1 st International Conference of Recent Trends in Information and Communication Technologies Vehicular Road Side Backbone Network with Multiprotocol Label Switching Kashif Naseer Qureshi *, Abdul Hanan Abdullah, Raja Waseem Anwar Department of Communication, Faculty of Computing, UniversitiTeknologi Malaysia, Skudai, Johor, Malaysia Abstract Vehicular Ad hoc Networks (VANET) is an emerging type of data communication between vehicles in Ad hoc and infrastructure modes. In recent years the VANET applications have been gained attention and popularity in transportation sector and implemented for vehicle safety, traffic control and for infotainment services. The high performance and quality of service are always considered on the top priority. To improve quality of services in network various solutions have been proposed and applied in layers 2 and 3 in network layer model. In this paper, we proposed Multiprotocol Label Switching (MPLS) based road side backbone for vehicular Ad hoc networks. Through this switching forwarding method the packets across the network and the contents of labels attached with the IP packets and based on layer 2 and layer 3. The proposed model use for forwarding the data and work as a backbone networks to improve network efficiency and quality. The results showed the better performance of network in the terms of delay, throughput, and packet loss in highway areas. Keywords. Multiprotocol Label Switching MPLS, Vehicular Ad hoc Network VANET, Quality of service (QoS). 1. Introduction In recent years, the vehicular industry has made convenient and affordable vehicles to provide safety and pleasant travelling. The Quality of Service (QoS) is main priority during packet forwarding between source and destination. The vehicular networks require high bandwidth, successful packet delivery and quality of service. Various different types of applications need high quality for delivering the services such as voice over IP (VoIP), video conferencing (VC) [1], multimedia services, video Security etc. The quality of service and delivery in time are the main and significant requirements of these application and services. Researchers have done considerable work for analyze and improve QoS performance in wireless and wired communication. In IP networks, some quality of services protocols have been applied for reducing delay, latency, and jitter problems such as Intserv (Integrated services), Diffserv (Differentiated services) and Multiprotocol label switching (MPLS). The MPLS technology allows a certain degree of quality in packet forwarding. In this paper, we implement MPLS as a backbone in vehicular *Corresponding author:kashifnq@gmail.com IRICT 2014 Proceeding 12 th -14 th September, 2014, Universiti Teknologi Malaysia, Johor, Malaysia
Kashif Naseer Qureshi et. al. /IRICT (2014) 204-210 205 Ad hoc networks for increasing the quality of service (QoS). The rest of the paper is organized as follows. Section 2 gives a brief overview of Multiprotocol label switching and in section 3 illustrates the vehicular Ad hoc network and its architecture. In section 4 we discuss the proposed MPLS based model and the section 5 is base on simulation setup and 6 demonstrates the simulation results and performance of proposed model. In last the conclusion section presents the concise output of paper and give future direction for researchers. 2. Multiprotocol label switching Almost all IT companies always search for an effective and appropriate solution for their wide area networks. They have been purchased the frame relay or ATM (Asynchronous Transfer Mode) leased lines. The VPN (Virtual private network) is a layer 2 technology and provides maximum security and a tunneled route for data traffic over the Internet. These technologies provide security against hackers and protect from intruders, which are threats for backbone networks. These types of networks have been faced the scalability problem [2]. The MPLS (Multiprotocol label switching) was introduced to overcome these issues [3], because it is an efficient and effective technique for forwarding the packets across the network. It is using the contents of the labels attached to the IP packets [4]. MPLS (Multiprotocol label switching) is based on layer 2 (data link) and 3 (network layer) and called as a layer 2.5 technology. This technology is similar to the virtual circuit concept like ATM (Asynchronous Transfer Mode) to find the next hop for packets searching in the routing table and take time but through MPLS, routers forwards the packets by looking at the label of a packet. The Attached labels have functions routing and forwarding like layer 3 and perform separately like layer 2 switching functions The MPLS runs with any layer 2 technologies and one of the main feature such as frame relay, ATM or Ethernet. There is another noticeable feature of MPLS is traffic engineering. The two protocols Routing Information Protocol (RIP) [5] and Open Shortest Path First (OSPF) [6] are usually rely on protocol design and some metrics for search the shortest path and ignore delay, packet loss, throughput, jitter and congestion. In case of MPLS, the determinations of the best route prepare service requirement of packets and the shortest path" is not important. Figure 1. Domain Architecture with Multiprotocol label switching (MPLS) 3. Vehicular Ad hoc Networks The vehicular Ad hoc network is a new emerging technology where vehicles nodes exchange and disseminate the data packets to provide services and comfort. Vehicular Ad-Hoc network (VANET) has some unique characteristics like high mobility, dynamic topology etc. In VANET the nodes are operational with the
Kashif Naseer Qureshi et. al. /IRICT (2014) 204-210 206 help of on- board units, sensors, and with roadside infrastructure. The different types of applications working in network for public-safety and infotainment such as accident detection-map, internet services etc. The two types of communication present in VANET: vehicle-to-vehicle (V2V) and vehicle-to-roadside (V2R) and hybrid communication The V2V communication is between vehicles and the V2R architecture is based on roadside unit (RSU). and hybrid is a mixture of both communications The road side communication needs a backbone network for connect the one unit with other roadside units and for internet services [7]. The complete design of VANET backbone setup shows below in Figure 2. The applications of VANET categorized into safety and infotainment applications, such as warning and accident detection applications fall in safety type and internet services fall in infotainment category. The safety application need high accuracy and all the packets should be reached in time without any delay if the packets are not in time so the result of that is life loss or other accidents. On the other hand the infotainment applications need quality of services. For efficient communication need reliable and fast backbone architecture to provide quality of services (QoS). We proposed a MPLS based backbone architecture to improve QoS in VANET [8-9]. Figure 2. MPLS based backbone architecture of VANET Basically there are two types of communication present in VANET wired and wireless depends on the environment. In wired network, the information is carrying through cables such as servers attached with roadside units or the main station. Whereas in the wireless network the data is sent from one vehicle to another vehicle and vehicle to infrastructure through, wireless medium. Wireless network is further divided into two groups: infrastructure wireless network and Ad hoc network. 4. MPLS Based Backbone Architecture The Multiprotocol Label Switching technology introduced in late 90's for improve quality of service and design for high-speed optical backbone networks. Recently
Kashif Naseer Qureshi et. al. /IRICT (2014) 204-210 207 this technology provides a better high processing power and most essential ingredient for wireless domain. The MPLS technology particularly takes extraordinary measures for quick processing of layer 2 TCP/IP protocol stack header and decreases the end-to-end delay. Through MPLS in vehicular Ad hoc networks the many Quality of Services (QoS) metrics will improve and like round trip delays, packet loss, fault tolerant paths, and structural management services. Due to high mobility of vehicles, utilizing MPLS in vehicle to vehicle communication may not have a positive effect, so we implement this technology in backbone network or in base stations, which are connected with a wired network. The vehicles send their data through the base station for further analyzing and processing to the wired infrastructure. After usage of MPLS, the QoS metrics will be more efficient and gain higher results. Below we test the proposed architecture and explain simulation environment in detail. 5. Simulation Setup First we set the mobility model to test our proposed backbone network. In VANET the mobility models are divided into two types microscopic and macroscopic. The macroscopic, models the motion constrained are included such as roads, street, crossroads and traffic lights, and distribution of vehicles, generation of vehicular traffic density, flow also fall in this category. The movement of each vehicle node and vehicle behavior are included in microscopic type. The various types of mobility models proposed to generate the mobility, some of them shows in below Table 1. Table 1. Different types of mobility models and their usage S/No Mobility Models Usage 1 Manhattan Mobility Model (MHM) For vehicle node movement 3 Random Waypoint Model (RWM) For research Community 2 Freeway Mobility Model (FWM) For freeway motion behavior 4 Reference Point Group Mobility Model (RPGM) For military battlefield communication We use Manhattan Mobility Model (MHM) for mobility movement pattern for simulation, where streets defined by maps and use for urban environment. Different types of routing protocols proposed for VANET like topology based, position based and cluster based, etc. We set topology based protocol to test the environment because it is best for shortest path finding [10] and further these topology based protocols divided into reactive and proactive.. In proactive protocols every node keep information of all connected node in the routing table and create a high overhead for network. In reactive protocols, routes are discovered and maintained on demand. The Ad-hoc On Demand Distance (AODV) protocol is an example of reactive routing protocol [11]. AODV reduces many problems of proactive protocols by supporting a large number of networks and also reducing flooding of messages in the network. Using MPLS in
Kashif Naseer Qureshi et. al. /IRICT (2014) 204-210 208 vehicular Ad hoc network as a backbone service for enhancing the network efficiency and performance. Vehicle to infrastructure network where vehicle nodes are connected with the base station and that base station have own domain of service. The base station is connected with a wired network. The vehicle communicate with each other and with base station and all data will forward to MPLS backbone wired network and below simulation results shows the better performance of proposed backbone network. We use more appropriate and authentic simulation NS2 because it is a event driven. For mobility generation we use SUMO with NS2 and set Manhattan mobility model, where vehicles are allowed to move along to the grid of vertical and horizontal streets in the map. In simulation we set urban area and with multiple roads and traffic lights and intersections. In our scenario, we use 4 groups of vehicles and each group have 4 vehicle. The simulation parameters are below in Table 2. 6. Results Parameters Table 2. Simulation Parameters Values Network Area 500 * 600 M Radio Range 150 m Channel Wireless Traffic type CBR Routing Protocol AODV Number of Vehicles 16 Base stations 10 Speed of vehicles node 30 km/h Packet size 800 byte Transport protocol UDP The backbone networks and base stations are connected with wire and through base stations, communication is possible between vehicles and the wired MPLS based domain. The each base station has its own addressing domain in 3 levels. On the other hand the vehicles node are communicating with each other, and send the packets to base stations and further base station send the packets to server with the help of wired MPLS based network. We improving this communication and getting efficient QoS, the MPLS mechanism is fast and reliable compare to previous mechanisms. Through this technique, we achieve better QoS in term of end-to-end delay, throughput, and packet loss [12]. MPLS provides faster forwarding compare to IP nodes. The methodology of your paper should be described in this section. We compare vehicle-to-vehicle AODV based communication with Vehicle-toinfrastructure MPLS based communication. The results show the better performance of vehicle -to-infrastructure network. Graphs show the overall performance of V2V routing and V2I MPLS enabled routing.
Kashif Naseer Qureshi et. al. /IRICT (2014) 204-210 209. Figure 3. Throughput Figure 4. Packet Loss Figure 5. End-to-End Delay The above graphs show the better performance of proposed MPLS based backbone network. The AODV protocol based network performance in Vehicle-tovehicle network. showed packet loss and drop and poor performance. On the other hand when we check the performance of same network with vehicle-toinfrastructure MPLS based network, so performance is better than previous. Using of MPLS in the backbone network we overcome the packet loss problem.
Kashif Naseer Qureshi et. al. /IRICT (2014) 204-210 210 Conclusion In Vehicular network, the mobility is very high, and packets loss problem still get the intension of researchers. In this paper, the usage of MPLS in VANET in urban city areas is an efficient technique to overcome the packet loss and end-to-end delay and enhance throughput and quality of services. Vehicles are connected with each other and with base stations and base stations are typically connected with wireless, but we proposed the base stations are connected with MPLS mechanism and they are wired based network. Through simulation, we proved that this method is better and efficient in some QoS parameters such as throughput, end-to-end delay, packet loss. ACKNOWLEDGMENT This research is supported by the Ministry of Education Malaysia (MOE) and in collaboration with Research Management Centre (RMC) Universiti Teknologi Malaysia (UTM). This paper is funded by the GUP Grant (vote Q.J130000.2528.06H00). References:- 1. Gospodinov, M. The affects of different queuing disciplines over FTP, Video and VoIP Performance. in International Conference on Computer Systems and Technologies-CompSysTech 2004. 2004. 2. Ooms, D., B. Sales, W. Livens, A. Acharya, F. Griffoul, and F. Ansari. "Overview of IP multicast in a multi-protocol label switching (MPLS) environment." Informational Request for Comments 3353 (2002). 3. Daugherty, B. and C. Metz, Multiprotocol label switching and IP. Part I. MPLS VPNs over IP tunnels. Internet Computing, IEEE, 2005. 9(3): p. 68-72. 4. Alwayn, V., Advanced MPLS design and implementation. 2001: Cisco Press. 5. Hedrick, C.L., Routing information protocol. 1988. 6. Moy, John. "Open shortest path first (ospf) version 2." IETF: The Internet Engineering Taskforce RFC 2328 (1998). 7. Kiani, Hassan Shahzad, and Mirza Hamid Baig. Performance Evaluation of MANET Using MPLS. Diss. MS thesis, Bleking Institute of Technology, Sweden, 2010. 8. Fathy, M., S. GholamalitabarFirouzjaee, and K. Raahemifar, Improving QoS in VANET Using MPLS. Procedia Computer Science, 2012. 10: p. 1018-1025. 9. Vijayarangam, S. and S. Ganesan. QoS implementation for MPLS based wireless networks. in ASEE Conference. 2002. 10. Luo, Jie, Xinxing Gu, Tong Zhao, and Wei Yan. "A mobile infrastructure based VANET routing protocol in the urban environment." In Communications and Mobile Computing (CMC), 2010 International Conference on, vol. 3, pp. 432-437. IEEE, 2010. 11. Shastri, A., R. Dadhich, and R.C. Poonia, Performance Analysis Of On- Demand Routing Protocols For Vehicular Ad-Hoc Networks. International Journal of Wireless & Mobile Networks (IJWMN) Vol, 2011. 3: p. 103-111. 12. Nzouonta, J., et al., VANET routing on city roads using real-time vehicular traffic information. Vehicular Technology, IEEE Transactions on, 2009. 58(7): p. 3609-3626.