AMTR: THE ANT BASED QOS AWARE MULTIPATH TEMPORALLY ORDERED ROUTING ALGORITHM FOR MANETS

Jan Zizka (Eds) : CCSIT, SIPP, AISC, PDCTA - 2013
pp. 389–396, 2013. © CS & IT-CSCP 2013 DOI : 10.5121/csit.2013.3644
AMTR: THE ANT BASED QOS AWARE
MULTIPATH TEMPORALLY ORDERED
ROUTING ALGORITHM FOR MANETS
Debajit Sensarma 1
and Koushik Majumder 1
1
Department of Computer Science & Engineering, West Bengal University of
Technology, Kolkata, INDIA
debajit.sensarma2008@gmail.com,
koushik@ieee.org
ABSTRACT
Mobile Ad hoc networks (MANETs) are self organized by a collection of mobile nodes, which
are interconnected by multi hop wireless paths. Providing QoS support in MANETs is an active
research area, basically for supporting real time applications that are based upon interaction
between the routing scheme and QoS provisioning. The goal of QoS aware routing is to achieve
more deterministic behavior of the network i.e. a stable path, so that information carried by the
network can be delivered in an uninterrupted fashion and for better resource utilization. In this
paper we have developed AMTR, a new ant based QoS aware on-demand multipath routing
algorithm for MANETs with better route failure management and which will be highly adaptive
and energy efficient.
KEYWORDS
MANET, Ant Colony Optimization, TORA, QoS Routing.
1. INTRODUCTION
With the growth of internet, the demand for real time and quality of services (QoS) of mobile ad
hoc network has increased. So, demand of QoS-aware routing in case of mobile ad hoc networks
has also increased. The major objectives of QoS-aware routing are:-i) finding path from source to
destination satisfying user’s requirement. ii) optimizing network resource usage and iii) repairing
or re-computing the path quickly in case of path break or link failure or unwanted things like
congestion, without degrading the level of QoS. Again, centralized algorithms have scalability
problems, static algorithms have trouble keeping up-to-date with network changes, and other
distributed and dynamic algorithms have oscillations and stability problems. Ant based routing
[1-4] provides a promising alternative to these approaches. Ant colony provides a number of
advantages [5] due to the use of mobile agents and stigmergy (a form of indirect communication
used by ants in nature to coordinate their problem-solving activities). It provides scalability, i.e.
according to the network size the population of the agents can be adapted. It is fault tolerant, i.e. it
does not rely on a centralized control mechanism. Therefore, node mobility or link breakage does
not result in catastrophic failure. Besides this, it provides adaptation, where according to the
network changes agents can change, die or reproduced. Speed is achieved, because network
topology change can be propagated very fast. Agent acts individually, independent of other
network layers, so it gives modularity. Also, autonomy is provided because little or no human

390 Computer Science & Information Technology (CS & IT)
supervision is required. It also provides parallelism, i.e. agents operate in parallel. However,
besides these advantages, one of the biggest difficulties with ant colony algorithms applied in
network routing area is that multiple constraints often make the routing problem intractable [6].
The proposed algorithm is a QoS aware multipath on-demand routing algorithm. The advantage
of AMTR is that, it supports energy efficient multipath routing and as well as takes care of the
QoS constraints-delay, bandwidth, energy and drain rate which is very essential for real time and
multimedia applications.
The paper is organized as follows: Section 2 describes the mathematical model. In section 3 the
proposed algorithm combining the idea of ACO and TORA is illustrated. Section 4 explains the
performance analysis. Finally section 5 concludes the paper.
2. MATHEMATICAL MODEL
For mathematical analysis MANET is represented by a connected undirected graph. Let G (V, E)
represents the mobile ad hoc network. Here V denotes the set of network nodes and E denotes the
set of bidirectional links. QoS metrics with respect to each link e ∈E is delay (e) and bandwidth
(e). With respected to node n ∈V , it is delay (n), energy (n) and drain rate (n) which is the
energy dissipation rate of node ‘n’. Another QoS metric considered here is hop count. It is
important because multiple hops are used for data transmission in MANET. So, it is necessary to
find paths with minimum hops. The main motivation of this proposed algorithm is to find path
from source to destination which will satisfy the QoS requirements such as delay, bandwidth,
energy, drain rate and hop count.
Let, path (i, j) or R is entire path from node i to j where QoS constraints have to satisfied.
From an arbitrary node i to an arbitrary node j, delay, bandwidth, energy, drain rate and hop count
is calculated as-
delay (path (i, j )) or D (R) = ∑ ∈ ),(
)(jiPe
edelay + ∑ ∈ ),(
)(jiPn
ndelay
where, delay (path (i,j )) is the transmission and propagation delay of the path(i,j) and delay (n) is
the processing and queuing delay of node ‘n’ on path(i, j).
bandwidth(path(i,j)) or B(R)= ),(min jiPe∈ {bandwidth(e)}
where, bandwidth (e) is the available bandwidth of that link on path(i, j).
energy (path (i, j)) or E (R) = ),(min jiPn∈ { energy (n)}
where, energy (n) is the residual energy of node ‘n’ on path(i, j).
drain rate (path (i, j)) or DR (R) = ),(max jiPn∈ { drain rate(n)}
where, drain rate (n) is the rate of energy dissipation of node ‘n’ on path(i, j).
hop count (path (i, j)) or HC (R) = Number of nodes in the path.
2.1 CALCULATION OF PHEROMONE
Ant deposits pheromone during traversal of the link for finding a route. The quantity of
pheromone it deposited on each link (i, j) along the route R is noted by ji,τ∆ and it is a function
of global quality of route R. It is expressed by the following equation-
ji,τ∆ =
DRHCD
EB
λλλ
λλ
DR(R)HC(R)D(R)
E(R)B(R)
++
+
(1)

Computer Science & Information Technology (CS & IT) 391
Here λB, λE, λD, λHC and λDR are the weight factors which indicate the relative significance of the
QoS parameters during pheromone update on path (i, j). The quantity of the deposited
pheromone is defined only after finding the route.
The pheromone quantity of the link (i, j) is updated according to the following equation:
τi,j = σ.τi,j + ∆τ i,j (2)
2.2 CALCULATION OF PATH PREFERENCE PROBABILITY
Path Preference Probability is calculated in each intermediate node as well as source node upon
receiving of QRY Reply_Ant.
Suppose current node i receives QRY Reply_Ant from node j for destination d, then the Path
Preference Probability is calculated as-
∑∈
=
iNk
α6
ijd
α5
ikd
α4
ikd
α3
ikd
α2
ikd
α1
ik
α6
ijd
α5
ijd
α4
ijd
α3
ijd
α2
ijd
α1
ij
ijd
][DR].[E].[B].[η].[D][τ
][DR].[E].[B].[η].[D][τ
P (3)
Here α1, α2, α3, α4, α5 and α6 are the tunable parameters which control the relative weights of
pheromone trails, hop count, bandwidth, energy and drain rate respectively.
Ni is the set of neighbors of i and k is the neighbor node of i through which a path to destination is
known.
The relative metrics are calculated from source i to destination d via j as-
Dijd
d))(i,delay(path
1
=
ijdη
d))ath(i,hopcount(p
1
=
Bijd= bandwidth (path(i,d))
Eijd = energy (path (i,d))
ijdDR
d))i,rate(path(drain
1
=
Now, source as well as neighbors has multiple paths from source to destination. The path with
higher Path Preference Probability is selected for the data transmission.
3. PROPOSED ALGORITHM
The proposed algorithm is a multipath on-demand routing algorithm. Here QoS constraints-delay,
bandwidth, energy and drain rate are considered which are very essential for multimedia
applications.
This algorithm has three phases namely route discovery phase, route maintenance phase and
route erasure phase. In route discovery phase, multiple paths which satisfy the required QoS
constraints are created and stored in the cache. The route with higher path preference probability
is selected for the routing. In route maintenance phase, when a node fails to transmit packets, it
will check for the alternate route to the desired destination with better path preference probability.
If such path exists, it is chosen for the routing. In the next case, if the failure node has no
outbound link to other nodes in the network, then it will send an error message to the source and

immediately start local link maintenance. All routes going through the failure node are deleted
from the cache. In the next step, when source finds the error message, it will check for other
alternate route in the cache with next better path preference probability. If it exits, that route gets
selected for routing. Otherwise, a new route discovery phase is started. If network partition is
detected during route maintenance phase, then route erasure phase is invoked and all invalid
routes are deleted from cache.
Each node k∈ N has a height represented by as a quintuple: Hi= ( kτ ,oidk, rk , kδ , k ), where i) kτ :
A time tag indicating the time of link failure represents the reference level. ii) oidk: Originator-id,
id of node that defined the reference level. iii) rk: 1 bit used to divide each reference level into 2
sub-levels. iv) kδ : Integer used to order nodes with respect to a unique reference level and v) k:
Unique identifier of the node. Each node k (other than the destination) maintains a link-state array
with an entry LSk.j for each link (k, j) ∈ L, where j is a neighbor of k, L is total number of links.
The height of node k, Hk and that of its neighbor j, HNk,j determines the direction of the links, and
is directed from the higher node to the lower node. RRk is route-required flag of node k.
3.1. ROUTE DISCOVERY PHASE
Algorithm 1: Route Discovery
BEGIN
Step 1: Let, Source(S) wants to communicate with the destination (D) with the QoS constraints
bandwidth, delay, drain rate, energy. No path exists in the cache previously.
Step 2: S sets its height as NULL initially i.e (-,-,-,-, S).
Step 3: S sends a QRY Request_Ant packet to its neighbors and if the destination is not one hop
away from the intermediate node then node also sets its height as NULL and broadcasts the
packet.
Step 4: The QRY Request_Ant packet collects the information about bandwidth, delay, energy
and drain rate during traversal.
Step 5: When an intermediate node k receives a QRY Request_Ant, it has four options:
5.1.If node k has no downstream links and RRk is unset, it rebroadcasts the QRY
Request_Ant packet and sets RRk.
5.2.If node k has no downstream links and RRk is set, it discards the QRY Request_Ant
packet.
5.3.If node k has at least one downstream link and its height is NULL, it sets its height to Hk
= min {Hj | j∈Nk} + {0, 0, 0, 1, 0} and broadcasts a QRY Reply_Ant packet.
5.4.If node k has at least one downstream link and its height is non-NULL, and if a QRY
Reply_Ant packet has been broadcast since the link over which the QRY Request_Ant
packet was received became active, it discards the QRY Request_Ant packet. Otherwise
it broadcasts a QRY Reply_Ant packet. Also, if RRk is set when a link becomes active, it
broadcasts a QRY Request_Ant packet.

Step 6: QRY Reply_Ant follows the route corresponding to the QRY Request_Ant but in reverse
order.
Step 7: When intermediate or source node k receives an QRY Reply_Ant from a neighbor j, k
updates HNk,j to reflect the height of node j received in the QRY Reply_Ant and performs one
activity from the two options:
7.1.If RRk is set (implying the height of node k is NULL), node k sets Hk = min {Hj | j∈Nk} +
{0, 0, 0, 1, 0}, updates the link LSk, unsets RRk and broadcasts a QRY Reply_Ant packet
with the new information.
7.2.If RRk is unset, node k updates the links in LSk.
Intermediate nodes also update the pheromone table according to equation (2) and
calculate the path preference probability according to equation (3) and also update the
probability table, and set the lower level node as downstream node.
Step 8: Source also calculates the Path Preference probability.
Step 9: If the calculated path preference probability is better than the requirements then the path
is accepted and stored in the cache.
Step 10: The path with the better Path Preference Probability is selected for data transmission.
END
3.2. ROUTE MAINTENANCE PHASE
Algorithm 2: Route Maintenance
BEGIN
Here 2 cases can occur:
Case 1:When an intermediate node k detects the route failure, if it has an outbound link
(Hk> HNk,j) i.e. another unexpired route exists in the cache, the route with better Path Preference
Probability is selected for routing.
Case 2:When an intermediate node detects the route failure, if it has no outbound link the
following sequence of steps are performed:
Step 1: The node generates an Error packet and sends to the source node and all intermediate
nodes delete routes from their cache which contains that node.
Step 2: As soon as source receives the error packet, it deletes routes from the cache which
contain the node and if an unexpired route exists in the cache, the route with next better
Path Preference Probability is selected for routing.
Step 3: After sending the Error packet, the node where link failure occur, starts local link
maintenance according to the following conditions:
a. Generate: Node k has no downstream links due to link failure.
( kτ ,oidk, rk) = (t, k, 0), where t is the time of failure.

( kδ , k) = (0, k).
Node k defines a new reference level, if k has no upstream neighbors, it sets Hk = NULL.
b. Propagate: Node k has no downstream links due to link reversal follow-
ing receipt of an UPD and the ordered sets ( kτ ,oidk, rk) are not equal for all j ∈ Nk.
( kτ ,oidk, rk) = max {( kτ ,oidk, rk) | j∈Nk}
( kδ ,k) = {min { jδ | j∈Nk with ( jτ ,oidj, rj) = max {( jτ ,oidj, rj) } - 1, k}.
Node k propagates the reference level of its highest neighbors and chooses a reference
lower than all neighbors of that reference level.
c. Reflect: Node k have no downstream links due to link reversal following
receipt of an UPD packet and the sets ( jτ ,oidj, rj) are equal with rj =0 for all j∈Nk.
( kτ ,oidk, rk) = ( jτ ,oidj, 1)
( kδ , k) = (0, k).
Node k reflects back the reference level by setting the r bit.
d. Detect: Node i has no downstream links due to link reversal following receipt
of an UPD packet and the sets ( jτ ,oidj, rj) are equal with rj=1 for all j∈Nk and oidk=k.
( kτ ,oidk, rk) = (_, _, _)
( kδ , k) = (_, k).
Node k has detected a partition and route erasure phase invoked (described below).
e. Generate: Node k has no downstream links due to link reversal following
receipt of an UPD packet and the sets ( jτ ,oidj, rj) are equal with rj=1 for all j ∈Nk and
oidj∈k.
( kτ ,oidk, rk) = (t, k, 0), where t is the time of failure.
( kδ , k) = (0, k).
Node k experienced a link failure between the time it propagated a reference level and the
reflected higher sub-level returned from all neighbors. This link failure required no
reaction.
Step 4: Finally, after performing step 2, step 3, if the cache contains no unexpired routes, again a
new route discovery phase is started.
END
3.3 ROUTE ERASURE PHASE
Route erasure is initiated upon the detection of a partition. Node k sets its height and the height
entry for each j ∈Nk to NULL. However, if the destination is a neighbor, the corresponding height
array is set to ZERO. Node k then updates its link-state array LS, and broadcasts a CLR packet
that consists of a destination id, did and the reflected sub-level of node k, ( kτ ,oidk, 1). When a
node k receives a CLR packet from a neighbor j∈Nk, it reacts according to the following steps of
Algorithm 3:

Algorithm 3: Route Erasure
BEGIN
Step 1: If the reference level in the CLR packet matches that of node k, it sets its height and the
height entry for each neighbor j∈Nk to NULL (unless the destination is a neighbor, in which
case the corresponding height array is set to ZERO), updates all the entries in its link-state
array LS, and broadcasts a CLR packet.
Step 2: If the reference level in the CLR packet does not match that of node k, it sets the height
entry for each neighbor j ∈Nk (with the same reference level in the CLR packet) to NULL and
updates the corresponding link-state array entries.
Step 3: Lastly, the hop in error is removed from the node's route cache and all routes containing
the hop are eliminated.
END
4. PERFORMANCE ANALYSIS
AMTR is an adaptive routing algorithm. So, it is suitable for the network where node mobility is
higher and no centralized control exists. It takes the advantages of both TORA and Ant Colony
Optimization technique.
This algorithm finds multiple paths between the source and the destination. Again, ACO always
does not find the shortest path. Rather at the time of path set up, it also takes care of the QoS
requirements and for this, link quality is improved. Besides this, if multiple paths exist and as ants
store paths, which satisfies the QoS requirements, losing any one path cannot affect the
communication. This is extremely necessary for the real time and multimedia communication.
Use of multi path routing also increases the packet delivery ratio, decreases the packet loss rate. It
also utilizes the bandwidth properly and for this, throughput and network stability or lifetime
increase.
AMTR’s design is aimed at minimizing the aggregate bandwidth by minimizing the control
packets. It also minimizes the communication overhead by localizing algorithmic reaction to
topological changes. Especially in case of link failure, the number of nodes that must participate
in the reaction is minimized in comparison with the other routing algorithms. In the first phase of
route maintenance of AMTR, if link failure occurs and at that time a downstream link exists then
only the route with valid QoS requirements is selected for routing. Thus, it minimizes the number
of control packets and consequently, the available bandwidth utilized properly.
5. CONCLUSION
An ant based QoS aware multipath routing algorithm AMTR is proposed in this paper which
supports real time and multimedia applications. It is more adaptive and provides energy efficient
routing by considering node’s remaining energy as well as drain rate (i.e. energy dissipation rate)
as one of the QoS parameters. In high mobility cases it is very efficient in terms of quick route
maintenance. This algorithm takes care of end to end delay, available bandwidth, and hop count
as QoS parameters which increase the network throughput. It is a multipath routing which
increases network stability. AMTR reduces the number of control packets and utilizes the
bandwidth properly which is also very efficient especially in case of dense networks.

ACKNOWLEDGEMENTS
The authors would like to thank West Bengal University Technology, Kolkata, India for the
supports and facilities provided to carry out this research. The authors also thank the reviewers
for their constructive and helpful comments.
REFERENCES
[1] P., Deepalakshmi, S., Radhakrishnan, (2009), “Ant Colony Based QoS Routing Algorithm for Mobile
Ad Hoc Networks”, International Journal of Recent Trends in Engineering, Vol. 1, No. 1, pp. 459-
462.
[2] S. B., Wankhade, M. S., Ali, (2011), “Route Failure Management Technique for Ant Based Routing
in MANET”, International Journal of Scientific & Engineering Research, Vol. 2, No. 9.
[3] R., Asokan, A. M., Natarajan, C., Venkatesh, (2008),“Ant Based Dynamic Source Routing Protocol to
Support Multiple Quality of Service (QoS) Metrics in Mobile Ad Hoc Networks”, International
Journal of Computer Science and Security, Vol. 2, No. 3.
[4] S., Kannan, T., Kalaikumaran, S., Karthik, V. P., Arunachalam, (2010), “Ant Colony Optimization for
Routing in Mobile Ad-Hoc Networks”, International Journal of Soft Computing, Vol. 5, No. 6, pp.
223-228.
[5] I., Kassabalidis, M. A., El-Sharkawi, R. J., Marks, P., Arabshahi, A. A., Gray, (2002), “Swarm
Intelligence for Routing in Communication Networks”, IEEE Globecom.
[6] S. J., Mirabedini, M., Teshnehlab, M. H, Shenasa, A., Movaghar, A. M., Rahmani, (2008), “AFAR:
Adaptive fuzzy ant based routing for communication networks”, Journal of Zhejiang University
Science A, Vol. 9, No. 12, pp. 1666-1675.
[7] D., Sensarma, K., Majumder, (2012), “A Comparative Analysis of the Ant Based Systems for QoS
Routing in MANET,” Recent Trends in Computer Networks and Distributed Systems
Security,Communications in Computer and Information Science, Volume 335, pp 485-496.
Authors
Debaji Sensarma has received his B.Sc degree in Computer Science in the year 2009
from university of Calcutta, Kolkata, India and M.Sc degree in computer Science in
the year 2011 from West Bengal State University, Kolkata, India. He obtained his
M.Tech. degree in Computer Science and engineering from West He is now pursuing
PhD from the department of Computer Science and Engineering, University of
Calcutta, Kolkata. He has published several papers in International journals and
conferences.
Koushik Majumder has received his B.Tech and M.Tech degrees in Computer
Science and Engineering and Information Technology in the year 2003 and 2005
respectively from University of Calcutta, Kolkata, India. He obtained his PhD degree
in the field of Mobile Ad Hoc Networking in 2012 from Jadavpur University,
Kolkata, India. Before coming to the teaching profession he has worked in reputed
international software organizations like Tata Consultancy Services and Cognizant
Technology Solutions. He is presently working as an Assistant Professor in the Dept.
of Computer Science & Engineering in West Bengal University of Technology,
Kolkata, India He has published several papers in International and National level journals and conferences.
He is a Senior Member, IEEE.

AMTR: THE ANT BASED QOS AWARE MULTIPATH TEMPORALLY ORDERED ROUTING ALGORITHM FOR MANETS

More Related Content

What's hot (19)

Similar to AMTR: THE ANT BASED QOS AWARE MULTIPATH TEMPORALLY ORDERED ROUTING ALGORITHM FOR MANETS (20)

More from cscpconf (20)

Recently uploaded (20)

AMTR: THE ANT BASED QOS AWARE MULTIPATH TEMPORALLY ORDERED ROUTING ALGORITHM FOR MANETS