The Comparison between Three Routing Algorithms, AODV,

HOPNET, and PAR

Hossein Motaharinejad
Department of Computer Science, Shahid Bahonar University of Kerman, Kerman, Iran
[Link]@[Link]

Neda Kazeminejad
Department of Computer Science, Shahid Bahonar University of Kerman, Kerman, Iran
[Link]@[Link]

Abstract
Development of technology and machinery of human life is that in recent years calls for
universal access to computer networks. There are items such as mobile phones, tablets, and
laptops that are constantly moving and shifting. They have made routing and Serving even
more crucial. Mobile Ad Hoc Networks (MANETs) are a special type of decentralized
wireless networks without the need for any pre-existing infrastructure for communication
among the nodes and for this reason has become the focus of attention. In recent years, an
optimized routing protocol for this type of network has been a challenge for researchers. The
inspiration of nature has been associated with routing techniques enables the revolutionary
which the Ant colony due to its high position in the dynamic nature of routing is important.
In this article, the three routing protocol using is presented. Two ant colony based protocols,
PAR and HOPNET, will be compared with AODV. The goal of this comparison is comparing
three routing algorithm which are from different groups, reactive (AODV), proactive (PAR)
and hybrid (HOPNET). The results of comparing the three methods show the properties of
each protocol and its specifications based on usage.

Keywords: Mobile ad-hoc network, Ant colony optimization algorithm, Ad Hoc On-Demand
Distance Vector Routing (AODV), Probabilistic Ant Routing (PAR), HOPNET

1. Introduction
Wireless networks are classified as: Infrastructure-based wireless network and Infrastructure-less
wireless network. In an Infrastructure based wireless nodes connect to an external network like the
Internet or Intranet with the help of an access point. On the other hand an Infrastructure-less network
is a network in which mobile nodes communicate with each other through wireless links, such a
network is also known as an Ad-hoc network. For example, two laptops with wireless adapter cards
can set up an Ad-hoc network. An Ad-hoc network can be further classified as MANET, WSN
(Wireless Sensor Network), WMN (Wireless Mesh Network) (Cadger et al, 2013; J. P. Singh and
Gupta, 2013). MANET is a temporary network in which mobile nodes communicate without any aid
of centralized administration and may operate in either stand-alone fashion or connected to the
Internet. Because of features like quick deployment and easy to use makes them useful in a wide

1
number of applications like Military operations, emergency rescue operations and wireless sensor
network, etc.
In MANETs, routing is needed to find a path from source to destination which is done with the help of
routing protocols (J. P. Singh and Gupta, 2013). The nodes communicate through single-hop or
multihop paths in a peer-to-peer fashion by using wireless media (Budyal and Manvi, 2014). The
wireless communication devices are transmitters, receivers and smart antennas. These antennas can
vary in many types while nodes can be either fixed or mobile (Al-Ani, 2011). The nodes of the
MANET operate as end hosts as well as routers. Due to the mobility of the nodes, routing path is
affected by addition and deletion of nodes. Hence, the topology of the network may change rapidly
and unexpectedly (Budyal and Manvi, 2014). Because of the lack of centralized control, routing
becomes a central issue and a major challenge as the network topology is constantly changing. It is a
collection of mobile nodes that are dynamic and arbitrarily located in such a manner that the
interconnections between nodes are capable of changing on a continual basis (Gupta et al, 2012). The
vast amount of research is going on to make such a routing protocol, which can work in a changing
topology environment of a Hybrid Ad-hoc network (J. P. Singh and Gupta, 2013).
Routing algorithms in mobile Ad-Hoc Networks (MANET’s) should match the special characteristics
of MANETs. It should be self-configured, self-built and distributed routing algorithm. In addition, it
should be able to optimize more than one quality of service parameter (routing overhead, latency,
packet delivery ratio and etc.) to achieve better network performance. The remaining energy in each
node across the network should be fairly distributed. To solve the routing problem in MANETs, the
goal of the routing algorithm should be achieved, while taking into consideration the special network
characteristics such as mobility, limited energy, unfair energy usage distribution, limited bandwidth
and high bit error rate. The routing algorithm should deal with the rapid change in the network and it
should optimize more than one quality of service parameters (QOS) in the network. Many routing
algorithms have been developed for MANETs (Hussein and Saadawi, 2003). Depending upon the
mechanism and functionalities MANET routing protocols can be classified into three categories: the
table driven or proactive, on-demand or reactive and hybrid routing protocols (Sofat and Gupta, 2013).
Basically, Routing is the process of choosing paths in a network along so that the source can send data
packets towards the destination. Routing is an important phase of network communication because the
characteristics like throughput, reliability, packet delivery, congestion and so on depends on the
routing information (G. Singh et al, 2012).
Recently, a new family of algorithms emerged, inspired by swarm intelligence, which provides a novel
approach to distributed optimization problems. The expression “Swarm Intelligence” defines any
attempts to design algorithms inspired by the collective behavior of social insect colonies and other
animal societies. Ant colonies, bird flocking, animal herding and fish schooling are examples in nature
that use swarm intelligence. Several algorithms which are based on ant colony were introduced in
recent years to solve the routing problem in mobile ad hoc networks (Marjan Kuchaki Rafsanjani et al,
2010).
Swarm Intelligence based Routing with Opportunistic Routing represent sets of algorithms based on
biological models, inspired by highly dynamic environments, which are particularly adequate for
MANETs (Correia and Vazão, 2010). Ant Colony Optimization (ACO) is a subset of Swarm
Intelligence. It is introduced by Marco (1992) (G. Singh et al., 2012). The main idea of the ant colony
optimization is taken from the food searching foraging behavior of real ant colonies (Yoo and Han,
2013; Zhang and Feng, 2012). When ants are on the way to search for food, they start from their nest
and walk toward the food (Destination). When an ant reaches an intersection, it has to decide which
branch to go. While going, ants deposit a chemical substance named pheromone, which ants are able
to sense, which marks the route taken, and they are attracted to the marked paths. The concentration of
the pheromone on a certain path is an indication of its usage. The more pheromone that is deposited on
a path, the more attractive that path becomes. With time, the concentration of pheromone decreases
due to volatile effects. Evaporation clears the pheromone on longer paths as well as on less interesting
paths. The shortest paths are refreshed more quickly with ants, thus having the chance of being more
frequently explored. Intuitively, ants will converge towards the most efficient path due to the fact that
it gets the strongest concentration of the pheromone (Asokan et al, 2008; Ilie and Bădică, 2013).

2
Due to MANETs suffers from limited resources, therefore two parameters, routing overhead and end-
to-end delay, are important to protect resources and real-time connectivity. The goal of this paper is to
compare three routing protocol based on ant colony optimization and then explain the differences in
four parameters: routing overhead, packet delivery ratio, average end-to-end delay, and throughput.
The organization of this paper is as follows. In next section II, we give a brief overview about the
classification of Ad-Hoc routing protocols as On Demand, Table Driven and Hybrid Protocols.
Section III explains about some related works, in section IV we will be explain about the three
algorithms, AODV, HOPNET, and PAR, section V explain simulation and performance evaluation
and finally conclusion are given in section VI.

2. CLASSIFICATION OF AD HOC ROUTING PROTOCOLS

Multiple routing protocols have been developed for Mobile Ad hoc Networks. Depend on the
mechanism and functionalities MANET routing protocols can be classified into three categories (Sofat
and Gupta, 2013):
 Proactive (Table-driven)
 Reactive (On-demand)
 Hybrid

2.1. Proactive protocols

In such this approach, each node in the network holds one or more routing table which stores
information of next hops/nodes. All nodes keep on updating these tables periodically. The drawback of
this routing causes more overhead, consumption of more Bandwidth, not used for large network, If the
network topology changes too frequently, might be very high cost of maintaining the network,
information about actual topology might even not be used if network activity is low (Goyal and Gaba,
2013). The differences among the protocols lies in their routing table structure, number of tables,
updating frequency, use of control messages and the presence of a central node (Dhanda and
Chaudhry, 2013). We forward the packets irrespective of when and how routes are desired as there is
always the availability of the routes in the continuously updated routed tables (Talwar and Gupta,
2012). The main disadvantages of table-driven QoS are bandwidth consumption in transmitting,
routing tables and also saving the table of the routes that are not used in future (Reddy et al, 2006).

2.2. Reactive protocols

Reactive protocols are elected when we want to set up routes on demand. This route will be
established by the routing protocol in the situation when any node wants to initiate the communication
with another node to which it has no route. This type of protocol is generally based on broadcasting
the network with Route Request (RREQ) and a Route Reply (RREP) messages (Talwar and Gupta,
2012). Advantage of reactive protocols is less control overhead as compared to proactive or even
hybrid protocols for Mobile Ad Hoc Networks. Thus, reactive routing protocols have better scalability
than proactive routing protocols in mobile ad hoc networks (Dhanda and Chaudhry, 2013). The main
disadvantages of the on-demand approach are Dilation when the source node trying to find a route to
the destination node and excessive flooding can be led to the network clogging (Chen et al, 2004).

2.3. Hybrid protocols

Hybrid routing protocols synthesis the advantages of proactive and reactive routing. The proactive
tactic is used to discover and maintain routes to nearer nodes, while routes for far away nodes are
discovered reactively. In an Ad-Hoc network, a Hybrid routing algorithm can be implemented in a
hierarchical network architecture. The performance of the network depends on the distribution of the
proactive/reactive approaches for each level of the network hierarchy (Dhanda and Chaudhry, 2013).
The hybrid routing protocol disadvantage is that the nodes have to maintain high level topological
information which cause more memory usage and power consumption (J. P. Singh and Gupta, 2013).

3
3. Related work

One of the most important section in routing process is finding the best path with the lowest latency
and distance. Using an ant colony algorithm made a great impact on network routing so that it made
the new routing algorithm better than past. In order to appreciate the apprehension of this work, let us
review some of the most relevant proposals reported in the MANETs routing problem which used ant
colony approach.
In 1997 one of the first Routing protocol for mobile ad hoc networks, proposed by Schoonderwoerd et
al. (Schoonderwoerd et al, 1997; Sim and Sun, 2003). Their approach has provided a load balancing
algorithm in telecommunications networks by ants. Each node in the network has a pheromone table
included neighbors and all possible destination nodes. Ants travel between the nodes and determine
the optimal routes. The ant's life is also affected to update the pheromone tables. Their approach called
ABC algorithm which uses a proactive approach in routing.
Caro et al. proposed a proactive routing protocol named AntNet (Di Caro and Dorigo, 1998). AntNet
is one of the first Routing protocols based on an ant colony algorithm too. The mechanism which Caro
and Dorigo employed, was that nodes send ants to the destinations were selected randomly in regular
time periods. The main objective of ants, finding paths, allocate the quality of these paths and use this
information to update the routing table of nodes that pass through by ants. The routing tables contain
information about each node also shows a good route to go to a specific destination. Pheromone
information helps in routing data packets and ants routing. Data packets choose the path with more
pheromones. If a sufficient number of ants send to variant destinations, nodes will update their tables
about the best routes.
Marwaha et al. in 2002 combine the advantages of on-demand routing protocols, AODV, and
proactive ant algorithm. They created a hybrid method named Ant-AODV (Marwaha et al, 2002). The
ants use for reducing the latency of the network, making repetition route discovery process in the
attempt to decrement end-to-end delay and incrementing the probability of finding routs more quickly,
as compared to AODV. The goal of Ant-AODV is to gain the network topology, without any other
specific functions, as opposed to most other ACO algorithm approach so has an artificial pheromone
model based on the number of hops (Martins et al, 2010). The connections between the hops increase
by the ants, also the amount of rout request packets (RREQ) for a destination will be reduced.
Moreover using route error message (RERR) lead to inform link failures in the Ant-AODV network
(Gupta et al, 2012; G. Singh et al, 2012).
AntHocNet, an ACO based routing algorithm for MANETs, is another approach which proposed by
Di Caro et al (Di Caro et al, 2004). It is a hybrid algorithm that use reactive part to finding the path
and after that in proactive process maintains and improves the path. This approach combine the
advantages of both AntNet and ARA. Visited node maintains in the ant memory to detect cycles.
Sending Forward ants from the source to the destination using for route discovery process and a path
has been found, whenever a backward ant received. AntHocNet shown better performance, as
compared to AODV. Packet delivery ratio, and average end-to-end delay has been improved. The
conclusions shown ability for the dynamic environment (Kuchaki et al, 2010; G. Singh et al, 2012).

4. Algorithms

4.1. AODV
The AODV is a reactive protocol that combines the advantages of both protocols, Dynamic Source
Routing (DSR) and DSDV (Panda, 2012). If a node using AODV protocol for communication, send a
message to a destination node for which it does not have a valid route to, it initiates a route discovery
process to locate the destination node (Marwaha et al, 2002).
In the route discovery process a route request message (RREQ) is broadcasted to all neighbors, which
continue to broadcast the message to their neighbors (Jörg, 2003). The neighboring nodes in turn
broadcast the packet to their neighbors and the process continues until the packet reaches the
destination or until an intermediate node knows a route to the destination that is new enough. To
ensure loop-free and most recent route information, every node maintains two counters: sequence

4
number and broadcast_id. The broadcast_id and the address of the source node uniquely identify a
RREQ message. broadcast_id is incremented for every RREQ the source node initiates (Jörg, 2003).
When an intermediate node receives a RREQ if it has already received a RREQ with the same
broadcast_id and source address it drops the redundant RREQ and does not rebroadcast it (Perkins and
Royer, 1999). When an intermediate node forwards the RREQ message, it records the address of the
neighbor from which it received the first copy of the broadcast packet. This way, the reverse path from
all nodes back to the source is being built automatically. The RREQ packet contains two sequence
numbers: the source sequence number and the last destination sequence number known to the source.
The source sequence number is used to maintain freshness information about the reverse route to the
source while the destination sequence number specifies what actuality a route to the destination must
have before it is accepted by the source (Jörg, 2003). A RREQ is issued with a limited TTL. If no
RREP message is received within a certain time by the source node, then another RREQ is issued with
a larger TTL value. If still no reply, the TTL is increased in steps, until a certain maximum value is
reached (Vetrivelan and Reddy, 2008).
When the route request broadcast reaches the destination or an intermediate node with a fresh enough
route, the node responds by sending a unicast route reply packet (RREP) back to the node from which
it received the RREQ. So actually the packet is sent back reverse the path built during broadcast
forwarding. A route is considered fresh enough, if the intermediate node’s route to the destination
node has a destination sequence number which is equal or greater than the one contained in the RREQ
packet. As the RREP is sent back to the source, every intermediate node along this path adds a forward
route entry to its routing table (Jörg, 2003). If a node does not receive a reply then it deletes the node
from its list and sends RERR to all the members in the active members in the route (Kaur and Nayyar,
2013). The forward route is set active for some time indicated by a route timer entry (Jörg, 2003). A
routing table entry is said to be expired if not used within certain duration. These nodes in Maintaining
Routes process, are notified with route error (RERR) packets when the next-hop link breaks. In the
situation of link break, due to mobility node, each predecessor node, forwards the RERR to its own set
of predecessors. In this way all routes, which contain the broken link, are removed (Vetrivelan and
Reddy, 2008). After having learned about the failure, the source node may reinitiate the route
discovery protocol. Optionally a mobile node may perform local connectivity maintenance by
periodically broadcasting hello messages (Kuchaki and Motahari, 2015).

4.2. HOPNET
Wang et al. proposed a zone based routing algorithm named HOPNET in 2009 (Wang et al, 2009). In
this method which mixed the features of Zone Routing Protocol (ZRP), hybrid protocol which
combines benefits of proactive and reactive protocol, and DSR protocols, whole of the network is
divided to the overlap zones to achieve scalability and a hybrid method is used for routing. The
algorithm contain the local proactive route discovery within the node’s neighborhood (zone) and
reactive method between the overlap zones. The nodes can be part of one or more zone, because of the
overlap zones. The zones vary in size and radius length measured in hops. The nodes are two types:
Boundary nodes and interior nodes. Just the same as nodes, the tables are two types: Intrazone Routing
Table (IntraRT) which use for routing in zones and Interzone Routing Table (InterRT) which using for
routing between the zones. Route discovery occurs by intrazone and interzone routing. In intrazone
routing, each node periodically sends an ant in order to maintain the intrazone routing table updated.
When the source node wishes to send a data packet to a destination node within its zone, the packet
travels the trail pheromone.
Despite the pheromone does not contribute to route discovery in interzone routing, ants can still lay
down pheromoneon some paths. When a source node wishes to send a data packet to a node beyond its
zone, the border nodes from its zone transmit an ant to localize the destination. These ants hop
between the border zones until an ant localizes a zone with the destination. This ant propagation
through the border zones is called bordercast. When an ant achieve the destination, it will have a route
from source to destination node, so, this discovered route returns to source node by the ant and then
the data packet is transmitted(Okazaki and Frohlich, 2009).
HOPNET is efficient and comparatively better than traditional protocol, AODV, for delivery ratio and
end-to-end delay and is sufficient for large networks (Gupta et al, 2012).

5
 4.3. PAR
In 2009 Prasad et al. proposed a proactive algorithm for routing mobile ad hoc network named
Probabilistic Ant Routing (PAR), which is inspired by Ant colony optimization (ACO) metaheuristic
and uses ants for route discovery, maintenance and improvement. They increased the diversity in ACO
by modifying the transition rule. They could improve the results in delay and delivery ratio (Prasad et
al, 2009).
Each node maintains a list of neighbors according to HELLO MESSAGE received. Forward ant
agents (FANT) are probabilistic and explore the network to collect network traffic information. They
are routed on normal priority queue. If route to destination is available as present node, the FANT is
unicast otherwise it is broadcasted (R. Singh et al, 2010).

5. Performance Evaluation
In this section we analyze and compare the performance between three routing protocol, AODV, PAR
and HOPNET. The comparison is made based on the simulation results of this three algorithms. The
simulation program is written in the NS-2 simulator. The NS-2 simulator allows examining protocols
in various scenarios and comparing them with each other by varying the different parameters (End-to-
End Delay, Throughput, etc.). We measure routing overhead, packet delivery ratio, average end-to-end
delay and Throughput metrics. At the beginning of simulation the nodes spread in the environment
randomly and then move with uniform velocity between and toward the random destination.
The simulation parameters are as follows (Table 1):

Table 1. Simulation parameter

Parameters Value
Number of nodes 50
Type of protocol IEEE 802.11
Simulation surface 1200 1200
Maximum speed 20 m/s
Mobility model Random Way Point
Start time 10 s
End time 200 s
Antenna type Omni Antenna
Packet length 1000 bytes
Channel capacity 2 Mb/s

5.1. Routing overhead

Routing overhead is Network routing packets which send with routing protocol. Protocol with less
overhead is a better protocol.

6
 Figure 1. Routing overhead

As you can see in the figure 1, PAR has an acceptable routing overhead with up to 150 nodes,
however by the time, has a worse performance than AODV and HOPNET. AODV is a reactive routing
protocol and the same as other reactive protocols have the least routing overhead. HOPNET is a
hybrid protocol that merges the proactive and reactive approach. Accordingly, it has lower overhead
rather than pure proactive protocols, and more overhead rather than pure reactive protocols. In
proactive routing protocols, each node has all of the network nodes in the rows of the table and need
more memory. Also, each node in order to discover available paths need to send the periodic Hello
message.

5.2. Packet Delivery Ratio

A measure of the percentage of packets that have been successfully delivered to the destination node
(G. Singh et al, 2014).
(1)

7
 Figure 2. Packet delivery ratio

Results of simulating packet delivery ratio show in figure 2. As you can see, the results shown that
PAR rather than two approaches, AODV, and HOPNET has better performance. PAR is a proactive
protocol, and they have a ready path any time. This cause they have better performance in packet
delivery ratio than other two protocols.

5.3. End-to-End Delay

The End-to-End delay is equivalent to the time taken for a packet to be transmitted over the network
from source to destination. In other words, end-to-end delay is the total delay in every step from
source to destination point (G. Singh et al, 2014). Figure 3 shows the comparison of End-to-End delay
parameter.

Figure 3. Average End-to-End delay

8
End-to-end delay is a parameter that study the quality of exchange data packets in real time
connection. So that the method with less end-to-end delay parameter is more suitable for real-time
information exchange. Because of AODV is a pure reactive protocol, it has the largest end-to-end
delay. This protocol whenever there is a request for a path, starting the route discovery process. This
caused a delay until the path was founded and make higher End-to-End delay than other approaches.
HOPNET is a hybrid routing protocol, which acts as a reactive protocol between the zones and this
make higher End-to-End delay than PAR. But because of the proactive act inter the zones, have better
performance than AODV. Hereof the pure proactive routing protocols have better performance than
AODV and HOPNET. PAR with less delay has shown better performance compared with the others.

5.4. Throughput
Throughput in communication networks such as Ethernet or packet transmission, is the average rate of
successful delivery of a message in a communication channel. Throughput usually measured in bits
per second, packets per second or packets at intervals (Malany et al, 2009).

(2)

Figure 4. Throughput

As the definition implies, throughput is an average rate dependent on the successful delivery of a
message in a communication channel. In fact, in ways that have a higher message delivery rates and
moreover the less exchange routing packet rate, have higher throughput. As it shows in figure 4,
AODV presented higher delivery rate compared to other methods. Moreover, has less exchange
routing packet rate than PAR, since packets have more opportunity to fully benefit from bandwidth
communications. Lower transaction overhead cause improves throughput in HOPNET than PAR.

6. Conclusion

In this work we have tried to compare three reactive, proactive and hybrid protocols (AODV, PAR and
HOPNET), according to four parameters, as they proposed in the previous section. The simulation

9
results show that the HOPNET due to zoning, is saved in memory usage per node. AODV has highest
rate in Throughput. HOPNET has the second rank, higher than PAR. The more a Protocol use the
whole of the bandwidth and send packet successfully, the more Throughput rate achieve. But when it
comes to End-to-End delay, PAR is the best. PAR due to being proactive and having updated routes,
has lesser End-to-End delay than reactive protocol, AODV and hybrid protocol, HOPNET, so that it
help to send packets with a little delay. The proactive part of HOPNET help to gain the second rank in
this comparison, before AODV. Such this proactive part, needs overhead to gain the paths.
Consequently, the opposite trend can be seen as the End-to-End delay. Just like the overhead, in the
delivery ratio, we can see the same manner. PAR, HOPNET, and AODV.

References
Al-Ani, Razan. (2011). Simulation and performance analysis evaluation for variant MANET routing protocols.
International Journal of Advancements in Computing Technology, 3(1), 1-12.
Asokan, R, Natarajan, AM, & Venkatesh, C. (2008). Ant based dynamic source routing protocol to support
multiple quality of service (QoS) metrics in mobile ad hoc networks. IJCSS: International Journal of
Computer Science and Security, 2(3), 48-56.
Budyal, V. R., & Manvi, S. S. (2014). ANFIS and agent based bandwidth and delay aware anycast routing in
mobile ad hoc networks. Journal of Network and Computer Applications, 39(0), 140-151. doi:
[Link]
Cadger, Fraser, Curran, Kevin, Santos, Jose, & Moffett, Sandra. (2013). A survey of geographical routing in
wireless ad-hoc networks. Communications Surveys & Tutorials, IEEE, 15(2), 621-653.
Chen, Yuh-Shyan, Tseng, Yu-Chee, Sheu, Jang-Ping, & Kuo, Po-Hsuen. (2004). An on-demand, link-state,
multi-path QoS routing in a wireless mobile ad-hoc network. Computer Communications, 27(1), 27-40.
doi: [Link]
Correia, Fernando, & Vazão, Teresa. (2010). Simple ant routing algorithm strategies for a (Multipurpose)
MANET model. Ad Hoc Networks, 8(8), 810-823. doi: [Link]
Dhanda, Mamta, & Chaudhry, Shikha. (2013). Survey of Routing Protocols for Mobile Ad Hoc Networks.
International Journal of Advanced Research in Computer Science and Software Engineering, 3(4),
1026-1031.
Di Caro, Gianni, & Dorigo, Marco. (1998). AntNet: Distributed stigmergetic control for communications
networks. Journal of Artificial Intelligence Research, 9(3/4), 317-365.
Di Caro, Gianni, Ducatelle, Frederick, & Gambardella, Luca Maria. (2004). AntHocNet: an ant-based hybrid
routing algorithm for mobile ad hoc networks. Paper presented at the Parallel Problem Solving from
Nature-PPSN VIII.
Goyal, Nitin, & Gaba, Alka. (2013). A review over MANET-Issues and Challenges. International Journal of
Enhanced Research in Management & Computer Applications, 2(4), 16-28.
Gupta, Anuj K, Sadawarti, Harsh, & Verma, Anil K. (2012). MANET routing protocols based on Ant Colony
Optimization. International Journal of Modeling & Optimization (IJMO), ISSN, 2(1), 42-49.
Hussein, O, & Saadawi, T. (2003). Ant routing algorithm for mobile ad-hoc networks (ARAMA). Paper presented
at the Performance, Computing, and Communications Conference, 2003. Conference Proceedings of the
2003 IEEE International.
Ilie, Sorin, & Bădică, Costin. (2013). Multi-agent approach to distributed ant colony optimization. Science of
Computer Programming, 78(6), 762-774. doi: [Link]
Jörg, David Oliver. (2003). Performance Comparison Of MANET Routing Protocols In Different Network Sizes.
Computer Networks and Distributed Systems, University of Berne, Switzerland.
Kaur, Manjot, & Nayyar, Anand. (2013). A Comprehensive Review of Mobile Adhoc Networks (MANETS).
International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) 2(6), 196-
210.
Malany, A Boomarani, Dhulipala, VR Sarma, & Chandrasekaran, RM. (2009). Throughput and delay
comparison of MANET routing protocols. Int. J. Open Problems Compt. Math, 2(3), 462-468.
Martins, José Alex Pontes, Correia, S Luis OB, & Celestino, J. (2010). Ant-DYMO: A bio-inspired algorithm for
manets. Paper presented at the IEEE 17th International Conference on Telecommunications (ICT).
Marwaha, S., Chen Khong, Tham, & Srinivasan, D. (2002, 17-21 Nov. 2002). Mobile agents based routing
protocol for mobile ad hoc networks. Paper presented at the Global Telecommunications Conference,
2002. GLOBECOM '02. IEEE.

10
Okazaki, Alexandre Massayuki, & Frohlich, Antˆonio Augusto. (2009). Adapting HOPNET algorithm for
wireless sensor networks. IEEE Latin America Transactions, 8(4), 30-35.
Panda, Ipsita. (2012). A Survey on Routing Protocols of MANETs by Using QoS Metrics. International Journal
of Advanced Research in Computer Science and Software Engineering, 2(10), 120-129.
Perkins, Charles E, & Royer, Elizabeth M. (1999). Ad-hoc on-demand distance vector routing. Paper presented
at the Mobile Computing Systems and Applications, 1999. Proceedings. WMCSA'99. Second IEEE
Workshop on.
Prasad, Sunita, Singh, YP, & Rai, CS. (2009). Swarm based intelligent routing for MANETs. International
Journal of Recent Trends in Engineering, 1(1), 153.
Rafsanjani, M. Kuchaki, & Motahari, H. (2015). Analysis and Comparison Routing Protocols based on ACO and
Without ACO in MANET. Journal of Information and Computing Science, 10(1), 007-018.
Rafsanjani, Marjan Kuchaki, Asadinia, Sanaz, & Pakzad, Farzaneh. (2010). A hybrid routing algorithm based on
ant colony and ZHLS routing protocol for MANET Communication and Networking (pp. 112-122):
Springer.
Reddy, T. Bheemarjuna, Karthigeyan, I., Manoj, B. S., & Murthy, C. Siva Ram. (2006). Quality of service
provisioning in ad hoc wireless networks: a survey of issues and solutions. Ad Hoc Networks, 4(1), 83-
124. doi: [Link]
Schoonderwoerd, Ruud, Holland, Owen E, Bruten, Janet L, & Rothkrantz, Leon JM. (1997). Ant-based load
balancing in telecommunications networks. Adaptive behavior, 5(2), 169-207.
Sim, Kwang Mong, & Sun, Weng Hong. (2003). Ant colony optimization for routing and load-balancing: survey
and new directions. Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE Transactions
on, 33(5), 560-572.
Singh, Gurpreet, Kumar, Neeraj, & Kumar Verma, Anil. (2012). Ant colony algorithms in MANETs: A review.
Journal of Network and Computer Applications, 35(6), 1964-1972.
Singh, Gurpreet, Kumar, Neeraj, & Verma, Anil Kumar. (2014). ANTALG: An Innovative ACO based Routing
Algorithm for MANETs. Journal of Network and Computer Applications, 45, 151-167.
Singh, Jatinder Pal, & Gupta, Anuj Kr. (2013). A Review on Dynamic MANET On Demand Routing Protocol in
MANETs. International Journal of Advanced Trends in Computer Science and Engineering, 2(2), 21-
26.
Singh, Rajeshwar, Singh, Dharmendra K, & Kumar, Lalan. (2010). Swarm intelligence based approach for
routing in mobile Ad Hoc networks. International Journal of Science and Technology Education
Research, 1(7), 147-153.
Sofat, Smriti, & Gupta, Anuj K. (2013). Ant Colony Based Dynamic Source Routing. International Journal of
Advanced Research in Computer Science and Software Engineering, 3(10), 721-728.
Talwar, Bhawna, & Gupta, Anuj K. (2012). Ant Colony based and Mobile Ad Hoc Networks Routing Protocols:
a Review. International Journal of Computer Applications, 50.
Vetrivelan, N, & Reddy, AV. (2008). Performance Analysis of Three Routing Protocols for Varying MANET
Size. Paper presented at the Proceedings of the International MultiConference of Engineers and
Computer Scientists.
Wang, Jianping, Osagie, Eseosa, Thulasiraman, Parimala, & Thulasiram, Ruppa K. (2009). HOPNET: A hybrid
ant colony optimization routing algorithm for mobile ad hoc network. Ad Hoc Networks, 7(4), 690-705.
Yoo, Kwang-Seon, & Han, Seog-Young. (2013). A modified ant colony optimization algorithm for dynamic
topology optimization. Computers & Structures, 123(0), 68-78. doi:
[Link]
Zhang, Zhaojun, & Feng, Zuren. (2012). Two-stage updating pheromone for invariant ant colony optimization
algorithm. Expert Systems with Applications, 39(1), 706-712. doi:
[Link]

11