







































 

 P-ISSN : 2715-2448 | E-ISSN : 2715-7199 

Vol.4 No.1 January 2023 

Buana Information Technology and Computer Sciences (BIT and CS) 

 

11 | Vol.4 No.1, January 2023 

 

An Enhanced Bio-Inspired Aco Model For Fault-Tolerant Networks 

 

Samuel W Lusweti 1  

Masinde Muliro University of Science 

and Technology  

Email: lusweti015@gmail.com

Collins O Odoyo 2 

Masinde Muliro University of Science 

and Technology  

Email: codoyo@mmust.ac.ke 

 

 ‹β› 
 

Dorothy A Rambim 3 

Masinde Muliro University of Science 

and Technology 

Email: drambim@mmust.ac.ke

Abstract—This research mainly aimed at establishing the 

current functionality of computer network systems, evaluating 

the causes of network faults, and developing an enhanced model 

based on the existing ACO model to help solve these network 

issues. The new model developed suggests ways of solving packet 

looping and traffic problems in common networks that use 

standard switches. The researcher used simulation as a method 

of carrying out this research whereby an enhanced algorithm 

was developed and used to monitor and control the flow of 

packets over the computer network. The researcher used an 

experimental research design that involved the development of 

a computer model and collecting data from the model. The 

traffic of packets was monitored by the Cisco Packet Tracer tool 

in which a network of four computers was created and used to 

simulate a real network system. Data collected from the 

simulated network was analyzed using the ping tool, observation 

of the movement of packets in the network and message delivery 

status displayed by the Cisco Packet Tracer. In the experiment, 

a control was used to show the behavior of the network in ideal 

conditions without varying any parameters. Here, all the 

packets sent were completely and correctly received. Secondly, 

when a loop was introduced in the network it was found that the 

network was adversely affected because for all packets sent by 

the computers on the network, none of them was delivered due 

to stagnation of packets. In the third experiment, still, with the 

loops on, a new ACO model was introduced in the cisco packet 

tracer used to simulate the network. In this experiment, all the 

packets sent were completely and correctly delivered just like in 

the control experiment. 

Keywords—ACO, packets ,Loops, Networks, Algorithm  

 

Abstrak—Penelitian ini bertujuan untuk menetapkan 

fungsionalitas sistem jaringan komputer saat ini, mengevaluasi 

penyebab kesalahan jaringan, dan mengembangkan model yang 

ditingkatkan berdasarkan model ACO yang ada untuk membantu 

memecahkan masalah jaringan ini. Model baru yang 

dikembangkan menunjukkan cara-cara untuk memecahkan 

perulangan paket dan masalah lalu lintas di jaringan umum yang 

menggunakan standard switches. Peneliti menggunakan simulasi 

sebagai metode untuk melakukan penelitian ini dimana algoritma 

yang disempurnakan dikembangkan dan digunakan untuk 

memantau dan mengontrol aliran paket melalui jaringan 

komputer. Peneliti menggunakan desain penelitian eksperimental 

yang melibatkan pengembangan model komputer dan 

mengumpulkan data dari model tersebut. Lalu lintas paket 

dipantau oleh alat Cisco Packet Tracer di mana jaringan empat 

komputer dibuat dan digunakan untuk mensimulasikan sistem 

jaringan nyata. Data yang dikumpulkan dari simulasi jaringan 

dianalisis menggunakan alat ping, pengamatan pergerakan paket 

dalam jaringan dan status pengiriman pesan yang ditampilkan 

oleh Cisco Packet Tracer. Dalam percobaan, kontrol digunakan 

untuk menunjukkan perilaku jaringan dalam kondisi ideal tanpa 

memvariasikan parameter apa pun. Semua paket yang dikirim 

diterima dengan lengkap dan benar. Ketika sebuah loop 

diperkenalkan pada jaringan, ditemukan bahwa jaringan 

terpengaruh secara negatif karena untuk semua paket yang 

dikirim oleh komputer di jaringan, tidak ada satupun yang dikirim 

karena stagnasi paket. Dalam percobaan ketiga, masih dengan 

loop aktif, model ACO baru diperkenalkan di pelacak paket cisco 

yang digunakan untuk mensimulasikan jaringan. Dalam 

percobaan ini, semua paket yang dikirim dikirim dengan lengkap 

dan benar seperti pada percobaan kontrol. 

Kata Kunci—ACO, packets ,Loops, Networks, Algorithm  

 

I. INTRODUCTION 

One can rarely do anything with data that doesn’t involve a 
computer network since these networks enable us to share 
information and other resources [1]. However, these 
networks have to be maintained well to keep providing users 
with these functionalities. These computer networks can fail 
to work especially when a network device fails or the 
communication link malfunctions or is being overused 
against its capacity [2]. The existing networks are more 
complex than the conventional networks therefore it becomes 
hard to create, install, manage and keep them up and running 
efficiently all the time [3]. Due to these challenges, new 
technologies are needed to be employed in managing these 
dynamic networks that are at the center of business 
transactions today. There exist similar problems in real-life 
situations and their biological solutions which are naturally 
evolved and can also be applied in networking paradigms to 
help curb the drawbacks [3]. These are commonly referred to 
as Bio-inspired systems. A bio-inspired system depicts a 
strong relationship between a proposed algorithm aimed at 
solving a certain problem and a biological or natural system 
possessing similar capabilities [4]. There exists a necessity to 
employ bio-inspired systems in computer networks because 



 

12 | Vol.4 No.1, January 2023 

 

living organisms like the ant colonies look better organized 
in their daily activities than the current internet [4]. This is 
because of the resilience to failure by biological systems to 
both internal and external environmental factors [5]. There 
exist many optimization algorithms including Particle Swarm 
Optimization, genetic algorithm, leaping frog among others. 
In general, ACO is the most popular and most successful 
algorithm that has ever been used in combinatorial 
optimization problems [6]. 

 

II. LITERATURE REVIEW  

2.1. Loops 

Loops can occur in a network whenever there exists more 
than one path (redundant links) at layer 2 between two 
endpoints or there are multiple connections between any two 
switches in the network or there is a physical connection 
between two active ports of the same switch. This loop 
creates broadcast storms during the forwarding of broadcasts 
and multicasts by switches. These switches repeatedly 
rebroadcast the message hence flooding the network [7].  

This means that packets that are sent along a given path will 
eventually be stuck in that network forever making cycles, [8] 
unless some mechanisms are invoked which will help in 
flushing such packets out of the cycle. Whenever you are 
using link-state protocols for instance OSPF, the forwarding 
loops may transiently occur if the routers in use adapt their 
own forwarding route tables in response to a change in 
topology [9]. The adverse effects of the loop on Ethernet 
switched networks include a reduction in bandwidth, memory 
clogging, and packet loss [7]. These and many other 
challenges in computer networks need to addressed using 
more intelligent mechanisms especially the use of 
metaheuristics like bio-inspired systems. This research paper 
concentrates on packet loops and provides an enhanced bio-
inspired mechanism to help solve the problem. 

2.2 Bio-inspired systems 

Biology is frequently being employed as an inspiration for 
research in computer science [2] and other fields like 
engineering, mathematics, energy [10], and business. Ant 
algorithms are in use today mainly to solve optimization 
problems instead of the problems and challenges that they 
were originally or initially developed to solve [11]. A direct 
approach to getting a solution to the combinatorial 
optimization problems is an exhaustive search [12], whereby 
the agents in these biological systems whose analogy is used 
to create bio-inspired systems, enumerate all possible 
solutions and choose the best one. The method of ants which 
is a bio-inspired system has proved to outshine other general-
purpose algorithms for optimization like the genetic 
algorithms. This is evident especially when employed in 
combinatorial optimization problems which require the 
interaction of cooperating agents [11]. A good number of 
such intelligent algorithms are therefore being developed 
with the aim of solving various complex problems. Whereas 
some studies try to explore the application of bio-inspired 
algorithms theoretically, others are continuously working to 
improve the functionality of the algorithms [10]. This 
becomes a green light for the growth and development of 
artificial intelligence systems which mimic the behavior of 
ants, especially during foraging. These algorithms include 

Neural Networks, Particle Swarm, Genetic Algorithm, and 
Ant Colony Optimization algorithm among others [10].  

2.2.1 ACO architecture and design 

During the early years of 1990s, there was an introduction of 
the ACO model proposed by M. Dorigo and his companions 
as a metaheuristic optimization algorithm that is naturally 
inspired for solving hard combinatorial optimization 
problems [13]. ACO is in the category of metaheuristics 
which are probabilistic algorithms for obtaining good 
solutions to hard combinatorial optimization problems with a 
reasonable time of computation [14]. The other examples of 
metaheuristics include tabu-search, evolutionary 
computation, and simulated annealing [15] [16]. The foraging 
behavior of real ants inspired the creation and deployment of 
ACO in various fields. ACO is a very popular and modern 
optimization paradigm that gets its motivation from the 
scenario of how ant colonies find the shortest routes between 
their home and the food source [6]. During the process of 
searching for food (foraging), the forward ants start by 
randomly exploring the environment that surrounds their nest 
before they locate the source of food and deposit pheromone 
trails [17]. Despite its well-known popularity and application, 
the theory of ACO is still under development and is in its 
infancy stages; therefore a solid foundation of its theory is 
required [18]. The ACO model has been modified in many 
forms since its introduction. However, these forms have a 
common architecture. The following figures show an image 
of real ants in a colony foraging and a flow chart of activities 
carried out by ants in an ACO algorithm during their foraging 
activities. 

2.2.2 How ACO works 

Ants are self-organized biological systems exhibiting three 
main principles of self-organization mechanisms which 
include interaction between individuals, feedback loops, and 
local state evaluation [19]. ACO algorithm works following 
an indirect communication among the simple agents of a 
colony, known as artificial ants, enabled by their artificial 
pheromone trails as their media of communication [20]. The 
foraging ants thus communicate by laying pheromone 
chemicals on the ground as they search the environment for 
food [18]. The other ants are consequently attracted by the 
laid pheromone trails and therefore tend to closely follow 
previous ants. In the case whereby the foraging ants discover 
different routes between the nest and a source of food, the 
shortest path typically gets filled with pheromone quicker 
than the longer path [18]. A fascinating aspect of ants is their 
ability to find the shortest routes to the source of food. This 
is made possible only by the ability of these ants to follow the 
laid down pheromone trails by the predecessor ants taking in 
mind that most ants are almost blind meaning visual aspects 
are not in use [21]. The more the ants take the shortest route, 
the more pheromone gets deposited, until almost all ants 
follow the shortest path [18]. 

2.3 Application of ACO in Networks 

There have been many areas in which bio-inspired systems 
have been successfully applied, some of which have been 
mentioned in the sections above. However, a few of them 
have been applied in computer networks to enhance their 
functionality and improve their resilience to faults. The main 
area of concern being adaptive routing. A good example is 
the ACO algorithm for network analysis and adaptive routing 



 

13 | Vol.4 No.1, January 2023 

 

[10]. Many issues in networking are formulated as 
multidimensional optimization problems. As dimensions of 
networks are increased both in terms of the number of nodes 
and spatially, the centralized control of communication in 
these networks becomes impractical. In comparison with 
biological systems, an individual alone is of less interest 
compared to the collective behaviour of the system of a larger 
number of the same individuals like the ant colonies [10]. As 
a result, Bio-inspired systems like ACO have been developed 
and successfully applied in network node research and design 
due to the appealing analogies existing between biological 
systems and large computer networks [2]. ACO has been 
applied in computer network routing problems in different 
formats such as AntNet, AntHocNet, ACR [6], and Ant-
BasedControl(ABC) [22] among other algorithms. ABC was 
the first routing algorithm applied in circuit-switched 
networks for instance telephone wire networks [23]. This 
algorithm was deployed on a simulated version of the British 
Telecom network, which formed the basis [20] of more 
research on this area of network routing problems. A highly 
successful application of the ACO to the dynamic routing 
problems is the ANTNET algorithm, which was proposed by 
Marco D and Di Caro [24] [25] [26]. This ANTNET being 
successful was recommended and applied in packet-switched 
networks in this case the internet for adaptive routing [20]. 
Later on, the ANTNET became widely useful in mobile ad-
hoc networks, to solve the routing problems and was used as 
ANTHOCNET posting exemplary results [27]. ANTNET 
[28] and ANTHOCNET [29] are two today very well-known 
ACO-based routing algorithms. ANTNET is a proactive 
routing algorithm while the ANTHOCNET is a reactive 
routing algorithm. They possess a very high delivery rate and 
find paths whose lengths are very close to the length of the 
shortest route [28],[30].  

2.4 Challenges faced by ACO  

If ACO is used in network routing, the packets mimic the real 
ants. However, it still has some challenges. In this case, a 
problem arises when an ant (packet) is stuck in a cycle (loop) 
and is forced to revisit an already visited node. These loops 
bring about the stagnation of ants [31] and are undesirable in 
networks because they cause packet latencies [6] and the 
packets caught in a loop are eventually destroyed when ACO 
is implemented. The large number of routes or paths makes it 
complex to manage the routing tables while concurrently 
increasing the probability of having packet loops [32]. 
Although research was done and a new ACO was developed 
known as MACO, the research does not claim that MACO 
will fully eradicate stagnation in ants, thus offering a basis for 
more research to be done on ACO [31]. These reasons 
necessitated research on how best to re-route the packets that 
are caught into a loop and stagnated without destroying them 
and by preventing stagnation and still making them reach 
their intended destination in the new ACO algorithm. 

 

III. METHOD 

A model was developed using python and pygame simulator 
to show the movement of ants. This study used the agile 
family as a model of SDLC since these methods are meant to 
quickly adapt to changing requirements, and minimize costs 
of production while upholding the quality of the software 
under development [33]. Agile is a combination of both 
incremental and iterative types of SDLC [34]. 

IV. RESULTS AND DISCUSSION 

In this chapter, the researcher presents the results of the 
simulation experiments performed. In these experiments, the 
researcher used the Cisco Packet Tracer simulator to test 
various configurations of the network as shown below.  First 
of all, the packet tracer is launched and a simple network 
configuration of four computer devices (two desktop 
computers and two laptops) are configured in the network. A 
router is used to bridge between two different network classes 
each having two computers. These classes include B having 
a default gateway of 172.16.0.1 and class C with a default 
gateway of 192.168.0.1. These computers are interconnected 
on the network using two 24-port. The so-developed network 
of computers is then subjected to different conditions and 
tested as shown by the screenshots below. 

4.1. Simple-ACO 

The old model called Simple-ACO was applied in various 
fields including computer networks. S-ACO ants were used 
to implement loop elimination which in turn improve the 
performance of the system [12]. While moving backwards 
through the path that in in their memory, the ants update the 
pheromone concentration on the paths they pass through. The 
following example shows how ants in S-ACO eliminate loops 
as using backward pass presented by Marco Dorigo [12]. 

 

 

Figure 1. Final route, no loop but also node 4 is removed 

In ACO, the private memory of ant is used to guarantee the 
probability of an ant building a feasible solution. However, 
during the process of finding these feasible routes, loops may 
appear making the ant rotate in a cycle for long. As a result, 
the ant ends up wasting time and resources [6]. When trying 
to avoid these loops, if an ant is forced to go back to an 
already toured node, the nodes having that cycle are removed 
from the private memory of the ant and information about the 
nodes is completely destroyed. If an ant moves in a loop for 
a time which is greater than half its age (Time To Live), the 
ant is consequently destroyed [6]. The model describes the 
movement of an ant from source to destination. From the flow 
diagram, if an ant is forced to return to an already visisted 
node, the node or ant will be destroyed. This saves the system 
from staying in a loop forever, but the killed ant never 
reached its destination. If the node is detryoed, then it is 
removed from the network making the node lose network 
connection. 



 

14 | Vol.4 No.1, January 2023 

 

 

 

Figure 2. Flow chart of Original ACO model  
showing how to eliminate loops 

pseudocode 
if (k ∈ V )/∗  CHECK IF THE ANT IS IN 
A LOOP AND   REMOVE IT ∗/ 
hops_cycle ← get cycle_ length(k,V); 

     hops_fw ← hops_fw – hops_cycle; 

else 

     hops_fw ← hops_fw + 1; 

     V [hops_fw] ← k; 

     T ,[hops_fw] ← Tk→n; 

end if 

 

4.2. Enhanced ACO model showing physical movement of 
ants 

 

Figure 3. Loops formed are assumed to be non-existent  
(for instance node 4 is not removed) 

In the figure above, if a loop is formed in the modified ACO 
model, it will be assumed by the ants as the ants will still be 
reroute out of the loop if they are caught in the loop for more 
than half of their TTL. In this case no ant will be killed and 
no node will be destroyed. The ants are made to be a little 
more intelligent in that, when an ant is caught in a cycle, its 
assigned time to live will be used to determine if the ant is in 
a loop. If the ant takes rotates in a loop for a time interval 
which is greater than half of its TTL, then it is rerouted out of 
the loop to its intended destination. 

 

 

Figure 4. Flow chart Enhanced ACO model  
showing how to eliminate loops 

Pseudo-code of enhanced ACO model 

if (k ∈ V )/∗  CHECK IF THE ANT IS IN 
A LOOP AND REMOVE IT ∗/ 
 hops_cycle ← getcycle length(k,V; 

      hops_fw>TTL/2; 

else 

hops_fw ← hops_fw + 1; 

V [hops_fw] ← k; 

      T ,[hops_fw] ← Tk→n; 

          end if 

 

In the enhanced algorithm above, if an ant is caught in a cycle, 
hops_fw>TTL/2 , the else statement function hops_fw ← 
hops_fw + 1 is invoked to get the ant out of the loop 
unconditionally. The ant selects the next node described in its 
foraging instructions. This enhanced algorithm is then 
applied in the simulated computer networks to help reroute 
packs intentionally forced to return to selected nodes. This 
forced return is done by introduction of forced loops. 

4.0 Control experiment 

Pinging the devices in the control experiment for instance 
from laptop 1 having IP address 172.16.0.3 to laptop 0 and 
PC 0 having IP addresses 192.168.0.3 and 192.168.0.2 
respectively, still shows communication taking place as 
shown by the command prompt screen in Figure 5 below. 



 

15 | Vol.4 No.1, January 2023 

 

 

 
Figure 5. Control results 

4.1 Simulator with loops but without the algorithm  

The simulated network is configured as shown below. There 
is a modification from the control experiment by the 
introduction of physical loops on the network. Figure 6 below 
shows the configuration. 

After simulating packets in this network, the message 
remained in progress forever as shown in the bottom right 
corner of the figure below. Secondly, the messages in the 
network simulator kept on rotating between the loops created 
and the two switches 0 and 4. 

 

 
Figure 6. Loops without ACO result 1 

When we open the command prompt and ping from laptop 1 
having IP address 172.16.0.3 to laptop 0 and PC 0 having IP 
addresses 192.168.0.3 and 192.168.0.2 respectively, the 
ICMP message fails to be delivered and shows time out on 
the command prompt and failed on packet tracer as shown in 
figure 7 below. 

 

 
Figure 7. Loops without ACO result 2 

4.2  Simulator with loops and with the algorithm  

The researcher was able to run the algorithm and then 
simulated the network having the loops. Secondly, the main 
algorithm is imported in the programming mode under the 
TCP Python package of each computer device running in the 
networking. The algorithm displays the activity of random 
movement of ants on the pygame.  

The simulation mode of the Cisco Packet Tracer shows 
message delivery in the network with green ticks as shown in 
the figure below.  At the bottom right corner of the simulator, 
there is an indication of the successful delivery of messages. 

 

 
Figure 8. Loops with ACO result 1 

Pinging the devices for instance from laptop 1 having IP 
address 172.16.0.3 to laptop 0 and PC 0 having IP addresses 
192.168.0.3 and 192.168.0.2 respectively, indicates 
communication taking place as shown by the command 
prompt screen below. 

 

 
Figure 9.  Loops with ACO result 2 

 

4.4 Interpretation of results 

In the above simulation, the researcher used three 
experiments to test and validate the data collected. The first 
which is a control experiment provides an ideal network with 
no variations in the network working under optimal 
conditions. It was discovered that the network messages were 
delivered properly between the devices on the network 
without any problem showing 100% success for the 4 
messages sent over the network. In the second experiment, 
the researcher subjected the simulated network to forced 
loops as it normally happens in colleges and universities 
(mostly done by students). In this experiment, it was 
discovered that the network went completely down and could 



 

16 | Vol.4 No.1, January 2023 

 

not recover from the looping packets which completely failed 
to be delivered to the intended destination. Pinging the 
network showed that the 4 messages send had a 100% failure 
rate. In the third experiment, the ACO Algorithm was 
simulated alongside the network with the loops still on. This 
experiment showed packets being delivered normally in the 
simulation mode by showing success to all ICMP packets 
sent. The ping tool also showed a 100% success rate for all 
the 4 test packets sent by various devices on the network.  
This shows that the algorithm intervened to reroute the 
packets that would have otherwise gotten locked up in the 
loop as shown in experiment 2 where the packets kept 
rotating in the loop thereby breaking communication between 
the computers on the network. The algorithm helped the 
network in rerouting any packets that would be caught in the 
loop to reach their destination. From the above experiments, 
five average Round Trip Times(RTT) were collected from 
control experiment and five from enhanced experimen. The 
control experiment had the following RTT as shown in figure 
8 and figure 9 (6,0,0,2,0) averaging to 8/5 =1.6ms. The 
experiment of enhanced algorithm showed the following 
results as RTT as witnessed from figure 16 and figure 17 
(0,0,2,0,16) avearaging to 18/5=3.6ms. Although the 
algorithm’s results show that the packets get to destination on 
average 2 ms later than in the case of the control, it is still 
witin the allowed RTT. The 2ms delay may have been 
brought about by the cycling packets before getting out of the 
loop. For optimal network performance , studies have shown 
that RTT<500ms [75]. 

4.5 How the enhanced ACO algorithm works 

The ACO model was first developed and applied in computer 
networks to help the network packets reach their destination 
very fast using the shortest route possible by mimicking the 
ants through their foraging behavior and intelligence. 
However, the model could not solve the problem of 
stagnation of ants and thus the new model developed in this 
research helps to get the stagnated ants out of the loop. The 
algorithm is then imported into the Cisco Packet Tracer and 
the packet tracer translates the ants into real packets of the 
network. When the packets are caught in a loop, they use their 
random intelligent movement to get out of the loop, once they 
realize there are talking too long to reach their destination. 
The movement of the packets (similar to the ants after 
importing into packet tracer) is as shown below. 

 

 
Figure 10. Ants in model foraging 

Figure 10 above shows packets finding the routes from the 
source of food to the nest (dest). 

 

 
Figure 11. Ants in model caught in a loop 

The figure above shows packets (ants) caught in a loop for 
sometime. 

 

 
Figure 12. Ants getting out of the loop 

This figure shows (packets) ants getting out of the loop after 
rotating in the loop for a few moments. The will be 
unnoticeable in the real network since it happens at a very 
high speed. 

4.6 Discussion  

According to research done by Kwang Hong Sim and Weng 
Hong Sun [31], when implementing ACO the following 
issues arise: stagnation of ants and  poor adaptiveness of ants. 
Stagnation will occur if a network reaches its convergence 
too early (prematurely). This leads to the following problems: 
1) congestion of packets (ants) on the optimal path, 2) 
reduction of probability of ants choosing other routes. The 
research further indicated that congestion and low probability 
of choosing other paths by ants will consequently lead to the 
following problems in dynamic computer networks: a) The 
selected path may become non-optimal if congested b) The 
congested path may be disconnected due to network failure. 
Due to these issues, Sim and Sun noted that the network 
would eventually have a low degree of sensitivity due to 
changes in topology or link failure. In their research, Sim and 
Sun [31] proposed a new improved model called MACO 
which they noted that it enhanced the adaptiveness of the 
network and reduced the chances of stagnation of ants. In 
MACO, the ants are made to deposit pheromones of different 
colors and observed that ants will follow routes with similar 
colors reducing congestion. However, they conclude that 
their research does not guarantee that MACO would fully 
eradicate stagnation of ants as it only reduced the chances of 
stagnation. Secondly, another research was done by Gianni 
Di Caro on the application of ACO to adaptive routing in 
telecommunications networks [8]. The research showed that 
loops that lead to stagnation of ants should be avoided as 
much as possible because they cause very high packet 



 

17 | Vol.4 No.1, January 2023 

 

latencies. To implement loop avoidance, Di Caro explained 
using an equation for avoiding loops in networks. For 
instance, if an ant is forced to return to a node that it had 
previously visited, the nodes that composed the loop are taken 
out of the memory of the ant, and all information describing 
those nodes is destroyed. Consequently, if an ant moves on a 
cycle for a time that is greater than its half-life, the ant is 
destroyed. This destruction solves the problem of loops; 
although, another problem arises whereby the affected nodes 
and ants(packets) will be destroyed which means they will 
not communicate over the network as intended. Considering 
the finding of the above two kinds of research, the researcher 
carried out this study to help strike a balance between 
stagnation and the destruction of ants. From the results above, 
the problem of stagnation of ants is more highly reduced than 
that of MACO suggested by Sim and Sun [31] by making sure 
the ants that are caught in a loop simply get out of the loop 
using a new optimal route as shown in figure 10. This will 
help us make sure that all the packets in the network reach 
their intended destination as opposed to destroying stagnated 
ants as found out in the research by Di Caro [8]. 

 

V. CONCLUSION AND RECOMMENDATIONS 

The results show that this research can add some contribution 

to the body of knowledge. When well adopted in schools, 

colleges, homes, and offices, the algorithm produced in this 

research study can go a long way in improving the reliability 

of the computer networks that are constantly under attack by 

forced loops. However, the algorithm does not completely 

guarantee that it will solve all the problems faced by 

computer networks. Further research needs to be done to 

improve it further. 

REFERENCES 

[1]  Meyers, M., "Introducing Basic Network Concepts." 

2010, 

https://www3.nd.edu/~cpoellab/teaching/cse40814_fall

14/networks.pdf (Accessed June. 3, 2022). 

[2]  Hylsberg R. Jacobsen, Q. Zhang,T. Skjødeberg 

Toftegaard, "Bioinspired Principles For Large-Scale 

Networked Sensor Systems: An Overview", Sensors, 

Vol. 11, No. 4, Pp. 4137-4151, 2011. DOI: 

10.3390/S110404137. 

[3]  Jisha Mrriam Jose, "Bio-Inspired Networking," 2011   

HYPERLINK 

"Http://Dspace.Cusat.Ac.In/Jspui/Bitstream/12345678

9/3195/1/Bio-Inspired%20networking.Pdf"  

Http://Dspace.Cusat.Ac.In/Jspui/Bitstream/123456789/

3195/1/Bio-Inspired%20networking.Pdf  (Accessed 

Jan. 23, 2020) 

[4]  Jamalipour A. "Bio-Inspired Networking", The 

University Of Sydney: IEEE, 2009.   HYPERLINK 

"Http://Www.It.Is.Tohoku.Ac.Jp/~Kato/Workshop200

9/01.Pdf"  

Http://Www.It.Is.Tohoku.Ac.Jp/~Kato/Workshop2009

/01.Pdf  (Accessed Feb. 20, 2020) 

[5]  Dressler, I.F., "Self-Organization in Autonomous 

Sensor/Actuator Networks [SelfOrg." 2010. 

[6]  Caro D.G., Dorigo M. "Ant Colony Optimization And 

Its Application To Adaptive Routing In 

Telecommunication Networks, 2004". (Doctoral 

Dissertation, Phd Thesis, Faculté Des Sciences 

Appliquées, Université Libre De Bruxelles, Brussels, 

Belgium). 

[7]  Iufoaroh S.U, Chukwumaobi .O,Oranugo C.O 

"Tracking The Effects Of Loops In A Switched 

Network Using Rapid Spanning Tree Network," 

International Journal Of Research In Electronics And 

Communication Technology (IJRECT 2015), Vol. 2, 

No. 3, 2015. 

[8]  Balakrishnan H, "Network Routing - II Failures, 

Recovery, And Change," 2009.   HYPERLINK 

"Http://Web.Mit.Edu/6.02/Www/S2009/Handouts/L21

.Pdf(Accessed"  

Http://Web.Mit.Edu/6.02/Www/S2009/Handouts/L21.

Pdf(Accessed  Jan. 6, 2021) 

[9]  Francois P., Bonaventure O. "Avoiding Transient Loops 

During The Convergence Of Link-State Routing 

Protocols." IEEE/ACM Transactions On Networking, 

15(6), 2007, Pp. 1280-1292 

[10]  Kumar K.A, "Bio Inspired Computing – A Review Of 

Algorithms And Scope Of Applications", Expert 

Systems With Applications, Vol. 59, Pp. 20-32, 2016. 

DOI: 10.1016/J.Eswa.2016.04.018. 

[11]  Afshar M.H, "A New Transition Rule For Ant Colony 

Optimization Algorithms: Application To Pipe Network 

Optimization Problems", Engineering Optimization, 

Vol. 37, No. 5, Pp. 525-540, 2005. DOI: 

10.1080/03052150500100312. 

[12]  Marco D., Thomas S., "Ant Colony Optimization", 

London: Bradford Book, 2004. 

[13]  Dorigo, M., Vittorio M.,Alberto C., "Ant System: 

Optimization By A Colony Of Cooperating Agents." 

IEEE Transactions On Systems, Man, And Cybernetics, 

Part B (Cybernetics) 26, No. 1 (1996): 29-41 

[14]  Glover F., Gary A., "Handbook Of Metaheuristics." 

Vol. 57. Springer Science & Business Media, 2006 

[15]  Glover F, "Tabu Search—Part II,," ORSAJ. Comput., 

Vol. 2, No. 1, Pp. 4-32, 1990. 

[16]  Glover, F., Laguna, M.,"Tabu Search Kluwer 

Academic". Boston, Texas, Dordrecht, 1997. 

[17]  Dorigo M., Blum C., "Ant Colony Optimization 

Theory: A Survey," Theoretical Computer Science, Vol. 

344, Pp. 243-278, 2005. 

[18]  Sudholt, D. And Thyssen, C., "Running Time Analysis 

Of Ant Colony Optimization For Shortest Path 

Problems". Journal Of Discrete Algorithms, 10, 2012, 

Pp.165-180 

[19]  Dressler, F., "Benefits Of Bio-Inspired Technologies 

For Networked Embedded Systems: An Overview". In 

Dagstuhl Seminar Proceedings. Schloss Dagstuhl-

Leibniz-Zentrum Fr Informatik. 2006. 

[20]  Dorigo, M., Stützle, T.,. "Ant Colony Optimization: 

Overview And Recent Advances. Handbook Of 

Metaheuristics", 2019, Pp.311-351. 

[21]  Cordon O., Herrera F., Stutzle T. "A Review Of Ant 

Colony Optimization Metaheuristic: Basis, Models And 

New Trends," 2002. 



 

18 | Vol.4 No.1, January 2023 

 

[22]  Mandeep Kaur, A.S "A Review On Ant Colony 

Optimization In MANET," International Journal For 

Science And Emerging, Vol. 19, No. 1, Pp. 1-6, 2014.   

[23]  Schoonderwoerd, R., Holland, O.E., Bruten, J., 

Rothkrantz,. "Ant-Based Load Balancing In 

Telecommunications Networks". Adaptive Behavior, 

5(2), 1997, Pp.169-207.   

[24]  Marco. D., Gianni C., "Ant Colonies For Adaptive 

Routing In Packet-Switched Communications," In Fifth 

International Conference On Parallel Problem Solving 

From Nature, 1998. 

[25]  Marco. D., Gianni C., "Mobile Agents For Adaptive 

Routing.," In Proceedings Of The 31st International 

Conference On System Sciences, 1998. 

[26]  Marco. D., Gianni C., " Antnet: Distributed Stigmergic 

Control For Communication Networks," Journal Of 

Artificial Intelligence Research, Vol. 9, Pp. 317-365, 

1998 

[27]  Gianni D., Frederick D. ,Luca M.G , "Using Ant Agents 

To Combine Reactive And Proactive Strategies For 

Routing In Mobile Ad Hoc Networks," International 

Journal Of Computational Intelligence And 

Applications, Vol. 5, No. 2, P. 169–184, 2005. 

DOI/10.1142/S1469026805001556. 

[28]  Marco. D., Gianni C., "Ant colonies for adaptive routing 

in packet-switched communications networks.," In 

Proceedings of the 5th ACM International Conference 

on Parallel Problem Solving from Nature, p. 673–682, 

1998. 

[29]  Di Caro, G., Ducatelle, F., Gambardella, L.M. 

"AntHocNet: an adaptive nature‐inspired algorithm for 

routing in mobile ad hoc networks". European 

transactions on telecommunications, 16(5), 2005, 

pp.443-455.  

[30]  Gunes M., "Ara - the ant-colony based routing 

algorithm for manets.," In Proceedings of the 2002 

International Conference on Parallel Processing 

Workshops, pages, pp. 79-89, 2002.   

[31]  Sim, K.M., Sun, W.H., November. "Multiple ant-colony 

optimization for network routing". In First International 

Symposium on Cyber Worlds, 2002. Proceedings, 2002, 

(pp. 277-281). IEEE.  

[32]  ManiezzoV., "Ant Colony Optimization", 2001.   

[33]  Young D., "Sofware Development Methodologies," 

2013.   

[34]  Sharma, S., Sarkar, D., Gupta, D.,." Agile processes and 

methodologies: A conceptual study". International 

journal on computer science and Engineering, 4(5), 

2012, p.892. 

[35]  Rasmussen J., "What is the basis for classifying a call as 

poor in Lync 2013 QoE?," 20 September 2013. 

[Online]. Available:  https://docs.microsoft.com/en-

us/archive/blogs/jenstr/what-is-the-basis-for-

classifying-a-call-as-poor-in-lync-2013-qoe. (Accessed 

July. 2, 2022). 

 

 


