic management scheme provided better solution in this context. Because in this approach base station does not require to calculate fitness function until it is requested by the vehicle. So lesser times calculation of fitness function will be done as a result less burden on base station and also less utilization of the resources in this scheme. Our contributions to this paper are summarized as follows- We propose a counter-based traffic management scheme for avoiding congestion from the road and path selection algorithm to provide an optimal route to vehicles. The proposed Counter-based approach allows vehicles to take new direction at each intersectional area and Path selection algorithm suggests best route based on the fitness score. Our traffic management scheme is completely decentralized and capable of working in the small failure cases. Our scheme is best in terms of reducing journey time, efficiency of fuel consumption and lesser overhead in it. III. SYSTEM MODEL AND ASSUMPTIONS This section describes assumptions and the system model inspired from one of the most unique characteristics of vehicular networks i.e. (generally vehicles move along the road differ from mobile ad-hoc network where nodes can freely move in two dimensional area). Then, we discuss the concept of Counter-based solution to avoid congestion on the road. A. Assumptions The fundamental assumption in our vehicular networks scenario is, drivers can determine the next road to move. This assumption is reasonable considering advances in GPS technologies, navigation systems and other intelligent vehicular devices. A base station is installed at an intersection point and we also assume that vehicles are capable of sending their next road number to the base station. Vehicles that have wireless communication capability are also assumed to efficiently communicate with other vehicles in scenario. Base stations are assumed to have information about other base stations and can also exchange 252

3 information if needed in vehicular networks. The radio range of base stations and vehicles are assumed to be sufficient enough to cover the width of the road. Typically, the radio range is at least 500 meters. This ensures that vehicles can communicate with a base station when they pass by it. B. Modeling of Vehicular Network Scenario In this model, Vehicular Network is considered as a graph G = (I, E), where the nodes in I represent intersections and the edges in E represent roads which connect intersections. An intersection may be a real road junction where a base station is installed. Vehicles movement can be modeled as moving from an intersection to another intersection along the road edges. In this scheme base stations range is assumed to be greater than the road. A vehicle in the radio range can communicate with the base station directly. IV. TRAFFIC MANAGEMENT SCHEME Our traffic management scheme is based on the following subsections where we have described counterbased approach and path selection algorithm. A. Counter Approach This section describes proposed Counter approach, which is based on random trip planning of the vehicles. In this approach vehicle chooses its new direction at each intersection point that is known as "stochastic-turn". The basic concept behind this approach is to inform the requested vehicle about the dense condition on the next road. All vehicles that enter into or leave this ad hoc network must pass one of the base stations, and are required to exchange information with the base station in the proposed scheme. Figure 2 shows the base station identifiers such as BS_a, BS_b, BS_c, BS_d and roads are indexed as per system model consideration such as Road1, Road2 etc. shown in fig. 2. These Base stations maintain counter value i.e. (C1, C2) Shown in figure 2 for every vehicle passes nearby them. Base stations do this for each direction for each vehicle separately. Counter value decreased by the base station as vehicle goes out of the coverage area of that base station. Figure 2, shows the demonstration of proposed Counter- based mechanism. We assign an identifier for a base station and a number for each direction of the intersection where vehicle selects the next road R (1<R<N) at an intersection based on counter value i.e. the number of vehicles on that road. Counter C1 is maintained by base station and it is required when two lanes are available on the road and C2 is a passive counter which activates only in the situation of single lane scenario where base stations need to share the counter values maintained by them. B. Path Selection Algorithm The proposed path selection algorithm is employed by the base station in its decision making process and helps the driver to take the optimized route to his destination. This algorithm ensures the route optimization for each of the vehicles requesting for decision function. route selection algorithm which not just optimize the time and fuel efficiency but also reduces overhead comparative to other existing algorithms. Path selection algorithm is employed by our traffic management scheme in its decision making process. Vehicle starts its journey by sending its current location and next road number to the Base Station installed at the junction. 1. Vehicle sends its information and next road Number to the base station on which it wants to move. 2. Base station gives the Counter value (number of vehicles) to the requested vehicle. If C< threshold then vehicle will follow the requested route. 3. If C>threshold then route will we chosen by vehicle based on Cost Function. Cost function presented in the next section. The following two cases need to be considered in above algorithm. Case 1: Two Lane (Counter sharing not required) Case 2: Single Lane (Counter sharing required) Case 1: In a "Two lane" network scenario, sharing of the counter value does not required. Because vehicles in this scenario, are on a different lane so no question of getting in to road congestion because of vehicles which are opposite in direction. The requested base station will give counter value of requested road. If counter value C<threshold then vehicle will follow the path. Case 2: In a "Single lane" network scenario, counter sharing required. This is because when vehicles are in opposite direction creating road congestion as it is not the case with "Two Lane" scenario. Here requested vehicle needs whole counter value i.e. (C1 + C2). Figure 2 Counter approach demonstration 253

4 A. Fitness Function The fitness function presented in equation (1) is proposed to choose a vehicle s route so that journey time, congestion and fuel consumption and gas emissions are minimized. It consists of weighted cost components including T which encourages a routing solution with the minimum possible user journey time and F which makes sure fuel consumption and gas emissions are factored in [3]. R nv = w 1 T nv + w 2 F nv (1) We go for fitness function i.e. R nv only when counter values exceeding the threshold. Individual components of fitness function are weighted by (see equation (2)), for getting more emphasis on a particular outcome. The detailed description of each component T nv, F nv can be seen in [7]. The more important a cost component is considered to be to the solution, the smaller the weighting factor associated with it and therefore the stronger its contribution to the overall score R nv [4]. B. Simulation Results We have experimented with few selected vehicles in network scenario. It has been observed that as more and more vehicles are increasing on the road causing road congestion most of the times. As a result of this congestion, delay in journey time. One of the important goal of this paper is to reduce the average journey time of the vehicles. It can be clearly seen in fig. 3 that our suggested approach significantly reduces the average journey time of the vehicles. Following figure shows the experimental results of average journey time in two of the cases we considered. (2) V. EXPERIMENTS AND RESULTS A. Simulation Experiments In order to evaluate the efficiency of the suggested solution following experiment were performed. Firstly we set up mobility of the vehicles and created traffic scenario using traffic simulator using SUMO (Simulation of Urban Mobility). A dedicated vehicular traffic simulator is used to generate the realistic movement of vehicles in a trace file, containing vehicle related information such as time taken by the vehicles and location and etc. SUMO (Simulation of Urban Mobility) is a microscopic, space-continuous road traffic simulation package. Microscopic means that the simulator models the movement of every single vehicle on the street, mostly assuming that the behavior of the vehicle depends on both the vehicle s physical abilities to move and the driver s controlling behavior. The common behaviors, such as changing lanes or stop before the stop signs and traffic lights, are all supported in SUMO. We have also used MOVE (realistic mobility generator for vehicular networks) to get mobility traces of the vehicles [1], [9]. We performed our simulation in the following cases: Case (1): We consider this case as a default case for each vehicle in network scenario i.e. each of the vehicles follow their normal route as they have to travel in the normal traffic conditions. Case (2): In this case, each vehicle drives to its own destination according to the traffic management solution provided. Figure 3 Vehicles journey time optimization Figure 4 Comparative fuel consumption in congested and suggested route Fig. 4 illustrates comparative results of fuel consumption of the selected vehicles in the scenario. For these experiments, we experimented with vehicular scenario having 6 intersectional area and around 400 vehicles in vehicular network. In figure 4 the results for average fuel consumption in terms of (liter/km) are shown. It can be clearly notice that the significant reduction in terms of consumption of the fuel. 254

5 ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their valuable remarks and the time they have invested to review this paper and approved the work we have done. REFERENCES Figure 5 Distance traveled vs. Journey time vs. Fuel consumption in congested scenario [1] ealistic mobility generator for vehicular networks. [2] Bernardos, C.; Festag, A.; Maxemchuk, N.; Pinart, C. & Weib, C. Advances in vehicular communications networks [Guest Editorial] Network, IEEE, 2010, 24, 4-5 [3] Collins, K. & Muntean, G. A vehicle route management solution enabled by Wireless Vehicular Networks INFOCOM Workshops 2008, IEEE, 2008, 1-6 [4] Collins, K. & Muntean, G. Route-Based Vehicular Traffic Management for Wireless Access in Vehicular Environments Vehicular Technology Conference, VTC 2008-Fall. IEEE 68th, 2008, 1-5 Figure 6 Distance traveled vs. Journey time vs. Fuel consumption in suggested scenario Fig. 5 and Fig. 6 are showing comparative results in terms of traveled distance, journey time and fuel consumption of the vehicles. In both the figures, distance shown in meter and fuel consumption in liter/km and time in seconds. Fig. 5 shows the experimental results of default i.e. congested scenario in the network and fig. 6 shows in scenario suggested by "Counter-based Traffic Management Scheme". VI. CONCLUSION Vehicular networks provide new opportunities to develop innovative and advanced solutions for traffic monitoring and management. In this context, we have proposed Counter approach that deals with vehicular congestion on the road followed by path selection algorithm which insures optimal route selection by the vehicles. We also examine the performance of suggested solution in different scenarios. The results of this scheme are in shorter journey time and fuel efficiency of the vehicles meaning that more free time for commuters and as well as saving of fuel. Future scope of this paper includes the route suggestion based on pattern evaluation of certain area i.e. frequent dense situation in a particular area so that vehicles are better informed about realistic traffic condition on the road. Consideration of different parameters so that different social and environmental issues can be covered. [5] Dikaiakos, M.; Florides, A.; Nadeem, T. & Iftode, L. Location-aware services over vehicular ad-hoc networks using car-to-car communication Selected Areas in Communications, IEEE Journal on, IEEE, 2007, 25, [6] Dornbush, S. & Joshi, A. StreetSmart traffic: Discovering and disseminating automobile congestion using VANET's Vehicular Technology Conference, VTC2007-Spring. IEEE 65th, 2007, [7] Fries, R.; Hamlin, C.; Chowdhury, M.; Ma, Y. & Ozbay, K. Operational impacts of incident quick clearance legislation: a simulation analysis Journal of Advanced Transportation, John Wiley & Sons, Ltd., 2011, n/a-n/a [8] Jerbi, M. & Senouci, S. Characterizing Multi-Hop Communication in Vehicular Networks Wireless Communications and Networking Conference, CNC IEEE, 2008, [9] Karnadi, F.; Mo, Z. & Lan, K. Rapid generation of realistic mobility models for VANET Wireless Communications and Networking Conference, WCNC IEEE, 2005, [10] Krajzewicz, D.; Hertkorn, G.; R\össel, C. & Wagner, P. SUMO: Simulation of Urban Mobility 255

6 Proc. of the 4th Middle East Symposium on Simulation and Modelling, 2002, [11] Lakas, A. & Chaqfeh, M. A novel method for reducing road traffic congestion using vehicular communication Proceedings of the 6th International Wireless Communications and Mobile Computing Conference on ZZZ, 2010, [12] Peng, Y. & Chang, J. A novel mobility management scheme for integration of vehicular ad hoc networks and fixed IP networks Mobile Networks and Applications, Springer, 2010, 15, [13] Schrank, D. & Lomax, T. Urban mobility study Texas Transportation Institute, College Station, TX, 2003 [14] Shinkman, M.; Buchanan, M.; Unit, E. & Britain, G. Driving change: How policymakers are using road charging to tackle congestion The Economist Intelligence Unit, 2006 AUTHORS Tarun Prakash received the B.E. degree from Madhav Institute of Technology and Science (MITS), Gwalior, India in 2009 and M.Tech. degree in computer science and engineering with the specialization in Advanced Networks from ABV-Indian Institute of Information Technology and Management, Gwalior, India in His research interests are in the area of wireless networks and vehicular networks. Dr. Ritu Tiwari is an Assistant Professor in the IT Department of Indian Institute of Information Technology and Management Gwalior. Her field of research includes Biometrics, Artificial Neural Networks, Speech Signal Processing, Robotics and Soft Computing. She has published around 60 papers in various national and international journals/conferences. She has received Young Scientist Award from Chhattisgarh Council of Science & Technology and also received Gold Medal in her post graduation. [15] Tamura, K. & Hirayama, M. Toward realization of VICS-Vehicle Information and Communication System Vehicle Navigation and Information Systems Conference, 1993., Proceedings of the IEEE-IEE, 1993, [16] Zhang, X.; Hong, J.; Fan, S.; Wei, Z.; Cao, J. & Ren, Y. A novel real-time traffic information system based on wireless mesh networks Intelligent Transportation Systems Conference, ITSC IEEE, 2007,