Professional Documents
Culture Documents
ISSN: 2321-8134
IJFEAT
INTERNATIONAL JOURNAL FOR ENGINEERING APPLICATIONS AND
TECHNOLOGY
Abstract
VANETs are the networks that rely on their own vehicles to provide accident avoidance systems. Such networks have clear
distinctive features such as restricted mobility and high speed of nodes. One of the mainly vital challenges in such a network is
reliable routing of data packets among the vehicle users. Ant Colony Optimization is used to search diverse routes between nodes
in the network to aid in link failures. To attain scalability the network has to divide into many zones. The use of proactive approach
is to discover a route within a zone and reactive approach is to discover routes between zones. To reduce broadcasting and
overcrowding the local information stored in each zone is used. The algorithm make resourceful use of the network bandwidth is
scalable and is strong to link failures. The hybrid algorithm proved to be more efficient in terms of End to End delay, Jitter, Energy,
and Throughput for the network. In this paper the review of routing protocol which is proactive as well as reactive with Ant colony
optimization is shown.
Index TermsAnt Colony Optimization (ACO), IARP, IERP and Vehicular Ad-hoc Network (VANET).
I.
INTRODUCTION
http: // www.ijfeat.org (C) International Journal For Engineering Applications and Technology
MANTHAN-15
ISSN: 2321-8134
periodically to the neighbors. This scenario may cause extra
overhead particularly in the high mobility network. Though,
routes to destinations will always be obtainable when needed
[8]. Proactive protocols frequently depend on shortest path
algorithms to determine which route will be selected as they
usually use two routing strategies: Link state strategy and
distance vector strategy.
II.
OVERVIEW OF PROACTIVE
ROUTING PROTOCOLS
A.
DESTINATION
SEQUENCE
DISTANCE
VECTOR ROUTING (DSDV)
DSDV protocol it is an most primitive ad hoc routing protocol
in which it implements the distance vector approach and uses a
shortest path algorithm to employ simply single route to
destination which stored in the routing table, the information
about all accessible nodes with the entire amount of hops
required to reach these nodes are always found in each routing
table and each entry in the routing table is labelled with a
sequence number initiated by the destination node. To continue
routes accurately, every node should periodically broadcast its
routing table to its neighbors. The loop free routs, excludes
additional traffic caused by regular updates and reduces control
message overhead guarantees by DSDV protocol [10]. Though,
by maintaining nodes randomized conclusion which allows
each node to formulate a judgment whether to forward or
discard a packet.
B. OPTIMIZED LINK STATE ROUTING
PROTOCOL (OLSR)
OLSR protocol utilizes the link state strategy; it
provides a routing table contains information with reference to
every probable route to network nodes. Each node have to send
its updated information to various selective nodes when the
changed in network topology occurs, which rebroadcasts this
information to its other selective nodes. The nodes which are
absent in the particular list can just read and process the packet.
Some researchers thought that OLSR has simple process which
allows it to integrated different operating systems, also it
workings fit in the dynamic topology, though, OLSR may
cause network overcrowding; because of common control
packets which sent to switch topology changes, additionally
OLSR do not take into account the high assets capabilities of
nodes (like transmission range, bandwidth, directional antenna)
[10].
C.
http: // www.ijfeat.org (C) International Journal For Engineering Applications and Technology
MANTHAN-15
furthermore proactive protocols at regular intervals update the
routing information which lets these protocols to perform well
in low mobility networks.
III.
OVERVIEW OF REACTIVE
ROUTING PROTOCOLS
ISSN: 2321-8134
routing overheads by adding all route information to each data
packet, further, as the network cover greater distance and
together with extra nodes, the overhead will normally increase
and as result network performance will be degraded [9].
C. TEMPORALLY ORDERED ROUTING
ALGORITHM (TORA)
TORA is a distributed routing protocol using multi hop
routing; it is designed to reduce the communication overhead
related to frequent network changes. This protocol does not
employ a shortest path algorithm; therefore the routing
structure does not stand for a distance. TORA build up a
directed graph which includes the source node as the tree root.
Packets must be running from higher nodes to lower nodes in
the tree. If a node broadcasts a packet to a specific destination,
its nearer node will broadcast a route replay if it has a
downward link to the destination, if not, it drops the packet.
TORA ensures the multi path loop free routing; since the
packet always flows downward to the destination and don't
flow upward back to the sending node [4]. The advantage of
TORA offers a route to every node in the network and
decreases the control messages broadcast. Though, it causes
routing overhead in maintaining routes to all network nodes.
IV.
http: // www.ijfeat.org (C) International Journal For Engineering Applications and Technology
MANTHAN-15
nodes. If the destination is a member of the local zone of the
node, it sends a route reply on the reverse path back to the
source. The source node uses the path saved in the route reply
packet to send data packets to the destination.
V.
ISSN: 2321-8134
The ability of ants to self organize is based on four principles.
They are positive feedback, negative feedback, randomness and
multiple communications.
This is used to improve the good result.
When ants move from one node to another, the concentration
of the pheromone along that trail increases. This helps other
ants to travel in this path.
e feedback This is mainly used to destroy bad
solution. It can be done by decay of pheromone concentration
with respect to time. The rate of decay is problem definite. Low
decay rate encourages the bad result not being destroyed for
longer time and higher decay rate destroys good solution early.
Path to be taken by ant is completely random
hence there is possibility of generation of new solutions.
Multiple interactions The solution is found by interaction
of different agents, so one ant cannot find the food, as the
pheromone would decay. Hence more ants can find food
faster in food searching process [6].
Zone-based ant colony VANET algorithm In our algorithm the
network is divided into zones. For routing the packets, we
follow a proactive approach within the zone and a reactive
approach between the zones. The radius length measured in
hops determines the size of the zone. A vehicle can exist within
two overlapping zones and the zones can vary in size. A
vehicle is categorized as interior vehicle, boundary vehicle and
exterior vehicle. All the vehicles within a zone having a hop
distance of less than the radius are known as interior vehicles.
The overlapping vehicles within the zones with the hop
distance equal to the radius are known as boundary vehicles
and the vehicles with the hop distance greater than the radius
are known as exterior vehicles. Figure 3 explains about the
vehicles with an example. If the source vehicle is S and the
radius of the zone is 2 then vehicles A, D, F are boundary or
border vehicles, and vehicles B, C, E are interior vehicles. All
other vehicles are exterior vehicles i.e. the vehicles outside the
zone.
http: // www.ijfeat.org (C) International Journal For Engineering Applications and Technology
MANTHAN-15
proactively updates the information within the zone, whereas,
the Inter zone routing table tracks the information between the
zones, on demand. During route discovery and maintenance,
we have used ve different types of ants. These are: internal
forward ants, external forward ants, backward ants, notication
ants, error ants. The data structure of the ant contains Source,
Destination, Sequence number, Type, Hops, Speed, Position
and Path.
Source: Source node address is stored in source eld.
Destination: This eld stores the destination address. This
eld is left blank for internal forward ant and stores the
destination node address for external forward ants.
Sequence Number: Each ant is tagged with a sequence
number, stored under sequence number eld.
Type: There are ve types of ants which are described in
the Type eld. 0 is internal forward ant, 1 is external
forward ant, 2 is backward ant, 3 is notication ant, 4 is
error ant.
Hops: It is used to indicate the number of hops between the
node and all the nodes within its zone. This eld helps to
differentiate between a peripheral node and an internal
node. For internal forward ant, the zone radius is assigned
and for external forward ant, we leave this eld blank.
Speed: It is the speed of the vehicle.
Position: This eld contains the current position of the
vehicle.
Path: Path eld represents the sequence of nodes between
source and destination.
The phases are explained below as:
a) Route discovery within the zone: The Intra zone routing
table is used for route discovery within the zone. This table
will contain the information about all the vehicles within
the zone and the internal forward ant periodically (every
.20s) updates the vehicles information in the table. The
columns in the table represent all the vehicles within a zone
while the row represents the vehicles that are one hop away
from each of the vehicles that is the degree of each vehicle.
Therefore the size of the table is the number of vehicles
within a zone x the maximum degree of all the vehicles. In
this routing table, we have incremented and decremented
(evaporate) the pheromone concentration on the vehicles
based on the identied route. When source needs to send
information to the destination, it rst looks up the columns
of its intra zone routing table. If the destination is found
within its zone, the route discovery process is done.
Otherwise, the route discovery process continues between
zones.
b) Route discovery between the zones: When the vehicle fails
to nd the destination with the zone, Inter zone routing
table is used to identify the new route using the boundary
vehicles. External forward ants are sent to the boundary
vehicle for route discovery which will in turn forward to the
ISSN: 2321-8134
next boundary vehicle from its current location, until the
destination vehicle is found. If the destination vehicle is
found, then the backward ant will traverse back to the
source, based on the inherited route. However, if the
destination is not within the current vehicles zone or the
paths lifetime has expired, the route discovery process will
be repeated using the boundary vehicle from the current
zone.
c) Route maintenance: Route paths may break because of the
dynamic nature of the network. If the broken route is within
the internal zone, then it is repaired periodically using the
proactive approach, but if the Inter zone route breaks, then
the upstream vehicle of the broken link is used to store the
packets and nd an alternative path. When the alternative
route is identied, then a notication ant is used to update
the new route on the source vehicle. If an alternative route
is not identied, then an error ant is used to notify the
source vehicle about route failure [2].
VI.
CONCLUSION
VII.
REFERENCES
http: // www.ijfeat.org (C) International Journal For Engineering Applications and Technology
MANTHAN-15
ISSN: 2321-8134
http: // www.ijfeat.org (C) International Journal For Engineering Applications and Technology