VANET-Vehicular Ad-hoc Network
Back in the day when Karl Benz gave birth to the automobile industry in 1886, it was a big positive as it made life easy for the commuters. The car was built without any safety and it was put aside for a long time. Safety took a huge boost over the years, with the introduction of seatbelts, airbags and active safety components, today a car has a lot more safety devices to assist the driver. Despite being well equipped, cars today are more dangerous than other modes of transportation. For that, we need to upgrade safety by introducing the supreme form of safety which is digital communication among the vehicles.
VANET follows a basic concept which consists of a large web of vehicles and a wireless local area network (WLAN) technology which connects all the vehicles together to make communication possible.
The main aim of VANETs is to build an Intelligent Transport System (ITS). The safety-related applications represent the main aim of inter-vehicular communication. When an accident is detected, a vehicle can continuously broadcast information about this critical situation to the other vehicles. When a vehicle brakes suddenly, it gives information about the other vehicles to the driver. Dedicated Short Range Communication (DSRC) plays an important role in building V2V and V2I communication.
VANETs can provide a wide range of services like accident avoidance, a mechanism for regulating trafﬁc ﬂow, provision of internet access to the on-road public, information about the location of parking lots, restaurants, gas stations and infotainment applications such as playing games and listening to music.
Architecture of Vanet
Vanets can be divided into three domains
- The Mobile Domain– It consists of two parts.
- Mobile Device Domain- consists of all kinds of portable devices such as personal navigation devices and smartphone
- Vehicle Domain- consists of all kinds of vehicles such as cars and buses
- Infrastructure Domain– It consists of two domains
- Roadside Infrastructure Domain- Contains roadside entities like traffic lights
- Central Infrastructure Domain- Contains Infrastructure management centres such as traffic management centres and vehicle management centre.
- Generic Domain- Although the architecture of VANET varies from region to region. In V2X (Vehicle to Anything) communication mainly V2V communication, the architecture is a little different. It further comprises three domains
- In-Vehicle: It is composed of an onboard unit (OBU) and one or multiple application units. The connections between them are usually wired and sometimes wireless
- Ad-hoc- It consists of vehicles which have OBUs and RSUs (Roadside Unit). An OBU is a non-stationary unit and the RSU is the stationary unit. RSUs can communicate with each other directly or via multihop as well.
Infrastructure Domain- There are two types of infrastructure domain access
- Road Side Unit (RSU)
- Hotspot (HS)
OBUs communicate with internet through RSU or HS. OBUs can communicate with each other via cellular network when RSU or HSU are not present.
There are 4 types of communication architecture in Vanets
- In-Vehicle Communication– In vanet research, this is the most important domain. It can detect a vehicle’s performance and more importantly driver’s drowsiness and fatigue which is crucial in the safety of the driver and public.
- Vehicle to Vehicle Communication– It provides data exchange platform so that vehicle can share information among themselves to assist the driver.
- Vehicle to Infrastructure Communication– It allows the vehicle to communicate with the environment and infrastructure for real time traffic or weather updates. It provides environmental sensing and monitoring. It is one of the most important fields in Vanets.
- Vehicle to Broadband cloud communication– Vehicles communicate through cellular data mechanisms such as 3G/4G. Broadband cloud includes more traffic information and it monitors data and infotainment. It is useful for the driver in assistance and vehicle tracking.
The Open System Interconnection (OSI) model classifies similar communication functions into one of seven logical layers. The architecture varies in every region and hence there are different protocols and interfaces designed for them. For example DSRC in the US has a set of standard protocols and interfaces. Different protocols are designed to use at different layers. After amending IEEE 802.11 to IEEE 802.11p which adds standard Wireless Access in Vehicular Environments (WAVE), mainly to the physical (PHY) layer and the MAC sublayer. IEEE 1609 is a higher layer standard based on the IEEE 802.11p. IEEE 1609, represents a set of protocols that operates in the middle layers of protocol stack to support safety applications in vanets. Non safety applications are presented by another set of protocols which are IPv6 (Internet Protocol version 6), UDP (User Datagram Protocol) and TCP (Transmission Control Protocol) for network layer services and transport layer services.
Characteristics of Vanet
- High Dynamic Topology– The topology of vanet changes because of the movement of vehicles at high speed.
- Frequent Disconnected Network– As the network of vehicles widens the disconnection between vehicles become frequent when they exchange information.
- Mobility Modelling– The mobility of vehicle depends upon the traffic environment, roads structure, the speed of other vehicles, driver’s driving pattern etcetera.
- Battery Power and Storage Capacity– Due to high battery power and storage, vanets are helpful for making effective communication and making routing decision unlike manet (Mobile Ad-hoc Network)
- Communication Environment– Due to difference in communication environment between dense and sparse network, the routing approach of the dense and sparse network will be different. For example in dense areas, trees are considered to be obstacles and in sparse areas like highways trees are absent.
- Interaction with On-Board Sensors- For effective communication and routing decisions, the movement and current position of these nodes can easily be sensed by On-Board sensors like GPS.
Due to high dynamic topology, the design of efficient routing becomes highly complicated. The routing protocol of Vanet is classified into two categories
- Topology-Based Routing Protocol
- Position Based Routing Protocol
Topology-Based Routing Protocol
The topology based routing protocols use information within the network to send the data packets from source to destination. They are further sub-categorised into
- Proactive or Table Driven– They are based on shortest path algorithm and they keep the information of all nodes in the form of tables because these protocols are table based. These tables are also shared with their neighbours. Whenever any change occurs in network topology, every node updates its routing table.
- Route Discovery is not required
- Latency for real time applications is low
- A significant part of the available bandwidth is occupied by unused path
1.1 Fisheye State Routing (FSR) – It is a proactive routing protocol which maintains a topology table to nodes and updates the network information to the other nodes, which are in network.
- By exchanging only a part of routing information with neighbours only, the bandwidth is significantly reduced
- Routing overhead is reduced
- Changing in the routing table will not occur even if there is any link failure because it doesn’t trigger any control message for link failure.
- With increase network size, the storage complexity and the overhead processing of routing table is also increased
- The performance is poor in small ad-hoc networks
- Less knowledge about nodes which are farther
- Insufficient information for establishing route
- Reactive or On-demand- It is called Reactive or On-demand because it starts route discovery when a node needs to communicate with another node. It reduces the network traffic.
- For the upgradation of routing table the periodic flooding of network is not required. Flooding is required only when demanded
- It saves bandwidth as it is beaconless
- For route discovery the latency is high
- Excessive flooding of the routes causes disruption of the network
2.1 Ad-hoc On Demand Distance Vector (AODV) – It is reactive routing protocol that forms a route when a node requires to send data packets. It can unicast and multicast routing. It uses a destination sequence number which makes it different from other on demand routing protocols.
- The path is updated always because of destination sequence number
- Route redundancy and excessive memory requirements are reduced
- It responds to the link failure in network
- Can be applied to large scale ad-hoc network
- Time consumed for initial connection to setup and communication to establish is more
- If old entries are contained in intermediate routes then it can lead to inconsistency in the network
- Extra bandwidth is consumed in case of periodic beaconing
2.2 Dynamic Source Routing (DSR) – It forms route on demand and depend on source routing instead of table. DSR is beacon-less and does not require periodic hello packets. It utilises source routing and maintains active routing. It has two phases, route discovery and route maintenance.
- Periodic update is not required
- It uses caching which reduces load on the network for future route discovery
- It has small overload on the networks to obtain route between nodes
- Too many nodes in the network will lead to byte overload
- It’s efficiency reduces in high mobility
- Unable to repair broken links locally
- Unnecessary flooding will burden the network
2.3 Temporally Ordered Routing Algorithm (TORA) – It is a reactive and on-demand routing protocol. In TORA the node clearly initiates a query when it needs to send data to the destination. TORA creates, maintains and then erases the routes.
- When necessary, it creates direct acyclic graph
- It reduces network overhead, because all intermediate nodes don’t need to rebroadcast the message.
- Perform well in dense networks
- It is not scalable
Position based Routing Protocol
- Greedy Perimeter Stateless Routing (GPSR) – The location of the source node is found using GPS and the location of the neighbour node is found using beacons exchange. It consists of greedy mode and perimeter mode.
In greedy mode the source node chooses a node among its neighbour that is closest to the destination and forward packet to it. When the forwarding node finds no neighbour node close to the destination other than itself and even the destination is out of forwarding node’s reach the packet meets local optimum. In such condition, GPSR uses perimeter mode to overcome a local optimum problem.
When there is local optimum problem GPSR uses perimeter mode. It consists of two steps, first it creates graph planarization using relative neighbourhood graph (RNG) and second it uses right hand next neighbour node that relays the packet towards the destination.This protocol has not been considered appropriate in the city environment. The shortcomings of this protocol are
- Due to obstacles in the city environment its graphical planarization fails
- The perimeter phase uses long routes in relaying packets from source to destination end to end delay
- It creates routing loops, which increases routing overhead
- Geographic Source Routing (GSR) Protocol– It is a location based routing protocol designed for cities. The source node finds the location of the destination using reactive location services. It achieves shortest routing path between source and destination. It performed better than topology based routing protocols like Dynamic Source Routing (DSR) and Ad-hoc on Distance Vector (AODV) with respect to end to end delay and packet delivery ratio. The drawback of GSR is it selects junction statically without considering traffic density.
- Greedy Perimeter Coordinator Routing (GPCR)– It is proposed for cities and has two phases, the restricted greedy forwarding and the perimeter mode. The node that is present on the junction is called the coordinator node and it is responsible for making routing decision without considering digital map. The drawbacks of this protocol are
- Awareness of traffic is not known
- Perimeter phase causes delay in relaying the packet toward the destination
- Restricted greedy forwarding takes more hops than greedy forwarding
- Greedy Perimeter Stateless Routing Junction+ (GPSRJ+) – It is the enhancement of GPCR and avoids the redundant packet stop that increases the hop count. It uses two hop neighbour information to forecast which street its neighbouring vehicle will take.
- Anchor based Street and Traffic Aware Routing (A-STAR) – It establishes anchor path with high connectivity by using information based on statistically rated maps about city bus routes. A-STAR is better than GPSR and GSR because of its ability to achieve end-to-end connected path in the case of low traffic density.
- Traffic Flow Oriented Routing (TFOR) Protocol– It is a protocol for cities. Its main feature is to assume traffic flows while routing. It has two parts
- A junction selection mechanism based on trafﬁc ﬂows and the shortest routing path
- A forwarding strategy based on two-hop neighbour information
Greater is the traffic flow it will provide more connectivity which will increase the throughput of the network.
- Greedy Traffic Aware Routing (GyTAR) – It is a junction based geographic routing strategy for robust routes in cities. It has two modes
- A dynamic junction selection mechanism for accomplishing shortest path based on traffic density
- An improved greedy forwarding that forwards the packet in between two junctions
Due to the above modes the packet moves in the cities and present higher connectivity. The drawbacks of this protocol are
- The direction of the vehicle is not taken into consideration while choosing the next junction
- Protocol suffers from a local optimum problem that affects the performance of the entire network
- Enhanced Greedy Traffic Aware Routing Protocol (E-GyTAR) – It is an upgraded version of GyTAR. It also has two phases
- Dynamic junction selection mechanism based on directional density
- Improve greedy forwarding for routing in between junction
It achieves shortest routing path on the basis of directional density thereby helping the packet to reach the destination. The major drawback with this routing is that it selects junctions based on directional density and ignores the non-directional density ﬂows on a multi-lane road.
- Directional Geographic Source Routing (DGSR)– It is an improved version of geographic source routing with directional forwarding strategy. The source node uses location services to get to the destination node. It computes the shortest path from source to destination which is composed of sequence of junctions.
- Enhanced Greedy Traffic Aware Routing-Directional (E-GyTARD) – It is an enhanced version of E-GyTAR with directional forwarding. It consists of two mechanisms
- Junction Selection
- Directional Greedy Forwarding Strategy
It uses location services to get to the destination node. Junction is selected on the basis of directional traffic density and shortest distance to destination.
Simulation is an important step in implementing new technology in vanets. There are two components of vanet simulation
- Traffic Simulator– In order to analyse vanet characteristics and protocols, traffic simulators are required to generate position and movement information of a single entity in vanet environment.
- Network Simulator– To test and figure out the operability of vanet, a good network simulator should possess some features including a comprehensive mode, efficient routing protocols like AODV (ad hoc on demand distance vector), and communication standards like IEEE 802.11[p] and IEEE 1609 specifications.
Implementing VANET is a task filled with numerous technical challenges. One of the main areas of research is secure and private routing of VANET. Security is vital to prevent the system from cyber thefts. Vanet works in two environments in urban areas, first, cities where there are high buildings which are a hindrance to the network and second, highways where such obstacles are not present. To develop a routing strategy for these would be another area of research in the future. The topic of VANETis still very raw and can be researched widely in the future but once we are presented with the secure and reliable connection, VANET will surely be of great help in the road safety.
Visit Homepage for latest updates and Technical articles: Automotive Electronics
Visit our forum to discuss: Automotive Electronics Forum