Delay and Hop Count Estimation of Directional-Location Aided Routing Protocol for Vehicular Ad-hoc Networks

Vehicular ad-hoc network (VANET) is an integral component of intelligent transportation systems. Due to continuous movement of vehicles, network formation and deformation is very frequent. Furthermore, for safety applications, we wish to create the communication without any delay or collision. Therefore, communication among the vehicles is even more challenging. D-LAR (Directional– Location Aided Routing) is one of the position based routing protocol which works efficiently in VANETs. We want to analyze the performance of D-LAR protocol in a city traffic scenario. To do a quality performance evaluation, extensive simulations has been done in realistic environment created with SUMOatraffic simulator and NS-2 network simulator. The performance of D-LAR protocol has been compared with LAR on various metrics like routing overhead, packet delivery ratio and delay. Index Terms Delay, D-LAR, Hop count, LAR, Routing protocol, Vehicular ad-hoc networks NOMENCLATURE DIR Directional Routing, D-LAR – Directional-Location Aided Routing, ITSIntelligent Transportation Systems, LAR Location Aided Routing, PDR – Packet Delivery Ratio, VANETs Vehicular Ad-hoc Networks, 1.0 INTRODUCTION Traffic and transport management systems are using the new generation information and communication technologies to have intelligent transportation systems (ITS). Their aim is to ease the citizen’s Life by providing them facilities like smart parking, intelligent traffic management systems,etc.VANETacts as a vital innovationfor ITS. It’s a kindof network where vehicles can communicate with eachother as per therequirements. To do this, intelligentvehicles that have wireless transceivers and computerized control modules are required. There are two kinds of communication in VANETs. One is vehicle-to-vehicle communication which is among vehicles only. Another type of communication is vehicle-toinfrastructure Department of Computer Science and Information Technology, Jaypee Institute of Information Technology, Noida, U.P., India. Email: [email protected],[email protected] AmbedkarInstitute of Advanced Communication Technologies and Research, Delhi, India. Email: [email protected] which is among the vehicles and roadsideinfrastructure units. Here,we are considering only vehicle-to-vehiclecommunication which is purely ad-hoc and completely infrastructure-less. The two vehicles can communicate when they lie in the communication range of each other. Vehicles movements are restricted as per the road-maps. Another important characteristic of VANETs is uneven distributions of vehicles on the roads. On top of that vehicles speed is irregular and generally it is high, therefore network formation and deformation is very frequent. Due to all these characteristics, doing the communication and maintaining it is very challenging. A number of researchers introduced a variety of routing protocols. These protocols have been grouped into two categories: position based and topology based routing protocols. Due to various characteristics of VANETs mentioned in the above paragraph, topology based routing protocols such as DSR, DSDV and AODV have poor adequacy and adaptability. Position based routing protocols works on the concept of greedy forwarding. Here, each node knows the positional coordinates of its neighboring nodes. After getting the positional information, source node selects the one as a next hop node who is geographically nearest to the destination. These protocols are also called as geographic or spatial aware routing protocols. In position based routing protocols, there is no requirement of establishment and maintenance of route. Also, these protocols do not require the knowledge about the complete network connectivity. These protocols utilize the spatial knowledge like lanes and maps of cities for routing decisions. Therefore, position based routing protocols performance is better than topology based protocols. However, position based routing protocols require extra information in terms of its own position, neighboring nodes and destination position. To get its own position, all the nodes utilize the Global Positioning System. To get the destination node positional information, source node takes the help of any location service. After getting the destination position, source node mention it in the packet’s destination address. Neighboring nodes gather the positional information of each other by periodic exchange of beacon or hello messages. GPSR[ HYPERLINK \l "BKa00" 1 ], A-STAR2]}, GEDIR[ HYPERLINK \l "Iva99" 3 ], LAR4]}, GSR[ HYPERLINK \l "CLo05" 5 ], D-LAR 6]}etc. are different examples of position based routing protocols. D-LAR (Directional-Location Aided Routing) [ HYPERLINK \l "Ram12" 6 ]proposed by Raw etal.is a position based routing protocol. This is amalgamation of two routing protocols, LAR and DIR. LAR (Location Aided Routing) protocol has been BIJIT BVICAM’s International Journal of Information Technology Copy Right © BIJIT –2016; July December, 2016; Vol. 8 No. 2; ISSN 0973 –5658 1039 proposed by Young and Vaidya4]}. In this protocol, they proposed the ideaof limited flooding for reducing the route discovery overhead. By getting the destination location and speed information, source node constructs an area, called as request zone which covers both source and destination. Now, route discovery is restricted to this constructed request zone. It helps in reducing the route discovery overhead. D-LAR protocol utilizes this concept for limiting the search area of route selection. Once the search area has been limited, now next-hop node would be selected from this search area. For the selection of next-hop node,D-LAR uses the concept of directional routing (DIR). D-LAR selects the next-hop node amongst the available neighboring nodes that has the minimum angular deviation from the connecting line of source and destination. Raw et al. [ HYPERLINK \l "Raw15" 7 ]proposed DLAR with a mental make-up that it achieves a remarkable performance in city traffic scenario because routing overhead would be reduced with the partial flooding concept of LAR protocol and directional greedy forwarding provides a stable route in comparison to other greedy forwarding techniques. They have provided the results of D-LAR for hop count and path throughput metrics. As we know that delay is an important metric for safety applications of VANETsbut they have not validated their proposed idea for this metric, while proposing the idea they said it is better for safety applications. D-LAR protocol is based on LAR even then D-LAR results have not been compared with the LAR protocol. In this work, we have evaluated the performance of D-LAR protocol in terms of delay, packet delivery ratio and routing overhead metrics. Its performance has beencompared with the existing LAR protocol. We have also analyzed how delay gets affected by the both MAC layer and routing layer parameters. The remaining article is organized as follows. The next section presents a brief overview of D-LAR routing protocol. Section 3 presents the review of some articles where authors did the mathematical modeling of delay with respect to wireless networks. Hop Count and delay analysis is provided in section 4. Section 5 and 6 presents the simulation results and conclusion respectively. 2.0 OVERVIEW OF D-LAR PROTOCOL Among the available position based routing protocols in the literature 8]}[ HYPERLINK \l "Ber09" 9 ], D-LAR is one of the most suitable position based routing protocol specially meant for highly dynamic dense networks. D-LAR is the extension of LAR protocol. LAR limits the route discovery area with the knowledge of speed and location of destination. Suppose, the destination node }is moving with the speed and at time , it is at location. With this information source node can expect that the destination node would be somewhere in the circular region around attime . This circular region is of radius . This region is called as expected zone where source node can expect that destination node would be at time . After constructing the expected zone, source node startsdefining the request zone. Request zone is the smallest rectangle whose bottom left corner is the current location of and top right corner covers the expected zone. The sides of the request zone are parallel to X and Y axis. Figure 1: Working of D-LAR protocol Source node initiates the route discovery process by broadcasting a route request message to its neighbors. Route request message encompasses the coordinates of the request zone. When a neighboring node receives the route request message, it checks its own positional information with the content of the route request message. If the neighboring node lies within the request zone it forwards the message further to its neighbors else it will drop the message. It can be visualized from figure 1 that A, B, C, G, I and F are the neighbors of S. Source node sends the route request message to all its neighbors. However, only three nodes A, B and C propagate the message further because they lie in the request zone. In this way the message is not propagated to all the nodes of the network. There is an involvement of few nodes which lies in the request zone. The route would be searched in the direction of destination; as a result the routing overhead is reduced. S would choose node B as the next-hop node because it has the minimum angular deviation in comparison to other two neighboring nodes. By using the same selection criteria, node B will select the next-hop node as H and then similarly H selects I and so on. This concept of LAR is used by D-LAR protocol for broadcasting of control packets, so that within the request zone all the nodes knew the location information about itself and its neighbors. After this, D-LAR selects the next-hop node among the neighboring nodes within the transmission range and the request zone. The next-hop node would be the one which has the minimum angular deviation from the line connecting source to destination. By using this next-hop selection strategy, message will not go out of way. S D