Survey of Modified Routing Protocols for Mobile Ad-hoc Network

IJSRD - International Journal for Scientific Research & Development| Vol. 1, Issue 5, 2013 | ISSN (online): 2321-0613
All rights reserved by www.ijsrd.com 1210
Abstract— In last few years extensive research work has
been done in the field of routing protocols for Ad-hoc
Network. Various routing protocols have been evaluated in
different network conditions using different performance
metrics. A lot of research has been done how to modify
standard routing protocol in ad-hoc network to improve its
performance. The hop count is not only metric that gives
efficient routing path. There are various modified protocols
which make the use of other parameters along with hop
count to select the best routing path to the destination. In
standard Ad-hoc On-demand Distance Vector (AODV)
routing protocol only hop count is used for selecting the
routing path. In this paper we have studied variants of
AODV protocols with modified routing metric.
Keywords: Mobile Ad-hoc Network, cross layer, AODV,
HMAODV, Bypass AODV, AODV-BR, E-AODV, F-AODV.
I. INTRODUCTION
Mobile Ad-hoc Network (MANET) [1] is collection of
mobile nodes which are connected with each other without
any infrastructure. We can also call it as infrastructure less
network. An Ad-hoc network is highly dynamic in nature
means that nodes are free to move, they can join or leave the
network at any time. In ad-hoc network adjacent nodes
which are within each other’s range can communicate
directly while non-adjacent nodes which are not within each
other’s range have to communicate through one or more
intermediate node/s. The topology of network keep on
changing as nodes keeps on joining and leaving the network.
Hence we need to have efficient routing protocol to maintain
the communication between nodes.
The routing protocols for wireless ad-hoc network are
classified as proactive, reactive and hybrid. In proactive
routing, the routes in the network are continuously evaluated
so that when any node has data to send, it also knows the
path to destination. Each node maintains a routing table
which contains network related parameters and metric
values. These routing tables are exchanged between the
nodes periodically. It is also known as Table Driven routing
protocol. In reactive routing, the routes are discovered and
established only when they are needed. The nodes are
unaware of network topology. Each node has to go through
route discovery and route maintenance procedure. The
routes remain valid till the destination is reachable or until
route is no longer needed. So paths are established on
demand of nodes which avoids unnecessary broadcasting of
network information. It is also known as on-demand routing
protocol. Hybrid routing protocol combines both proactive
and reactive routing strategy. The routing is initially
established with any proactive route and then demand for
additional nodes through reactive flooding. So the network
is partitioned into two layers, inner layer uses proactive
while outer layer uses reactive routing.
The Ad hoc On-Demand Distance Vector (AODV) [2] [3] is
designed for ad-hoc networks. It is purely on demand
routing protocol as routing does not depend on fixed path
and routing information is exchanged only when needed. In
AODV routing protocol messages, sequence number is used
by the node to determine the freshness of the information
contained with the node. Route Request (RREQ), Route
Replies (RREP) and Route Errors (RERR) are the various
types of message used in AODV routing protocol. AODV
routing procedure can be characterized into following
phases
A. Route Discovery Process:
Route discovery process is started when any node wants to
send data to other node and it does not have valid path to
that node. Here each node maintains two different counters a
node sequence number and broadcast ID. Broadcast ID is
incremented every time source node initiates new RREQ
message. The RREQ message is broadcasted to all the
neighbors of source node .When any intermediate node
receives RREQ, it replies to source if it has path to the
destination node. If it does not have path to destination
node, it further broadcasts the RREQ message to its
neighbors. When the destination node receives the RREQ, it
sends RREP. Route Reply message is unicasted by the node.
B. Reverse Path Set Up:
As RREQ message is traversed from source to the various
nodes to the destination, the reverse path is automatically
setup from all nodes back to the source. In order to set up
this reverse path, each node stores the address of the
neighbor from which it has received the RREQ message.
C. Forward Path Set Up:
When RREQ message is received by destination node itself,
it replies to the source node. If an intermediate node has
route to the destination node, before sending RREP to
source it checks the freshness of the path information it has.
If the sequence number of the path is greater than or equal to
sequence number in the RREQ, then it replies with path to
source. If the sequence number in the RREQ message is
greater than that of path with the node then instead of
sending this path information to source, it broadcasts the
RREQ message to its neighbors.
D. Route Maintenance:
HELLO messages are used for maintaining active route.
Each node maintains a precursor list which contains the list
of nodes which needs to be notified in the case of link break.
When any node detects link break for the next node in its
Survey of Modified Routing Protocols for Mobile Ad-hoc Network
Ashish B. Gaigole1
1
M.Tech Scholar, Department of Computer Science and Engg.
1
Visvesvaraya National Institute of Technology (VNIT), Nagpur, Maharashtra, India

(IJSRD/Vol. 1/Issue 5/2013/043)
path towards the destination, it sends RERR message to all
the nodes in its precursor list. Once the source node receives
the notification of the broken link, the source node can
restart the route discovery process if it has data to send to
the destination.
II. LITERATURE SURVEY
The standard AODV routing protocol had been modified
previously considering different parameters as mobility,
energy level, congestion, bandwidth and node load. In this
section we are going to discuss some of these modified
AODV routing protocols.
Heterogeneity and Mobility aware AODV (H-MAODV) [4]
have considered the relative velocity and the distance
between each node and node which is one hop away from it
to calculate one weight metric. This weight metric is
calculated at each node and this parameter is added as an
entry in the RREQ message. If the calculated value of the
metric at any node is more than that in the RREQ, the value
of the metric is replaced else the parameter value is kept
unchanged. The destination node selects the path whose
metric value in RREQ is least among all the paths. This
protocol mainly focuses on increasing the lifetime of the
routing path. It improves the packet delivery ratio than basic
AODV protocol. The weight function (Fij) is the parameter
that allows nodes to select best path as shown in equation
1.1.
Where,
α and β are the weights satisfied .
Dij is the distance between node i and node j.
Tri is transmission range of node i.
Vrij is relative velocity between node i and node j.
Vrmaxij is maximum relative velocity between node i and
node j.
In Throughput Enhancement in AODV Routing Using
Mobility Awareness [5] the movement of the nodes is
considered for selecting best route to the destination. Here
each node broadcasts its current location in the network at
regular time intervals inside the HELLO packet. Hence each
node is aware of the mobility of its neighbors. The RREP is
sent that neighbor which has more stability or less
movement. So path to destination has maximum possible
stable nodes. Hence it increases route lifetime, reduces re-
route discoveries and link breakage. Here throughput is
improved.
Bypass AODV [6] uses cross layering to identify link breaks
due to mobility and set up the bypass path between the end
nodes whose link is broken keeping rest of the path same.
Here MAC layer is modified and channel has assigned
different states which are used to distinguish packet loss due
to link break. Two new messages bypass-RREQ and bypass-
RREP are used for bypassing purpose. The information
about bypass route is stored in bypass routing-table. This
bypass routing table is used when any node receives RREQ
or RREP message to invalidate that particular route entry.
Bypass AODV shows continuously improved performance
as node density increases. Bypass AODV results higher
packet delivery ratio than standard AODV as it buffers the
packets for transmission over bypassed route. It reduces the
invocation of unnecessary route error messages which in
turn reduces routing overhead in the network.
Backup Routing in AODV (AODV-BR) utilizes a mesh
network to provide multiple alternate paths to existing on
demand routing protocols without producing additional
control messages [7]. Instead of depending on single path,
alternative paths are maintained at each node by overhearing
route reply messages of the nodes within the range. As soon
as link breakage is identified by any node it broadcasts
packets to its neighbors hoping that at least one of them will
have alternate route to the destination. That node also sends
RERR message to the source node so that source can start
new route. This mechanism is similar to that of DSR’s
routing mechanism. In this scheme mesh link is used to skip
the broken link in the route while in DSR node uses route
cache information to replace the entire route till destination
node on routing path. This algorithm results into better
packet delivery ratio and end to end delay for mobile nodes
network. But this is not that much effective in heavily
loaded network as it is in lightly loaded network.
In [8], cross layering mechanism is used to improve route
selection metric in which route is selected based on various
parameters as hop metric, node load, bandwidth etc. The
cross layering mechanism integrates the parameters from
different stack layers. The parameters from network layer,
MAC layer and physical layer are combined. The modified
route selection mechanism is done by combining different
metrics into single metric to get better performance. The
result shows that when multiple metrics are used for route
selection, performance is improved in terms of end to end
delay, packet delivery ratio and route load especially for
high speed mobile network.
In E-AODV [9] the main goal is to route the data packets
through the nodes which are expected to have better residual
energy. The energy consumption rate specifies how fast
node is consuming its remaining energy. In F-AODV [10]
main goal is to optimize the data forwarding in MANET by
minimizing number of forwarding nodes. The selection of
forwarding node is based on maximum battery level and
queue occupancy. Maximum battery level and queue
occupancy information are injected in route request and
route reply messages.
In [11], two modified protocols E-AODV which considers
energy consumption rate and F-AODV which uses cross
layer forwarding mechanism are compared. Both E-AODV
and F-AODV give better packet delivery ratio for variable
node density. Due to high reactiveness of E-AODV and F-
AODV to the link changes compared to AODV, E-AODV
and F-AODV have lower routing overhead compared to
AODV protocol. For lower data rates all three protocols
perform approximately same but for higher data rates E-
AODV and F-AODV outperforms AODV.
III. CONCLUSION
We have studied and reviewed various modified AODV
routing protocols for mobile ad-hoc network. In standard
AODV routing protocol the basic metric used for selection
of path is hop count. The use of other parameters with hop
count gives more efficient routing path that result into
improved network performance. As some network
parameters are varied, there was dip in the performance. We

(IJSRD/Vol. 1/Issue 5/2013/043)
conclude that there is need of a protocol that can adapt itself
according to network parameters to perform optimally.
IV. FUTURE WORK
As future enhancement, energy parameter of node with hop
count can be used as path selection metric to establish more
efficient routing path. Residue Energy Battery Model along
with AODV routing can be used for this. The initial energy
of any node is given the parameter
DEFAULT_FULL_BATTERY_CAP. In our algorithm we
will use one threshold energy paramater which will be
initialised as shown in equation 1.2.
Where,
To get the optimal value of α that gives best results, analysis
needs to be done on this modification in route discovery
process. Every node will first check its residue energy with
this threshold value. If the node has energy equal or more
than that of threshold value then only broadcast the RREQ
packets to its immediate neighbors. This process will be
continued until route is established or RREQ message is
received at destination.
REFERENCES
[1] S. Corson, Mobile Ad hoc Networking (MANET):
Routing Protocol Performance Issues and Evaluation
Considerations, RFC-2501, 1999.
[2] C.Perkins, Ad hoc On Demand Distance Vector
(AODV) routing, RFC-3561, 2003.
[3] Luke Klein-Berndt, A Quick Guide to AODV
Routing, National Institute Of Standards And
Technology (NIST).
[4] Bisngar Ahmed, Zytoune Ouadoudi, RzizaMohamed
and Ouadou Mohamed, “A mobility aware Modified
AODV for heterogeneous mobile ad hoc networks”,
IEEE 2012.
[5] Dr.S.A.Hussain, Dr.E.Garcia and M.Idrees, “In
Throughput Enhancement in AODV Routing Using
Mobility Awareness”, Unpublished.
[6] Ahed M. Alshanyour, Uthman Baroudi, “Bypass
AODV: Improving Performance of Ad Hoc On-
Demand Distance Vector (AODV) Routing Protocol
in Wireless Ad Hoc Networks”, ICST February 2008.
[7] Sung-Ju Lee and Mario Gerla, “AODV-BR:Backup
Routing in Ad-hoc Networks”, Defence Advanced
Research Projects Agency(DARPA).
[8] Jihong Shi, Jiang Yu, Haiyan Li and Rong Zong,
“AODV Protocol Optimization Based on Cross-Layer
Design of WMN”, IEEE, 2011.
[9] Lamia Romdhani, Christian Bonnet, “Energy
consumption speed-based routing for mobile ad hoc
networks”, Tokyo, Japan, WWAN 2004.
[10] Lamia Romdhani, Christian Bonnet, “A cross-layer
feature for an efficient forwarding strategy in wireless
ad hoc networks”, 20th
IEEE, Vienna, Austria, AINA
2006.
[11] Lamia Romdhani, Christian Bonnet, “Energy-based
routing optimization in MANET: Cross-layer
benefits”, proceeding of 9th
international conference
on mobile and wireless communication networks,
2007.

Survey of Modified Routing Protocols for Mobile Ad-hoc Network

More Related Content

What's hot (17)

Viewers also liked (15)

Similar to Survey of Modified Routing Protocols for Mobile Ad-hoc Network (20)

More from ijsrd.com (20)

Recently uploaded (20)

Survey of Modified Routing Protocols for Mobile Ad-hoc Network