Communications and Network, 2013, 5, 611-617
http://dx.doi.org/10.4236/cn.2013.53B2110 Published Online September 2013 (http://www.scirp.org/journal/cn)
QoS Aware Routing Protocol to Improve Packet
Transmission in Shadow-Fading Environment for Mobile
Ad Hoc Netw ork s
T. Sangeetha1,2, K. K. Venk atesh1,2, Rajesh1,2, M. S. K. Manikandan1,3
1Department of Electronics and Communication Engineering, Madurai, India
2SACS MAVMM Engineering College, Madurai, India
3Thiagarajar College of Engineering, Madurai, India
Received July 2013
ABSTRACT
In Shadow-Fading model, it is difficult to achiev e higher Packet Delivery Ratio (PDR) due to the effects of large scale
fading. The main goal of this paper is to improve the PDR in Shadow-Fading environment. To achieve this objective a
protocol has been proposed that discovers backup routes for active sessions. These backup routes salvage the packets of
active session in case of active route failure. It is found by research that proactively maintaining backup routes for ac-
tive sessions can noticeably improve the PDR in Shadow-Fading environment. This protocol has been implemented
with a mechanism of having backup routes and simulations have been conducted by using both node disjoint paths and
link disjoint paths. Comparisons have been made between new protocol as well as AODV protocol. Simulation has been
carried out using Network Simulator 2 (NS2) and the results show that the proposed QoSAR protocol exhibits higher
PDR than AODV pro toco l in S hadow-fading environment.
Keywords: Shadow-Fading Model ; Adhoc On Demand Distance Vector Routing (AODV); Backup Routes; Node
Disjoint Paths; Link Disjoint Paths; Packet Delivery Ratio (PDR); Quality of Service Aware Routing
(QoSAR)
1. Introduction
In Mobile Ad h oc Netwo rks (MANE Ts), the prov ision of
Quality of Service (QoS) guarantees is much more chal-
lenging than in wireline networks, mainly due to node
mobility, multihop communications, contention for chan-
nel access, and a lack of central coordination. QoS guar-
antees are required by most multimedia and other time or
error sensitive applications. The difficulties in the provi-
sion of such guarantees have limited the usefulness of
MANETs. The QoS routing protocol is an integral part of
any QoS solution since its function is to discover which
nodes, if any, are able to serve applications’ requirements.
Consequently, it also plays a crucial role in data session
admission control.
MANET applications are nowadays used in military
and in emergencies, entertainments and outdoor industry
environments where centralized administration is diffi-
cult and costly to install [1]. Two nodes can communi-
cate if they are within each other’s radio range, otherwise,
routers if they are out of range, thereby it becomes mul-
tipath routing. These networks have several salient fea-
tures like rapid exploitation, robustness, flexibility, in-
trinsic mobility support, highly dynamic network topol-
ogy, the limited battery power of mobile devices, limited
capacity and asymmetric or unidirectional lin ks [2].
Routing has always been one of the key challenges in
Mobile Adhoc Networks and the challenge become more
difficult when the network size increases [5]. Man y multi
path routing protocols have been proposed for Ad Hoc
Networks [3]. The routing protocols mainly are either ta-
ble-driven (Proactive) or on-demand (reactive) routing
protocols. Many hybrid routing protocols having the com-
bination of functionality of both proactive as well as reac-
tive routing protocol are also proposed. The proactive
routing protocols periodically update the routing tables.
When there is a request to forward message the routes
are available in the routing table. On the contrary reac-
tive routing search the route when there is a request for it.
In the route search operation the reactive routing proto-
cols, find multiple paths for the same source and destina-
tion pair. One out of these multiple paths is selected to
forward messages to the destination.
The basic idea of On-Demand routing protocol is to

T. Sangeetha ET AL.
612
search route when it is required. Due to the mobility
characteristic of MANET nodes and Dynamic Topology
of the network, there are always multiple paths available
between the sources to destination pair. The On Demand
routing mechanism suggests using the route that is con-
sidered best according to the required Quality of Service
for the transmission, as in [3] hop count, is th e parameter
for selecting a route. Once the best route out of searched
route is chosen, the other routes are not given any con-
siderations. But maintaining information about these ad-
ditional routes will be more advantageous [2-4]. These
additional routes can be used as backup routes in the
events of link failures. In our paper the route failure can
be overcome by disjointness for shadow fading environ-
ments.
This paper mainly focuses on the idea of finding mul-
tiple paths (Multi-path) using Node Disjoint and Link
Disjoint paths to improve Packet Delivery Ratio (PDR)
of the network in Shadow fading environments. In addi-
tion to this, the study also targets to evaluate the perfor-
mances of AODV and QoS Aware Routing (QOSAR)
protocols based on PDR and throughput. The organiza-
tion of the rest of the paper is as follows. Section 2 QoS
routing challenges in MANET. Section 3 briefs about the
related work. Section 4 presents the discussion about pro-
posed routing protocols and in Section 5 Result and Dis-
cussion.
2. QoS Routing Challenges In MANET
The design of efficient routing protocols is a critical issue
for both wired and wireless networks. QoS-Aware Routing
(QoSAR) protocols are facing the following design is-
sues and constraints. These issues and constraints must
be kept in mind while designing these protocols.
2.1. Node Mobility
A MANET consists of mobile nodes. Nodes form the
network only when they are in the communication range
of each other. If they move out of range, link between the
two nodes is broken. At times, f ailure of a single link can
lead to the major network partitioning. Hence, mobility
of the nodes is a major challenging issue for a stable
network. Also, failure of certain links results in routing
decisions to be made again.
2.2. Lack of Centralized Control
The advantage of an ad-hoc network is that it is formed
spontaneously without fixed architecture. Participant nodes
or members can change their positions dynamically and
can join or leave the network independently. Due to this
nature of mobile ad-hoc networks it is difficult to pr ovide
centralized control. Th us it is difficult to achieve efficient
and fair Medium Access Control (MAC). Thu s those com-
munication protocols are preferred which utilize only lo-
cally available states and operate in a completely distri-
buted manner [6]. This increases complexity and an al-
gorithm’s overheads, as information about the participant
nodes must be collected efficiently. There is no central
entity to collect resour ce state information and admission
decisions. Instead of central administration, nodes must
make decisions based on available network resources,
which may lead to potential inaccuracies. Due to lack of
infrastructure the control and management of the network
is distributed among the nodes.
2.3. The Unreliable Wireless Channels
Due to interference from the other transmissions, mul-
ti-path fading and shadowing effects the received signal
are flat to bit errors. Such errors may lead to packets be-
ing not decidable. However, persistent packet error can
result in link failure, leading to re-routing, increased pack-
et delay and cong estion and packet dropping [7 ].
2.4. Multiple Paths
To send data from a source to destination, a path has to
be found beforehand. If a single path is established, send-
ing all the traffic on it will deplete all the nodes faster.
Also, in case of path failure, alternate path ac ts as a back-
up path. Thus, establishing multiple paths aids not only
in traffic engineering but also prevents faster network
degradation.
2.5. Node-Disjoint Paths
Multiple paths between the two nodes can be any link
disjoint or node disjoint. Multiple link disjoint paths may
have one node common among more than one path. Thus,
traffic load on this node will be much higher than the
other nodes of the paths. As a result, this node tends to
die much earlier than the other nodes, primary to the paths
to break down much earlier. Thus, the presence of node
disjoint paths prolongs the network lifetime by reducing
the energy fall rate of a particular node [8-9].
3. Related Works
This section briefs about the research work on the idea of
mul ti-path routing in M obile Ad hoc ne t works.
3.1. Ad Hoc On-Demand Distance Vector
Routing
In this paper titled “Ad Hoc On-Demand Distance V ector
Routing”, Feb- 1999, the authors C.E. Perkins and E.M.
Royer have presented a distance vector algorithm that is
suitable for use with ad-hoc networks. In AODV each
Mobile Host operates as a specialized router, and routes
are obtained as needed (i.e., on-demand) with little or no

T. Sangeetha ET AL.
613
reliance on peri odic a dverti sem ents . Their sim ulati on shows
that on-demand route establishment with AODV is quick
i.e., and accurate. AODV is a reactive routing protocol,
meaning that it provides a route to a destination only on
demand. Because AODV does not require global period-
ic routing advertisements, the demand on the overall
available bandwidth to the mobile nodes is substantially
less than in those protocols that do necessitate such ad-
vertisements.
The main difference between AODV and DSR [5,11]
stems out from the fact that DSR uses source routing in
which data packet carries the complete path to be tra-
versed. In AODV, the source node and the intermediate
packet nodes store the next hop information correspond-
ing to each flow for data packet transmission. In this
protocol, the source node floods the Route Request pack-
et in the network when a route is not available for the
desired destination. When an intermediate node receives
a request for route, it either forwards the request or pre-
pares a reply for that route if it has a valid route to the
destination. The validity of a route at the intermediate
node is determined by comparing the sequence number at
the intermediate node with the destination sequence num-
ber in the Route Request packet. It may thus obtain mul-
tiple routes to different destinations from a single Route
Request. Multiple Route Reply packets in response to a
single Route Request packet lead to heavy control over-
head. AODV avoids the counting-to-infinity problem of
other distance-vector protocols by using sequence num-
bers on route updates. AODV has the ability for both
unicast and multicast routing. But considering this case
the AODV protocol not perform well in shadow fading
environments its proved in our simulations.
3.2. QoS Routing Solutions
In this paper titled “A Survey of QoS routing solutions
for Mobile Ad Hoc Networks”, Apr-June 2007 the au-
thors L. Hanzo II and R. Tafazolli have presented a QoS
routing solutions that is suitable for use with ad-hoc
networks [10,13].
This paper offers an up-to-date survey of most major
contributions to the pool of QoS routing solutions for
MANETs published. It includes a thorough overview of
QoS routing metrics, bandwidth, factors affecting per-
formance and classify the protocols found in the litera-
ture. The aim of MAC protocol was to provide a basic
best-effort level of service to ensure network operation in
the face of an unpredictable and shared wireless commu-
nication medium and to maintain a network topology
view and routes in the face of failing links and mobile
devices. QoS routing protocols play a major part in a
QoS mechanism, since it is their task to find which nodes,
if any, can serve an application’s requirements. Therefore,
the QoS routing protocol also plays a major part in Ses-
sion Admission Control (SAC), since that is dependent
on the discovery of a route that can suppo rt the requ ested
QoS.
The majority of the solutions proposed in this litera-
ture till now have focused on providing QoS based on
two metrics: throughput and delay. Of these, the more
common is throughput. This is most likely because as-
sured throughput is somewhat of a “lowest common de-
nominator” requirement; most voice or video applications
require some level of guaranteed throughput in addition
to their other constraints. The proposed protocol will im-
prove the QoS based on the above two metrics.
4. Protocol Implem ent at ion
In this section, we describe the implementation of pro-
posed protocol, which exploits the knowledge of alterna-
tive or backup routes to a source’s destination in order to
improve the robustness of throughput-QoS assurances in
the face of route failures. The main goal of this paper is
to improve the Packet Delivery Ratio (PDR) in shadow
fading environments. Thus by using the proposed proto-
col Quality of Service (QoS) can be improved as much as
possible. Hence in the following sections we are going to
use the name as QoSAR (Qos Aware Routing) for the
propose d protoc ol .
In shadow fading environments, it is difficult to achieve
high PDR only with the use of primary routes from source
to destination. Idea behind QOSAR is to find multiple
routes between source and destination. Hence it improves
PDR as well as it gives solution to the route failures.
In the newly proposed QoSAR protocol, once the ses-
sion is ready to transmit, a backup route for that session
must be found. We have considered two mechanisms to
find multiple routes as follows:
1) Link Disjoint Paths
2) Node Disjoint Paths
4.1. Link Disjoint Paths
Link disjoint paths are paths between sources to destina-
tion which have no overlapping links. The backup route
is selected in such a way they are “sufficiently disjoint”
i.e., it includes no more than half of the links in the cur-
rent/primary route [12].
In Figure 1 three paths (shown in red color) between
sources to destination are called as link disjoint paths. If
the primary and backup routes are selected in such a way
that, they are sufficiently link disjo int then the route fail-
ure is minimized.
If a new backup route must be discovered, the search
packet, referred to as RReq backup carries a copy of the
session’s primary route. To avoid fully flooding the net-
work with the RReq-backup, once the RReq backup has
traveled at least half of the length of the primary route,

T. Sangeetha ET AL.
614
Figure 1. Link and Node Disj oint paths.
the disjointness condition is enforced and the packet is
dropped if the partially discovered route does not comply.
Also, the RReq backup Time to Live (TTL) is only one
greater than the primary route length, again to reduce the
extent to which the network is flooded [12].
4.2. Node Disjoint Paths
In node-disjoint paths no node is common other than the
source and destination. This work enables discovery of
two node disjoint paths from source to destination. In this
protocol Hello messages have been used to identify two
node disjoint paths. Three control packets are required
for setup two node-disjoint routes, which reduce control
overhead in the network and also setup backup route
faster that reduces the end to end delay.
In QOSAR protocol, each node periodically broadcasts
the HELLO messages to inform its neighbors that it has
not moved away. When a node receives a HELLO mes-
sage from a neighbor, a route to this neighbor is only
added to the routing table when the neighbor does not
already exists. If the neighbor exists, its lifetime is in-
creased. When the network topology changes and HEL-
LO, message does not received for a defined period of
time the route expires.
In the above Figure 1, Intermediate node 4 is a pri-
mary route (1-4-7) from source to destination. If any
failure occurred in the primary route it will take alternate
route as (1-2-3- 4) and (1-5-6-7) using node disjoint
paths. If the primary and backup routes are selected in
such a way that, they are sufficiently node disjoint then
the route failure is minimized.
5. Results and Discussion
Simulations have been conducted in the environment of
Network Simulator (NS2) . The simulations start with t he
small network of 7 nodes.
Figure 2 shows small network with minimum number
of nodes and the simulation parameters are listed in Ta-
ble 1. We con sidering node 0 as a TCP traffic source and
node 3 assumed to be TCP sink. This simulation results are
presented in Table 2.
Figure 2. Small network with 7 nodes.
Table 1. Simulation Parameters Used for small network.
Parameter Va lue
Simulation area size 100 m × 100 m
Number of nodes 7
Number of traffic source 1
Simulation time 100 s
Traffic source type TCP
Propagation model Shadowing
Path loss exponent 2, 2.2
Table 2. Simulation Results for small network.
Path Loss Parameter AODV Proposed Protocol
2
Packets sent 35983 42896
Packets received 32547 42567
PDR 90.451 99.233
Throughput 4.72127 × 107 7.3808 × 107
2.2
Packets sent 5132 19948
Packets received 4196 18498
PDR 81.7615 92.7311
Throughput 876858 1.4964 × 107
After observing the performance of QoSAR protocol
in small network, then our simulations focused to ob-
serve the performance in large network. We have taken a
network of 50 nodes in which node movements are ran-
dom. Initially the network is as shown in Figure 3. The
simulation parameters are presented in Table 3. Three
CBR flows with packet size of 100 0 bytes are estab lished
between Nodes 37 and 18, Nodes 19 and 47 and Nodes
29 and 7. Propagation model used as shadowing model
with path loss exponent varied from 1.6 to 2.8. The node
movements are random and the speed of mobile node
varied from 4 - 20 m/s.
Performances are analyzed for these two cases:

T. Sangeetha ET AL.
615
Figure 3. Large network with 50 nodes.
Table 3. Simulation Parameters Used for large network.
Parameter Va lue
Simulation area size 500 m × 500 m
Number of nodes 50
Node movement Random
Node Speed when mobile 4 - 20 m/s
Node pause time when mobile 50 s
Number of traffic source 3
Simulation time 200 s
Traffic source type CBR
Propagation model Shadowing
Path loss exponent 1.6 to 2.8
 Finding backup routes by using Node Disjoint Paths
termed in this paper as QoSAR Node Disjoint.
 Finding backup routes by using Link Disjoint Paths
termed in this paper as QoSAR Link Disjoint.
The following graphs show the performance of QoSAR
protocol.
The Figure 4 is clearly shows that the Packet Delivery
Ratio (PDR) higher for proposed protocol (QOSAR) com-
pared to AODV protocol. AODV protocol achieved low
PDR due to link failure. This link failure can be achieved
in our prop os ed protocol by the use of backup ro utes.
From Figure 5, the throughput assurance reliability
achieved for proposed protocol increases with the speed
of mobile nodes. It has been observed that QoSAR Node
Disjoint and Link Disjoint protocol exhibits higher
throug hp ut than AOD V protocol .
Still considering the Figure 6 QoSAR not only im-
proves PDR and Throughput but it also reduces the delay
considerably. From Figure 7, we can decide the Loss
ratio also increases with increasing speed of mobile nodes
in AODV protocol. Our proposed protocol gives much
better performance compared to AODV.
Figure 4. Speed vs Packet Delivery Ratio.
Figure 5. Speed vs Throughput.
Figure 6. Speed vs Delay.

T. Sangeetha ET AL.
616
Figure 7. Speed vs Loss Ratio.
Figures 5, 6 and 8 confirm that the protocol proposed
in this paper, in general, able to uphold throughput guar-
antees more in shadow fading environments. We can
overcome route failure using backup routes. Compari-
sons show that QoSAR protocol provides better quality
of service.
In this scenario we consider constant number of mo-
bile nodes only variation in path loss exponential and the
results obtained for packet delivery ratio. The Figure 8
shows that effects on the PDR and path loss exponential.
As shown in Figure 8. When increasing the path loss
exponent the packet delivery ratio deceased to zero by
both proposed and AODV protocol. Considering another
case we can obtain high the packet delivery ratio for
QoSAR protocol.
The simulation para meters for varia ble nodes show n in
Table 4 and simulation results for variable nodes are
shown in Table 5. In these simulation there are 10 - 50
mobile hosts randomly distributed in 100 × 100 meters.
As per the performance analysis when increasing the
number of mobile nodes in shadow fading environment
the QOSAR disjoint paths achieve better performance
compared to AODV. (i.e) considering the number of mo-
bile node as 20 the PDR is achieved 95.2165 for AODV
protocol and 99.8169 is achieved for QOSAR disjoint
paths. While increasing the mobile nodes the source and
destination has multiple paths to transmit and receive
packets. So we can achieve higher throughput for QO-
SAR disjoints paths.
6. Conclusion
In MANET, it is difficult to provide QoS assur ances due
to node mobility, contention for channel access, lack of
centralized coordination etc, as discussed previously. Si-
mulation shows that backup routes improves Throughput
and Packet Delivery Ratio ( PDR) and also reduces Delay
Figure 8. Path loss exponent vs Packet Delivery Ratio
(PDR).
Table 4. Simulation Parameters for variable nodes.
Parameter Value
Simulation area size 100 m × 100 m
Number of nodes 10 - 50
Number of traffic source 1
Simulation time 100 s
Traffic source type TCP
Propagation model Shadowing
Path loss exponent 2
Table 5. Simulation Results for variable nodes.
Number of
nodes Parameter AODV QoSAR (Node disjoint Paths)
10
Packets sent 36,089 54,879
Packets received 32,829 54,609
PDR 90.9668 99.508
Throughput 4.78226 × 107 1.21047 × 108
20
Packets sent 51,448 66,644
Packets received 48,987 66,522
PDR 95.2165 99.8169
Throughput 1.01773 × 108 1.79061 × 108
30
Packets sent 50,831 64,862
Packets received 48,630 64,740
PDR 95.67 99.8119
Throughput 1.002 × 108 1.69629 × 108
40
Packets sent 54,190 66,269
Packets received 51,617 66,158
PDR 95.2519 99.8325
Throughput 1.13166 × 108 1.77077 × 108
50
Packets sent 53,503 61,604
Packets received 51,083 61,512
PDR 95.4769 99.8507
Throughput 1.10681 × 108 1.53093 × 108

T. Sangeetha ET AL.
617
and loss ratio in shadow fading environments. As per the
performance analysis the QoSAR protocol gives better
performance in shadow fading environments.
However, it was found that with severe shadowing in-
duced signal strength fluctuations, the backup routes was
less significant, although merely proactively seeking back-
up routes still improved to achieve QoS. This suggests
that routing protocols benefit from proactively requiring
that backup routes exist. However, the more than one
backup route is counter-productive due to the excess
overhead incurred when initiating state information setup.
The required level of link disjointness between data ses-
sions’ primary and backup routes was also studied [12].
Thus it is suggested that by using node disjoint and link
disjoint paths, we can achieve better Quality of Service
(QoS) in shadow fading environments.
REFERENCES
[1] R. Ramanathan, J. Redi and B. Technologies, “A Brief
Overview of Ad-hoc Networks: Challenges and Direc-
tions,” IEEE Communications Magazine, Vol. 40, No. 5,
2002, pp. 20-22.
http://dx.doi.org/10.1109/MCOM.2002.1006968
[2] R. Shi and Y. Deng, “An Improved Scheme for Reducing
the Latency of AODV in Mobile Ad Hoc Networks,” 9th
International Conference for Young Computer Scientists,
IEEE Computer Society, 2008, pp. 594-598.
[3] H.-W. Tsai, T.-S. Chen and C.-P. Chu, “An On-Demand
Routing Protocol with Backtracking for Mobile Ad Hoc
Networks,” WCNC IEEE Communication Society, 2004,
pp. 1557-1561.
[4] G. Di Caro, F. Ducatelle and L. M. Gambardella, “Special
Issue on Self-Organization in Mobile Networking An-
tHocNet: An Adaptive Nature-Inspired Algorithm for
Routing in Mobile Ad Hoc Networks”, European Trans-
actions on Telecommunications, Vol. 16, 2005, pp. 443-
455.
http://dx.doi.org/10.1002/ett.1062
[5] R. Bai and M. Singhal “DOA: DSR over AODV Routing
for Mobile Ad Hoc Networks,” IEEE Transaction on
Mobile Computing, Vol. 5, No. 10, 2006, pp. 1403-1416.
[6] L. Hanzo II. and R. Tafazoli, “A Survey of QoS Routing
Solutions for MANETs,” IEE E Communications Surveys
and Tutorials, Vol. 9, No. 2, 2007, pp. 50-70.
http://dx.doi.org/10.1109/COMST.2007.382407
[7] L. Hanzo II. and R. Tafazolli, “Admission Control
Schemes for 802.11-Based Multi-Hop Mobile Ad-hoc
Networks: A Survey,” IEEE Communications Surveys &
Tutorials, Vol. 11, No. 4, 2009.
[8] C. E. Perkins and E. M. Royer, “Ad Hoc On-Demand
Distance Vector Routing (AODV),” IETF RFC 3561,
2003.
[9] C. E. Perkings, E. M. Roy er and S. R. Das, “Performance
Comparison of Two On-Demand Routing Protocols for
Ad Hoc Networks,” IEEE Personal Communications,
Febuary 2001.
[10] L. Chen and W. B. Heinzelman, “QoS-Aware Routing
Based on Bandwidth Estimation for Mobile Ad Hoc
Networks” IEEE Journal on Selected Areas in Commu-
nications, Vol. 23, No. 3, March 2005.
[11] D. B. Johnson and D. A. Maltz, “Dynamic Source
Routing in Ad Hoc Wireless Networks,” Mobile Compu-
ting, T. Imielinski and H. Korth, Eds., Kluwer Academic,
Vol. 353, 1996, pp. 153-181.
[12] L. Hanzo and R. Tafazolli, “QoS-Aware Routing and
Admission Control in Shadow-Fading Environments for
Multirate Manets,” IEEE Transactions on Mobile Com-
puting, Vol. 10, No. 5, 2011.
http://dx.doi.org/10.1109/TMC.2010.208
[13] L. Hanzo and R. Tafazolli, “A Survey of QoS Routing
solutions for Mobile Adhoc Networks,” IEEE Communi-
cations Surveys and Tutorials, Vol. 9, No. 2, 2007, pp.
50-70.
http://dx.doi.org/10.1109/COMST.2007.382407