Survey on Various DTN Routing Algorithms for Vehicular ADHOC Network (VANET)

- Vehicle Adhoc Network (VANET) permits self-organized infrastructure. Link or Connectivity breakage or long delay (intermittent connection) is an important issue in the VANET which arises due to the dynamic topological changes. Mainly two algorithms Vehicle assisted Data Delivery (VADD) and Opportunistic routing algorithm supports for handling delay/disruption tolerant network (DTN). Intention of DTN is to create dynamic link between the sender and receiver vehicle in the case of link disruption or absence of path between two nodes and forwards the data using store carry and forward. VADD carry and forward the data as much as possible in disconnected network using intersection mode, straightway mode or destination mode. Opportunistic routing can deal the unreliable data and intermittent connectivity in VANET by considering the multiple nodes as a next hop forwarder instead of considering pre-selected single node to be the best forwarder. This paper gives the study about various opportunistic routing algorithms for handling delay/disruption tolerant network.


Introduction
The development of ITS (Intelligent Transportation System), to improve road safety and disseminate the emergency information in network, automobile industry triggered the researchers in the field of VANET. communication Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) are the two types of vehicular Adhoc network which is supported by VANET using the IEEE standard 802. 11p.VANET is the deputize class of MANET here the nodes are naturally mobile vehicles. There is a high possibility for the vehicles to mobile anywhere because of its natural characteristics. Because of high mobility there is a high chance for frequent link disconnection and it cannot provide reliable data transfer [1]. Due to the spare connectivity, broken connectivity, high movement and latency, long delay, high error rates data communication gets interrupted between the vehicles, that can be handled and communication takes place between the vehicles by using the disruption tolerant network (DTN). In this network vehicles carries the data until it identifies the appropriate node to deliver or forward the data via the node. This is called as store-carry and forward mechanism. Vehicle DTN supports for many applications like emergency information broadcasting, advertisements, displaying the vehicle parking slot availability etc. Opportunistic routing and VADD are the two routing algorithms used in VANET DTN.
A node carries the data until it meets the trusted intermediate node or the destination node in its coverage range. Due to the unpredictable nature of VANET using the pre-computed paths will lose the data while forwarding so the dynamic path selection at junctions will provide the optimal solution for reliable communication. VADD selects the path will low packet delivery in two situations. When messages route by node are located in the middle of a road and when the located in a junction. Opportunistic routing is trajectory-based protocol uses both the vehicle movement and the geographical information provided by Global Positioning System (GPS). In Opportunistic routing nodes stores the data until it receives a suitable node. For picking the appropriate next hop each node calculates the closest point in their trajectory in track of the destination [2].

Routing Issues and Challenges in VANET
There are varieties of routing protocols are used in VANET. Broadcast routing, Cluster based routing, Geocast routing, Position based routing, Topology based routing and hybrid routing. Each and every routing methodology is having some challenges during the implementation. In this survey positionbased routing is considered and one important issue is acknowledged, disruption or delay is considered as an important issue and handling this situation is a big challenge for researchers. Opportunistic routing algorithm provides the solution for this issue by implementing this algorithm in Delay/Disruption Tolerant Network.

A location aided controlled spraying routing (LACS)
In this paper, a Location Aided Controlled Spraying (LACS) routing algorithm is proposed to deal with the thoughtprovoking issues in DTN routing. In LACS there is no need of global knowledge about the network and only the contacted nodes carrying the routing information alone need this algorithm. Two stages of routing process are used, (i) controlled spraying routing stage and (ii) single-copy routing stage. In controlled spraying routing stage, the current node uses the scattering technique depends on the position of interaction node. In the single-copy routing present node forecasts the terminus node location using Semi Markov process and then decides to onward the message to the mark node. Random Way Point (RWP) model is used for simulation. Comparing to the traditional model LACS produced higher delivery ratio and less transmission when the buffer space increased, the transmission delay increased if buffer space is less than 20MB [3].

An Opportunistic Routing for Data Forwarding Based on Vehicle Mobility Association (OVMA)
Identifying the appropriate data forwarding path for data conduction with short delay is an imperative role in VANET. In this paper vehicle mobility association-based routing protocol for data forwarding in VANET is proposed. Forwarding to the extra intermediate nodes can be reduced with this routing protocol. The Manhattan Mobility model is used for the vehicle movement. Probability for data conduction is fabricated into the Hidden Markov Model (HMM) decision. By using this model, the forwarding movement probability is set to 0.5 and the possibility of left turn and right turn is 0.25 respectively. Each vehicle carries the information about the associated vehicle, so that the routing decision can be done based on the vehicle density. The OVMA protocol can escalation the network lifetime, expand the performance of data delivery ratio, and shrink the data delay and routing overhead when compared to the other well-known routing protocols. The simulation result shows that the routing overhead increases progressively when the number of vehicles increases because the routing frequency is high. Finally, the path discovery is considerably slow because of increasing the number of nodes participating in the network [4].

Joint Link State and Forwarding Quality: A Novel Geographic Opportunistic Routing in VANETs (LF-GOR)
Important challenges of routing protocols in VANET are it suffers from frequent link disconnection and delay in packet delivery. To address this dispute, Geographic Opportunistic Routing protocol is introduced established on link state and Forwarding quality inside nodes (LF-GOR). For determining the candidate nodes set of the transfer node in LF-GOR, one hybrid approach is used namely filtering and prioritizing which considers the position information, link state and node's forwarding quality. In the existing, link state only considers the transmission time q but in this paper addition to the transmission time q it considers the retransmission limit ξ also for considering the link state it is notable one. Forwarding quality inside a node is done by introducing the new layer DaMo (Data Monitor) between the MAC layer and Network Layer of OSI model. This layer is used to monitor the incoming and outgoing packet traffic through DaMo.200 nodes are taken for the simulation when 140 nodes are used it is having the peak packet delivery rate. This delivery rate is not increased gradually when the nodes are increased. It uses the influence of link state to improve the network throughput it worked really well but throughput is not gradually increasing when the no of nodes is increased. On the whole this algorithm is having acceptable end to end delay [5].

Location and Direction Aware Opportunistic Routing in Vehicular Ad hoc Networks (LDAOR)
Routing in VANET is considered as an important issue, meanwhile of its exceptional feature such as deficiency of energy constraints and frequent link breakage. Designing routing algorithm for highly dynamic network is the challenging task, so the appropriate opportunistic routing called Location and Direction Aware Opportunistic Routing (LADOR) is introduced in this paper where the transmission is performed using store-carry and forward mechanism. LADOR has two phases i) neighbor node selection phase ii) determining the priority of message for transmission. Neighbor node is selected by using vehicles locations, vehicles directions, and prioritising of messages from buffers, based on contact antiquities and neighbour nodes position to destination. When multiple nodes approach the carrier node, the neighbor node is selected, which is closer to the destination. The end-to-end delay is high when the no of nodes is small because of the parameters used by the LDAOR, but the overhead is less in all situations that can be appreciated. Additionally, traffic flows from buffers are considered in this paper which will supports for improving the delivery ratio. A message in a buffer only sent to the neighbor node which is very nearer to the destination node, because of this additional sending and receiving of messages are reduced, so that the bandwidth can be efficiently used. LADOR not only increasing the delivery rate, it also avoids flooding and traffic flows [6].

An opportunistic routing based on symmetrical traffic distribution in vehicular networks (OSTD)
Finest route selection is the important performance metrics for resourceful data delivery with low end to end in VANET. In this paper adaptable routing algorithm is introduced based on the symmetrical traffic distribution (OSTD). This algorithm firmly considers the vehicular dissemination for the calculation of efficacy function which is used to estimate the network route. Considering the vehicular distribution for route selection is very important than other factors such as vehicle density and Euclidean distance. This OSTD algorithm is designed as two-phase algorithm i) intersection selection phase and ii) Next hop selection phase. Line running method is used for improving the data delivery ratio. The result shows that the data delivery ratio is gradually reduced once the data distribution rate is increased. Data delivery ratio is increased when the numbers of nodes are increased gradually. Due to the week wireless link the fraction of packet loss is increased when the packet transfer rate is improved. In OSTD end to end delay is very less, even in the increased quantity of nodes and data distribution rate [7].

A Trusted Routing Protocol Based on GeoDTN+Nav in VANET (GeoDTN+NAV)
Trust management between nodes becomes more difficult in the VANET because of the characteristics of high mobility. In this paper this problem is solved by introducing the new reliable routing protocol based on GeoDTN+Nav by using Bayesian faith supervision model. This GeoDTN+NAV is outfitted with Virtual Navigation Interface (VNI), which is used to find the nearby vehicles and provides navigation communication data with dependable format.VNI contains two elements (Nav-info, confidence). Nav_info provides the thorough route to the destination and vehicle movement direction. Confidence provides the probability to reach destination node. This protocol comprises of four phases i) initialization Phase, ii) Route discovery, iii) Route establishment and iv) Route deletion. In simulation results shows that correct ratio of received packets is high and number of packets forwarded by this routing protocol is also considerably high [8].

LSGO: Link State aware Geographic Opportunistic routing protocol for VANET (LSGO)
Due to the unstable linkage connectivity and high topological deviations robustness and competent data delivery is the issue in VANET. With the help of broadcasting characteristic and additional backup links the opportunistic routing can improve the consistency of routing. In this paper such routing algorithm is introduced based on the state of the link and location called link state aware geographic opportunistic routing (LSGO). Based on the vehicle location and link quality this protocol improves the reliability of data by selecting forwarders and prioritizing. This protocol contains three parts i) link quality estimation, ii) candidate node set selection done by link transmission rate and iii) priority scheduling algorithm based on the time. In the simulation the results show that packet delivery ratio is great, packet dropping rate is low, end to end delay is acceptable and routing overhead is increased with the increase in vehicle speed and number of vehicles [9].

Connectivity-aware minimum-delay Geographic routing (CMGR)
For the routing issue raised in VANET due to the continuous topological changes, Connectivity-aware Minimum-delay Geographic Routing (CMGR) is introduced in this paper. CMGR supports efficiently for both the sparse and dense network. For extending the chance of packet reception in sparse network CMGR considers the connectivity of routes in its path selection. In dense network CMGR determine ample connectivity between nodes and selects the route among them with minimum delay. IPv6 is considered as network protocol and anycast addressing is used in this paper. Navie approach is used to select the route for selecting the appropriate route with maximum value o the minimum vehicle density. In this paper two limitations and its solution are addressed i) For moving destinations, vehicle tracking mechanism and ii) Sparse junction packet forwarding decision. Vehicle tracking mechanism increases the control overhead because of generating beaconing messages at the junction. The simulation result shows that packet delivery ratio is increased but not gradually increased when the numbers of vehicles are increased. Delivery delay is high in sparse network and it is gradually getting decreased in dense network. Drawback of this mechanism is dropped data packet ratio is increased when the vehicle count is increasing [10].

Road traffic and geography topology based opportunistic routing for VANETs (ORRIS)
In this paper opportunistic routing is proposed based on topographical location and path traffic flow analysis for handling the issue transmission delay and packet delivery ratio. Generally, data transmission takes place between two vehicles using dedicated short range communication. For making communication between two vehicles this ORRIS algorithm use the Road Side Unit (RSU), since the structure of the network is restricted to the spatial dissemination of road and store carry and forward mechanism is used in VANET. This algorithm is generated for straight road communication which uses greedy forwarding and intersection road communication which considers the angle between two nodes as important parameter. Simulation ratio shows the packet delivery success ratio is high when the numbers of vehicles are increased, transmission delay is decline and steady and dropped packet ratio is also considerably low [11].

Multi-Metric Opportunistic Routing for VANETs in Urban Scenario (MMOR)
Multi-hop routing is a prominent issue in VANET, based on that this paper introduces an opportunistic algorithm called Multi-Metric Opportunistic Routing (MMOR). This algorithm identifies the best opportunistic next forwarding node for transmitting the data from source to destination. MMOR contains three portions i) Collection of neighbour node information ii) Forwarding candidate set selection and iii) Unscrupulous forwarding. Nearby node information can be collected with the help of node fortified with device spot and electrical map. In forwarding candidate set selection it considers the distance between candidate node and destination node, Candidate node packet process load, candidate node moving direction and velocity and neighbor node density. Since MMOR takes multi metrics into consideration for best routing the packet drop ratio is low, simulation results show throughput is low and end to end delay is high when the numbers of nodes are increasing gradually. In the conclusion self-adaptive routing according to the dynamic traffic network is considered as future work [12].

A Localized Efficient Forwarding Algorithm in Large-scale Delay Tolerant Networks [LOOF]
In VANET effective forwarding can be done by using global or local information in delay tolerant environment. In this paper by considering the local information alone an opportunistic algorithm is created called Local Optimistic Opportunistic Forwarding [LOOF]. By using the available local information this algorithm works optimistically and the simulation results shows that packet delivery rate is high and delay is very low [13].

Specific Opportunistic Routing at Road Intersection in VANETs (SOR)
Self-Organized network can be generated without the support of any permanent infrastructure in VANET and routing is a thought-provoking issue in due to the high motion, dynamic topology and high density of vehicles. In this paper positionbased routing algorithm is proposed for road intersection. Advantage of the position-based algorithm is forwarding is done with the awareness of destination node location, individual position and the position of one-hop neighbor so that forwarding turn out to be easy. This kind of algorithm will not support for rural area because of terrain effects and GPS cannot extend its service in the tunnel because of the absence of signal in such environment. This algorithm is executed based on the shortest path and best junction selection. Drawback of this algorithm is performance is getting reduced in high mobility, when the nodes count increased data delivery delay is increased and packet delivery delay is decreased.

Comparative Analysis
The following Table 1 shows the analysis of various Opportunistic routing algorithm contribution in VANET for attaining various parameters.

Conclusion
Routing in VANET is a challenging task because of its intermittent connectivity and high mobility. These characteristics result in degrading the routing performance of using traditional routing protocols. In order to handle the delay/disruption two new routing algorithms VADD and Opportunistic are suggested for delay/disruption tolerant network. In this paper various Opportunistic routing algorithms are discussed and its various parameters like Delivery ratio, Network Overhead, Transmission delay, Reliability, Throughput, traffic and packet loss and security were analysed.