ENGINEERING SCIENCE AND TECHNOLOGY INTERNATIONAL RESEARCH JOURNAL, VOL.1, NO.3, SEP, 2017
Performance Evaluation of Routing Protocols in VANET on Highway: Police Cars Communication Scenario
16
ISSN (e) 2520-7393 ISSN (p) 2521-5027 Received on 23rd Aug, 2017 Revised on 21st Sept, 2017 www.estirj.com
Irshad Rahim Memon1, Wajiha Shah2, Wanod Kumar2, Khalid Hussain Mohammadani3, Sufyan Ali Memon4 1,2
Department of Electronic Engineering, Mehran University of Engineering & Technology, Jamshoro, Pakistan. Faculty of Engineering, Science and Technology, Isra University Hyderabad, Pakistan.
3,4
Abstract: Highways lead to main entrance of the city. They are occupied with different types of vehicles and need to be monitored in terms of accidents, traffic jams and any type of unwanted activity through an authority such as national highway police cars. Therefore, we need to enhance our security procedures during real time audio communication between highway police cars. This requires an enhanced and reliable system for voice communication established between police cars. Highway police cars mostly use traditional systems and face serious problems in voice communication. Call losses occur due to an unreliable communication system which further leads to difficulties in terms of finding or catching the suspicious person or look for an unwanted activity. One of the best solutions to this problem is to use the new dedicated technology for vehicular communication that is called VANET. Moreover, for reliable wireless communication among fast moving police cars, we need stable and efficient routing protocols. Therefore, in this paper, we assess the execution of two fundamental routing protocols, Dynamic source routing protocol and Adhoc on Demand Distance Vector routing protocol, for a temporary network of vehicles in a squad car correspondence situation. Execution of both protocols is being assessed on basis of quality of service parameters namely packet delivery ratio, average delay, throughput and normalized routing load.
Keywords: Vehicular ad-hoc network, Routing Protocols, Quality of Service (QoS), Voice Communication.
1. Introduction
V
ehicular Ad-hoc Network (VANET) is a multi-hop specially appointed system that comprises of different vehicles (hubs) that communicate with each other through remote connections utilizing an inferred form of IEEE 802.11 convention, IEEE 802.11p. It is a form of the multihop wireless ad-hoc network which is all around propelled by financial estimation of cutting edge intelligent transportation systems (ITS) whose main aim is to reduce traffic congestions, street mishaps and so on. VANET can bolster an assortment of uses, for example, street deterrent cautioning, vehicles impact shirking, well-being message dispersals, activity data administrations etc. [1, 2]. VANET is comprised of both vehicle-to-vehicle (V2V) communication and vehicle-to-roadside (V2R) communication which are important requirements for advanced intelligent transportation systems (ITS). V2R communication is an infrastructure-based wireless communication scenario, like cellular networks, WiMAX, and Wi-Fi somewhere intelligent vehicular nodes communicate with the roadside base station or access point. A roadside unit (RSU) cannot always provide network coverage required by advanced ITS system, therefore need to achieve a continuous connectivity, V2V communication is needed that helps in producing the best network coverage and the performance. V2V communication is a purely multi-bounce ad-hoc network arrangement as mobile ad hoc Network (MANET). In V2V communication scenario, vehicles traveling on street progressively self-sort out themselves by exploiting their wireless link interfaces. The Corresponding author or Email address:
[email protected]
reactive protocol works on demand of network. A reactive protocol generally uses a flooded discovery mechanism to find routes for data to be sent and after discovering the data is communicated to other users through established routes. It does not depend on any table data. So, its main advantage is that it consumes less power as it works on demand and is proven to the best for real-time communication. However, its disadvantage is that it takes more time to find a route as compared to proactive protocol because if routes breakup then it does flooding mechanism again and find routes in the same fashion to establish a link. It does not have any information of previous nodes in his queue memory which has already made routes. It starts making routes with new storage queue while neglecting previous ones [3,4]. A classification of routing protocols is shown in Figure 1.
Figure. 1: Classification of Routing Protocols Furthermore, this paper has been fashioned in following sections: Section 2 details about the related work. Section 3 discusses the system design while Performance evaluation
I.R. MEMON et.al: PERFORMANCE EVALUATION OF ROUTING PROTOCOLS IN VANET ON HIGHWAY……
is carried out in Section 4. Finally, this paper concludes in Section 5 followed by references.
2. Related Work In [5] authors have proposed an updated routing method in VANETs. Their proposed routing technique is based on algorithm namely junction based multipath source routing where geographical VANET multipath protocol has been created. The results reveal that choosing multiple paths is very much beneficial for VANETs. In their work, the execution of JMSR has been dissected utilizing NS2 simulator. A convenient VANET convention PFQ-AODV has been proposed in [6] which receives itself to the best course utilizing fluffy requirement Q-learning calculation in view of AODV system. This convention decides either remote connection is great while not considering numerous measurements. The proposed protocol gives data of other vehicle development through its neighbor vehicle regardless of the possibility that position data is inaccessible. It is additionally free of lower layers that make this proposed convention an adaptable, versatile and practical answer for directing in VANETs. The research challenges and opportunities related to real-time video streaming in vehicles have been discussed in [7] which can help to enhance street wellbeing and efficiency. For agreeable insightful transport frameworks (C-ITS),5.9 GHz frequency band has been allocated while for vehicles 802.11p has been selected as communication technology. C-ITS uses any type of wireless technology with respect to its requirement. IEEE 802.11p is a shortrange wireless communication which provides a great possibility of transferring information between vehicles without any detour around a base-station/access point by utilizing ad hoc network. In addition, there is no need of any framework establishment to run C-ITS applications. The performance evaluation of VANETs in sparsely populated traffic with the help of real vehicular traces has been carried out in [8]. The vehicular movement is incorporated using IDM-IM mobility model. Various protocols such as LAR1, AODV and DSR have been taken into account. The performance of all protocols is compared through packet delivery ratio and it has been found that LAR1 gives 100% PDR by looking at its outcomes with other protocols. The performance analysis is carried out in Glomosim 2.03 simulator. Authors in [9] have examined the execution of AODV, DSR and AODMV with blackholes attack (in terms of nodes) in VANETs. These black holes absorb all data packets. It has been found that DSR achieves the highest performance in terms of all parameters compared to other protocols. A detailed survey about VANETs has been carried out in [10] where authors have discussed protocols used in VANETs. The V2V communication has been used in this work. This work has discussed main features and objectives of VANETs along with its goals. A road scenario is taken in [11]. Authors evaluated the four-mobile ad-hoc routing protocols to simulate the real-time traffic. They used various call stress load to examine the routing protocols. They found as poor to DSR. AODV proved with good performance in their scenario. Performance analysis along with a comparison of three location-based services RLS, HLS and GLS with GPSR
17
routing protocol in a city environment has been carried out in [12]. The comparison has been performed by using different vehicles in a realistic environment with all three types of services present in them. All simulations were carried out in NS2 simulator. Performance Evaluation of AODV, AOMDV, DSR and DSDV of a real city scenario is observed in [13] with varying vehicle density using NS2 and it has been observed that DSR is efficient among all. Authors in [14] have examined the execution of AODV, DSR and DSDV observed in many types of traffic scenario. Main performance metric includes average energy consumption while other QoS parameters were also analyzed. Results show that AODV and DSR give efficient results as compared to other. All simulations were carried out in NS2 simulator. In this work, we assess the execution of two protocols in terms of QoS parameters such as packet delivery ratio (PDR), average delay, throughput and normalized routing load in a highway police car communication scenario on highways.
3. System Design This section has following subsections in which we discuss complete system design in term of realistic highway scenario, NS2 software, and simulation parameters. 3.1. Realistic Highway Scenario A realistic scenario of 10 km of the national highway of Pakistan is taken in to account shown in Figure 2 using Google map. Vehicles move with varying speeds on this road.
10Km Road
Figure. 2: A Scenario of National Highway of Pakistan captured via google map. Many highway police vehicles are put on duty here that basically watch speedy traffic along with suspicious/criminal activities. These police vehicles also help peoples if they are facing problems in terms of accidents, mechanical faults along with security issues. For these police vehicles to work smoothly, a reliable and efficient wireless communication based on VANET is essential instead of traditional communication. In this paper, we have considered highway of a 10km stretch. This road length can be increased depending on locations of vehicles as VANET is infrastructure less network.
Copyright ©2017 ESTIRJ-VOL.1, NO.3 (16-22)
I.R. MEMON et.al: PERFORMANCE EVALUATION OF ROUTING PROTOCOLS IN VANET ON HIGHWAY……
3.2.
NS2 Software
Figure 3 depicts the main flowchart of NS2 simulator that is mainly a GUI interface of NS2 simulator.
18
different speeds. In each scenario, four types of call durations of 50,100,150 and 200 seconds are considered. Two protocols namely AODV and DSR are analyzed with their four popular QoS parameters. A complete list of simulation parameters is provided in Table I.
4. Results and Discussion In this part, we will discuss the scenarios implemented in NS2 and then we evaluate the performance in terms of QoS parameters. 4.1.
Figure. 3: NS-2 Simulator Structure NS2 is a discrete level simulator that is very useful in providing research-based knowledge of various network topologies. It provides us ease to access any type of network in this environment that can be visualized and varied in terms of network parameters. TCP, UDP and other unicast and multicast routing protocol are very much supported by NS2. It is written with a combination of two languages namely C++ and script language. Script language called as OTCL that defines nodes, traffic, links, agents, sources and protocol used. This script language helps NS2 during simulation to define whole network environment. The Results are saved in .trace file. Trace file maintains series of events that occur during simulation. The file with .nam extension helps to visualize the whole network.
Scenarios of Highway in NS2 Animator
Figures 4, 5 and 6 depict highway scenarios of VANET of 50, 90, 170 Vehicles in NS2. Highway with the length of 10km is considered in this work. In these scenarios, two types of vehicles are shown with help of two colors. Blue nodes represent police vehicles while black nodes represent remaining common vehicles. These all nodes are mobile while police cars (nodes) can communicate with each other through common vehicles by using them as intermediate nodes.
Table.1. Simulation Parameters Parameters No nodes Area No scenarios Speed
Values 50,90,170 11000 x 2000 meters 3 Police=15m/s(54 Km/h) Cars=35m/s(120 km/h) Buses=28m/s(100 km/h) Truck=17m/s(61 km/h) Motorbike=18m/s(65 km/h)
Call length Packets size Rate Type of traffic / Routing agent Routing protocols Mobility model Type of channel MAC protocol Radio propagation type Interface queue type
50,100,150,200 sec 160 Byte 8Kb CBR/ UDP AODV, DSR Road model for VANET Wireless IEEE 802.11 Two ray ground AODV _Drop Tail Pri Queue DSR_CMUPriQueue
Queue Size Antenna model Frequency Total no of simulations Simulation time
50 Omni antenna 2.472 GHz 24 600 Seconds
Figure. 4: Highway Scenario with 50 Nodes
Figure. 5: Highway Scenario with 90 Nodes 3.3.
Simulation Parameters
In this research paper, three scenarios have been built on NS2 consisting of 50, 90 and 170 vehicles moving at Copyright ©2017 ESTIRJ-VOL.1, NO.3 (16-22)
I.R. MEMON et.al: PERFORMANCE EVALUATION OF ROUTING PROTOCOLS IN VANET ON HIGHWAY……
19
Figure 9 shows PDR of scenario 3 (consisting of 170 vehicles). AODV protocol gives a 90% average PDR in all real-time call durations. As number of vehicles are more in this scenario the congestion of packets transmission is caused which leads to packets drop. Even in this situation, the AODV protocol performs better than DSR protocol. The results also reveal that the performance of DSR protocol degrades with an increase in call duration.
Figure. 6: Highway Scenario with 170 Nodes 4.2. Simulations Results Figure. 8: PDR with 90 vehicles Three Scenarios are created in NS2 consisting of 50, 90, and 170 vehicles respectively moving with different speeds. In each scenario, four types of call duration 50, 100,150 and 200 (sec) have been considered. The results are discussed in terms of four basic QoS discussed above. Figure 7 shows PDR of Scenario1 (consisting 50 vehicles). For call duration of 50 sec AODV and DSR, both give almost 25% PDR. However, for a call duration of 150 sec AODV gives better PDR (20%) as compared to that of DSR which is 15%. PDR of both AODV and DSR protocols decreases rapidly as call duration increases; however, the PDR protocol of AODV is efficient than the PDR of DSR protocol. In this scenario, both protocols find less number of routes from source to destination due to low node density. If nodes are in the range of each other, the PDR is good; however, when the range increases the PDR gradually decreases for both AODV and DSR protocols. PDR results in this scenario are not good for all call durations.
Figure. 9: PDR with 170 vehicles Figure10 shows throughput of scenario1. During realtime call durations of 50 and 100 secs, the throughput of AODV and DSR is almost same but as the call duration of 150 sec is considered the AODV protocol achieves better throughput compared to DSR protocol.
Figure. 7: PDR with 50 vehicles Figure 8 shows PDR of scenario 2 (consisting of 90 vehicles). In this scenario, AODV protocol attains 90% PDR and DSR protocol gives an average 70% PDR. Because of the increased number of vehicles in this scenario, the PDR result is good for both protocols. In spite of congestion caused by increased number of nodes, the AODV protocol performs better than DSR protocol in terms of PDR.
Figure. 10: Throughput with 50 vehicles Figure 11 depicts throughput of scenario 2. As compared to scenario 1, the scenario 2 shows better results in terms of throughput. Here throughput of both AODV and DSR protocols gradually increases with increase in real time call durations. The throughput of AODV as compared to DSR protocol is more efficient.
Copyright ©2017 ESTIRJ-VOL.1, NO.3 (16-22)
I.R. MEMON et.al: PERFORMANCE EVALUATION OF ROUTING PROTOCOLS IN VANET ON HIGHWAY……
20
AODV protocol is less than or equal to 0.5 sec for all call durations.
Figure. 11: Throughput with 90 vehicles Figure. 14: Average delay with 90 vehicles Figure 15 illustrates average delay of scenario 3. In this scenario, the increased congestion cause more delay. The results are worse for DSR protocol with long call duration. The delay for AODV protocol is very small because of a greater number of routes available. Figure 16 depicts the NRL of scenario 1. In this scenario number of vehicles, traveling at different speeds, is a less hence small number of control packets are generated. DSR protocol results in higher NRL with an increase of call duration compared to AODV protocol. Figure. 12: Throughput with 170 vehicles Figure 12 depicts throughput of scenario 3. In This Scenario vehicles are increased to 170. The throughput of AODV increases with increase in real time call durations; however, the throughput of DSR protocol attains low value during the whole time of simulation. Figure 13 depicts average delay of scenario1. In the present scenario, the average delay of both protocols increases with increment in call duration. The average delay of DSR protocol is comparatively higher than that of AODV protocol for making real-time communication between vehicles.
Figure 17 illustrates NRL of scenario 2. Compared to scenario 1, the NRL is higher due to the increased number of communicating nodes. Moreover, their speed variations increase congestion of voice packets which results in more to be dropped. In this scenario, AODV protocol performs better than DSR protocol for all call durations.
Figure. 15: Average delay with 170 vehicles
Figure. 13: Average delay with 50 vehicles Figure 14 illustrates average delay of scenario 2. In this scenario, increased number of vehicles means possible communication routes are present. The delay for DSR protocol is always higher than AODV protocol. Delay for both protocols is low compared to scenario 1. The delay of Fig. 16: NRL with 50 vehicles Copyright ©2017 ESTIRJ-VOL.1, NO.3 (16-22)
I.R. MEMON et.al: PERFORMANCE EVALUATION OF ROUTING PROTOCOLS IN VANET ON HIGHWAY……
21
network." Electrical & Computer Engineering (ICECE), 2012 7th International Conference on. IEEE, 2012. [4] A. Shaikh, D. Vasan, and H. Mohammadani, Khalid, “Performance Analysis of MANET Routing Protocols – A Comparative Study,” vol. 83, no. 7, pp. 1–29, 2013. [5] Vinel, Alexey, and Magnus Jonsson. "Pre-study: Real-time video streaming over VANETs for increased road traffic efficiency and safety." (2013). [6] Sermpezis, Pavlos, Georgios Koltsidas, and Fotini-Niovi Pavlidou. "Investigating a junction-based multipath source routing algorithm for VANETs." IEEE Communications letters 17.3 (2013): 600-603.
Figure. 17: NRL with 90 vehicles
[7] Wu, Celimuge, Satoshi Ohzahata, and Toshihiko Kato. "Flexible, portable, and practicable solution for routing in VANETs: a fuzzy constraint Q-learning approach." IEEE Transactions on Vehicular Technology 62.9 (2013): 42514263. [8] Das, Sanjoy, et al. "Performance analysis of routing protocols for VANETs with real vehicular traces." Intelligent Computing, Networking, and Informatics. Springer, New Delhi, 2014. 45-56. [9] Sonia, Sonia, and Padmavati Padmavati. "Performance analysis of Black Hole Attack on Vanet's Reactive Routing Protocols." International Journal of Computer Applications 73.9 (2013): 22-26.
Figure. 18: NRL with 170 vehicles Figure 18 shows NRL of scenario 3. This scenario is comprised of 170 vehicles. NRL for this scenario is much higher compared to previous two scenarios because of the increased number of vehicles. NRL for AODV protocol remains low for all call duration, whereas, the NRL of DSR protocol increases with increase in call duration.
5. Conclusion In this paper, we have evaluated the performance of two routing protocols (AODV and DSR) for voice communication among police cars on a highway. The performance has been evaluated in terms of four QOS parameters, PDR, average delay, throughput and NRL. From results, it has been noticed that AODV protocol is better in terms of providing real-time communication between police cars in the modeled highway scenario. The results also reveal that the performance of DSR protocol deteriorates with increase in node density and call duration.
References [1] Conti, Marco, and Silvia Giordano. "Mobile ad hoc networking: milestones, challenges, and new research directions." IEEE Communications Magazine 52.1 (2014): 85-96. [2] Chang, Che-Yu, Hsu-Chun Yen, and Der-Jiunn Deng. "V2V QoS guaranteed channel access in IEEE 802.11 p VANETs." IEEE Transactions on Dependable and Secure Computing 13.1 (2016): 5-17.. [3] Rahman, Julia, Md Al Mehedi Hasan, and Md Khaled Ben Islam. "Comparative analysis the performance of AODV, DSDV and DSR routing protocols in wireless sensor
[10] Ranjan, Prabhakar, and Kamal Kant Ahirwar. "Comparative study of vanet and manet routing protocols." Proc. of the International Conference on Advanced Computing and Communication Technologies (ACCT 2011). 2011. [11] S.Abbasi, K.H. Mohammadani, Z.Hussain, JH.Awan, R. H.Shah. "Evolution of V2V Routing Protocols in Realistic Scenario of National Highway NH-5 Pakistan." Sci.Int.(Lahore) 28.5 (2016): 4711-4714. [12] Altayeb, Marwa, and Imad Mahgoub. "A survey of vehicular ad hoc networks routing protocols." International Journal of Innovation and Applied Studies 3.3 (2013): 829-846. [13] Ayaida, Marwane, et al. "A comparison of reactive, grid and hierarchical location-based services for vanets." Vehicular Technology Conference (VTC Fall), 2012 IEEE. IEEE, 2012. [14] Khairnar, Vaishali D., and Dr SN Pradhan. "Simulation Based: Study and Analysis of Routing Protocol in Vehicular Ad-hoc Network Environment." arXiv preprint arXiv:1403.6013 (2014).
About authors Irshad Rahim Memon is Professionally an Electronic Engineer with more than four-year experience in teaching at Undergraduate level. His area of interest includes wireless sensor network, image processing, control Engineering and VLSI design. He obtained his BE in Electronics Engineering from MUET, Jamshoro, Sindh, Pakistan in 2013. He also received M.E Degree in Electronic System Engineering from MUET Jamshoro, Hyderabad, Sindh Pakistan in 2016.Currently, he is serving as a Lecturer in the Department of Electrical Engineering at Isra University Hyderabad, Sindh, Pakistan Dr. Wajiha Shah is currently serving as Chairperson at the Department of Electronics Engineering MUET, Janshoro, Sindh, Pakistan.She has wide experience of teaching at Undergraduate and Postgraduate levels. She received Ph.D. Degree from Vienna University, Austria in 2010. Her area of interest centers in the field of Performances analysis of next-generation networks and Embedded System Design.
Copyright ©2017 ESTIRJ-VOL.1, NO.3 (16-22)
I.R. MEMON et.al: PERFORMANCE EVALUATION OF ROUTING PROTOCOLS IN VANET ON HIGHWAY……
Dr. Winod Kumar is currently serving as Associate Professor at the Department of Electronics Engineering MUET, Janshoro, Sindh, Pakistan. He has wide experience of teaching at Undergraduate and Postgraduate levels.He Obtained his MSc in Modern Digital and RF Wireless Communication and Ph.D. in Wireless Communication from University of Leeds, England in 2008 and 2012 respectively. His area of interest centers in the field of Wireless Communication, Wireless Sensor Networks, Digital Signal Processing and Digital Image Processing. Khalid Hussain Mohammadani: is a Computer Science Professional with over five-year experience in teaching networking, programming in C, Web development and other CS and Telecommunication courses. His area of interest includes wireless sensor network, mobile ad-hoc network, Localization and Network Coding. He obtained his BS in Telecommunication from University of Sindh, Jamshoro, Sindh, Pakistan in 2008. He also received M. Phil. Degrees in Computer Science from Isra University, Hyderabad, Sindh Pakistan in 2015. At present, he is pursuing Ph.D. in Computer science at Isra University Hyderabad, Sindh, Pakistan. Dr. Sufyan Ali Memon is professionally an Electronic Engineer with more than two-year experience in teaching at Undergraduate level.His area of interest includes Control Systems Design, Guidance Navigation & Control, Sensor Fusion and Data Fusion. He obtained his Ph.D. in Electronic System Engineering from Hanyang University, South Korea in 2016.Currently, he is serving as an Assistant Professor in Department of Electrical Engineering at Isra University Hyderabad, Sindh, Pakistan.
Copyright ©2017 ESTIRJ-VOL.1, NO.3 (16-22)
22
