A preview of this full-text is provided by IGI Global.
Content available from International Journal of Cloud Applications and Computing
This content is subject to copyright. Terms and conditions apply.
DOI: 10.4018/IJCAC.309937
Volume 12 • Issue 1
Copyright © 2022, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
*Corresponding Author
1
Cryptographic algorithms enable secure data communication over public insecure networks. Though
they enhance network security, complex cryptographic operations consume substantial amounts of
computing resources, introducing significant network overhead costs. This study aims to find the
cryptographic algorithm that can efficiently utilize network resources. The study evaluates three
cryptographic algorithms with different file formats on varying numbers of node densities. The
NS-3 simulator was used to measure latency, data throughput, end-to-end delay, packet delivery
ratio, and packet loss of files in text, image, and audio formats. The results find AES as better than
DES and 3DES for a large number of node densities for the three file formats in terms of latency,
data throughput, end-to-end delay, and packet delivery ratio. However, DES has the lowest packet
loss as AES records the highest packet loss. The findings provide researchers avenues for further
research and the practitioners the choice of suitable algorithms based on the overhead performance.
Advanced Encryption Standard, Data Encryption Standard, End-to-End Delay, Latency,, Packet Delivery Ratio,
Packet Loss, Throughput, Triple Data Encryption Standard
The rapid growth in the development of complex cryptographic algorithms coupled with the variety
and volume of data traversing the global network in a distributed systems environment raises the
issue of network performance (Unal et al., 2021). Cryptographic algorithms are methods used to
ensure confidentiality, authenticity, and data integrity (Kumari, 2017). Cryptographic algorithms
enable secure data communication over public insecure networks (Rojasree & Gnanajayanthi,
2020) by hiding the original information from the intruder (Pal, Datta, & Karmakar, 2022). Though
Samuel Asare, University of Ghana, Ghana
Winfred Yaokumah, University of Ghana, Ghana*
https://orcid.org/0000-0001-7756-1832
Ernest Barfo Boadi Gyebi, University of Ghana, Ghana
Jamal-Deen Abdulai, University of Ghana, Ghana
Volume 12 • Issue 1
2
cryptographic algorithms enhance network security (Makarenko et al., 2020; Jing et al., 2014),
cryptographic operations consume substantial amounts of computing resources (Patil et al., 2016).
This leads to significant overhead costs in resource-bounded systems (Braeken, 2022; Laszka,
Vorobeychik, & Koutsoukos, 2018), such as in computer networks. Overhead costs are introduced
because cryptographic algorithms depend on complex arithmetic computational functions to perform
their operations. Therefore, the encryption and decryption time, execution time, running time, key
strength, and memory affect the overall performance of the network (Mushtaq et al., 2017).
Besides, the usage of cryptographic methods on computer networks introduces significant
communication overhead cost (Braeken, 2022), which includes data throughput, end-to-end delay,
data loss, memory usage, and latency (Gueron, 2016). Similarly, communication networks also have
computational complexities, inefficient security, reduced throughput, and increased delay (Mohindra
& Gandhi, 2021). In this regard, previous works analyse the efficiency of cryptographic algorithms
in various contexts. Notable among them are the evaluation of image files with AES (Elhoseny et al.,
2020; Shaktawat et al., 2020) and the encryption of text files (Jintcharadze et al., 2021; Pal, Datta,
& Karmakar, 2022). Other studies assess the power utilization, different key sizes, CPU utilization
time, and the encryption speed of each of the algorithms (Parkar et al., 2021; Fazzat et al., 2020).
These studies focus on a specific data format (image only, text only, video, or audio only) or
networks with specified node densities or stand-only systems. However, the overhead performance
of these algorithms may differ on networks with different file formats (Zhang & Liu, 2021). This
study implements and evaluates popularly used cryptographic algorithms DES, 3DES, and AES
to show their overall communication overhead performance with a varying number of nodes with
different file formats. Thus, this study aims to find the cryptographic algorithm that can effectively
and efficiently utilize network resources. Specifically, this study (a) investigates the impact of DES,
3DES, and AES cryptographic algorithms on throughput, (b) ascertain the effect of the cryptographic
algorithms on end-to-end delay, (c) examines the impact of the cryptographic algorithms on latency,
(d) investigates the packet delivery ratios of the cryptographic algorithms.
Cryptographic algorithms are grouped into symmetric key encryption and asymmetric key encryption.
Symmetric key encryption uses a single key to encrypt and decrypt data (Ramesh & Suruliandi,
2013). There are various symmetric key algorithms, including DES, triple DES, AES, and IDEA
(International Data Encryption Algorithm) (Kumar et al., 2011). However, in asymmetric key
encryption, two different keys are used. One of the keys is used for encryption (public key) and the
other for decryption (private key). The public key is meant for general use. It is made available to
anyone on the network. Thus, any user who wants to encrypt plaintext should know the public key
of the receiver. But the private key is kept secret from the outside world. Only the authorized users
can decrypt the ciphertext through the private key (Agrawal & Mishra, 2012).
Examples of asymmetric encryption are Rivest, Shamir, Adleman (RSA), Digital Signature
Standard (DSS) which works together with Digital Signature Algorithm (DSA), Diffie-Hellman
exchange method, and Elliptic Curve Cryptography (ECC). Asymmetric cryptography is based
on the concepts of number theory, including prime number theorems, division of integrals into
modules, modular exponent, logarithm, and the Chinese theorem (Yan, 2012). There are strong and
weak cryptographic keys. Examples of weak cryptography algorithms include Ron’s Code or Rivest
Cipher 2 (RC2), DES, and 3DES (Cambareri et al., 2015). Also, Blowfish and Advanced Encryption
Standard (AES) are some examples of strong cryptography algorithms (Cambareri et al., 2015). The
RC2 uses one 64-bit key, DES uses one 64-bit key, 3DES uses three 64-bit keys, while AES uses
(128, 192, or 256) bits keys (Cambareri et al., 2015).
Volume 12 • Issue 1
3
Data Encryption Standard is an encryption algorithm that encrypts 64-bit blocks with a 56-bit
key (Singh, 2013). As a symmetrical encryption technique, DES uses the same key for encryption
and decryption. It is an iterative algorithm, like all modern block ciphers. For each plain text block,
encryption is done in 16 rounds (Lagendijk et al., 2012). Triple Data Encryption Standard has two
and three key versions. In the two-key version, the same algorithm runs three times but uses k1 for
the first and the last step. In other words, k1 = k3, and if k1 = k2 = k3, the 3DES is a single DES.
Advanced Encryption Standard code block and key sizes vary from 128 bits, 192 bits, and 256 bits.
However, the standard requirement of AES is 128-bits of the block size (Kales & Zaverucha,
2020). Unlike DES, AES does not have a Feistel structure. Festal networks do not encrypt the entire
block of each segment. For example, in DES 64/2 = 32 bits are encrypted at the same time. Conversely,
AES encloses all 128 bits in one iteration (Kales & Zaverucha, 2020). Also, AES consists of three
different types of layers. Each layer connects all 128 bits of the data path. Similar to DES, the key
schedule computes round keys or subkeys (k0, k1, . . ., knr) from the original AES key. But the Triple
DES (3-DES) uses three fixed DES keys with a total length of 168 bits. The 3DES takes three 56-bit
keys (k1, k2, and k3). It first encrypts with k1, decrypts with next k2, and lastly encrypts with k3
(Abood & Guirguis, 2018).
Previous studies discussed the performance of various encryption algorithms, key management, and
key distribution issues. Makarenko et al. (2020) compared cryptographic algorithms in the Internet
of Things (IoT) from the perspective of energy, power consumption, memory consumption, and
throughput. They found that SPECK and XTEA are good choices for cryptographic algorithms in IoT.
Riman and Abi-Char (2015) leveraged cloud resources to implement RSA (asymmetric encryption
algorithm), MD5 (a hashing algorithm), and AES (synchronization encryption algorithm) to encrypt
large amounts of data. The purpose of their study was to assess the performance of AES, DES, RSA,
and 3DES algorithms in terms of memory usage, CPU utility time duration, and encryption speed.
The study found that RSA takes the most amount of time and MD5 takes the least for encryption.
The findings also revealed that the packet delivery ratio is highest for AES but lowest for MD5 and
RSA algorithms. Again, the input delivery ratio decreases rapidly as the input file size increases.
Similarly, Kushwaha et al. (2018) presented a comparison of the commonly used symmetric
cryptographic algorithms (AES, DES, and 3DES) in terms of power consumption. The study compared
the encryption algorithms using different data types, such as text, audio, image, and video. The
encryption algorithms were implemented in Java during the simulation. The experiments used various
file formats with file sizes between 4MB and 11MB during the encryption. The simulated results
showed that AES has a better performance than other encryption algorithms. However, 3DES has
shown poor performance as it required high processing power. Since battery power is one of the main
limitations of the management nodes, the AES encryption algorithm was found to be the best choice.
Sharma and Kushwaha (2019) implemented symmetric key encryption algorithms, including DES,
AES, and Blowfish using the NS-2 network simulator to compare their performance with different
data types on some performance metrics. In experiments, algorithms encrypt three different file types,
including text, image, and video size (0.3KB - 1KB). For each file type, performance metrics comprising
encryption and decryption time, battery usage, and throughput were measured. The results showed that
AES had the best encryption time, and Bluefish had the best performance in terms of throughput but
used more battery power than other algorithms. Also, Ahmad et al (2015) discussed the performance
and efficiency of various block cipher algorithms, DES, 3DES, CAST-128, BLOWFISH, IDEA, and
RC2 symmetric key algorithms. The parameters used were the size of the input data (in the form of
text, audio, and video), the encryption time, the encryption bandwidth of each block cipher, and power
consumption. The study found that 3DES had lower power consumption and lower bandwidth than DES.
Likewise, Norouzi and Mirzakuchaki (2016) focused on enhancing security performance in
wireless ad hoc networks with encryption algorithms and existing transmission rates. Using MATLAB,
Volume 12 • Issue 1
4
the input text files with a minimum size of 50 bytes and a maximum of 300 bytes were measured
using two methods. For the first mode, data was transferred without the use of any encryption. In the
second mode, the data were transferred using DES, AES, and Bluefish encryption methods. During
the experiments, only one key was used to encrypt and encrypt the data. Based on the results obtained,
the study recommended AES as having the highest throughput. Kashani and Mahriyar (2014) used
several cryptographic algorithms to analyze the characteristics of video streaming on ad hoc networks.
The public key infrastructure approach was chosen to provide authentication at the network layer
level. The study implemented and analyzed various cryptographic schemes, including RC4, 3DES,
AES-128, AES-256, Salsa20-128, Salsa20-256, and the time required to encrypt different data sizes.
The results showed that RC4, 3DES, AES-128, AES-256, Salsa20-128, and Salsa20-256 took less
than 1500ms to encrypt a 1MB binary file, while 3DES consumes the longest encryption time.
Moreover, Marwaha et al (2013) analyzed DES, 3DES, and RSA encryption algorithms. The
ability of algorithms to secure data, the time it takes to encrypt data, and the bandwidth required for
the algorithm were evaluated. The performance of different algorithms varies depending on the input
data. The study concluded that the privacy provided by 3DES was much higher than that of DES and
RSA. Although DES used less memory power and time to encrypt and decrypt data, its security can
be easily hacked by brute force, making it the least secured algorithm. Likewise, Seth and Mishra
(2011) ranked AES and RSA against parameters such as calculation time, memory usage, and output
bytes. The results found that RSA took the most encryption time and also consumed a lot of memory.
DES took the least encryption time, and AES had the least memory usage.
The experiment was conducted using a CORE i5 64-bit processor with 8GB of RAM and a 1T hard
drive. The simulation program was compiled using the NS-3 network simulator, installed on Ubuntu
16.0 client OS and Oracle virtual machine for windows applications. The experiment was performed
severally to confirm that the results were consistent and valid for comparison with different algorithms.
The cryptographic algorithms were implemented with the C++ program and Crypto++ package.
Crypto++ package provided security features such as encryption, decryption, key generation, key
management infrastructure, and authentication and authorization capabilities.
The experiment started with 20 nodes and later increased to 30, 40, 50, and 60. Three cryptographic
algorithms DES, 3DES, and AES were evaluated to ascertain the network performance and data
privacy. Latency, end-to-end delay, throughput, packet loss, and packet delivery ratio were the metrics
for measurement. The NS-3 simulator used consisted of a probe, collector, and aggregator as the three
basic classes. The probe is a mechanism that controls the output of simulation data used to monitor
events. Its objects are linked to one or more trace sinks or collectors. Data generated by one or more
probe objects are handled by the collector, which performs transformations such as normalization,
reduction, and computation of basic statistics on the data. An aggregator, another type of object, is
the end of the data collected by a network of probes and collectors.
The experimental process is described below:
Step 1. DES, 3DES, and AES were implemented in C++, which transferred data from the source
node to the destination node in a network topology.
Step 2. Different data file sizes of at least 1M (text, image, and audio files) were added to the
Crypto++ library.
Step 3. The data collection helpers, GnuplotHelper and FileHelper, were used to analyze the
collected data.
Volume 12 • Issue 1
5
Step 4. GnuplotHelper creates a data file, a control file, and a shell script to generate the Gnuplot.
Step 5. The results generated were then compared with each other and also with other related work.
Five network performance metrics (latency, throughput, end-to-end delay, packet delivery ratio, and
packet loss) were used to evaluate the cryptographic algorithms (DES, 3DES, AES) on three data
file formats (text, image, and audio files).
Latency. The first metric is the response time in terms of latency. To measure the time the system
takes or reacts to a given input, the experiments are run and statistics are obtained using the pmap
utility. The experiments consider a network of 20 to 60 nodes with multiple links and examine the
response time of each node. Different encryption techniques may require different response times for
implementation. The response time depends on the number of operations performed by the algorithm,
the size of the key used, the initialization vectors used, and the type of operations. Latency introduces
an overhead cost of networks and for better performance latency should be as small as possible (Zhu
et al., 2016). Therefore, developing strong cryptographic algorithms with good latency should be a
priority (Sirajuddin et al., 2022). The formulae used to compute latency are given below:
Round Trip Time (RTT) = Time taken to destination (TP) + Time taken to the source (TR)
Average Throughput. This is the number of bits passing through a point in a second over a
communication channel. It is the data transmitted over a physical or logical channel, or through a
specific network node. It is generally measured as the number of packets per second received at the
destination (Subash, 2021). Throughput can be represented with the following formula:
Throughput = minimum of {TR1, TR2, …, TRn}
End-to-End Delay. This specifies the length of time a packet travels from the source node to
the destination application layer (Bondorf & Schmitt, 2016). It is also the average time between the
start of data transmission and the arrival of the data at its destination. The time it takes for a packet
of average size to be sent out is known as transmission delay and it is measured by
Delaytr = Packet length / Transmission Rate
The time it takes for a bit to travel from source to destination in the media is known as propagation
delay and it is also measured by
Delaypg = Distance / Propagation speed.
The process by which the router removes the header, detects an error, and delivers the packet to
the upper layer is known as processing delay and it is measured by
Delaypr = Time required to process a packet in a router or destination
The time a packet waits in input or output queues in a router is measured by
Delayqu = Time a packet waits in input or output queues in a router.
Finally, the total delay can be obtained by
Delaytd = (n+1)(Delaytr + Delaypg + Delaypr) + (n)(Delayqu)
Packet Delivery Ratio. It is the ratio of the number of packets received per the number of
packets sent in the network. When the packet delivery ratio increases, the performance of the network
increases. Conversely, when the packet delivery ratio decreases, the performance of the network
decreases (Subash, 2021). The packet delivery ratio can be calculated using the following formula.
Packet delivery ratio = ∑ Number of packets received
∑ Number of packets sent
Packet Loss. Packet loss usually occurred when one or more packets within a transmission are
sent successfully from a source, but the destination host fails to receive the packets. Packet loss can
occur due to a variety of factors, including defective network components such as hardware or drivers,
network congestion, or corrupted packets within the transmission. Packet loss can be recovered through
data retransmission to the destination host to complete requests successfully (Ali et al., 2018). It can
be computed as follows:
Packet loss = 100% * (transferred packets – retransmitted packets) / transferred packets
Volume 12 • Issue 1
6
In the experiments, DES, 3DES, and AES are individually tested with three data formats for latency,
end-to-end delay, throughput, packet delivery ratio, and packet loss. The results are presented and
discussed in the following sections.
Table 1 shows how latency varies with the network size for DES, 3DES, and AES in the same
simulation environment for text, image, and audio files. At node densities of 20, 30, and 40, DES
and 3DES show constant behavior of latency for the text, image, and audio files. At node densities
of 50 and 60, the latency varies substantially among the algorithms. Figure 1 shows that AES starts
with a higher latency for image and audio files but decreases as the node density increases. This win
margin is due to AES’s strong architecture and modular design. At a node density of 20, the latency
of DES and 3DES for the text and image files was 30ms but that of AES was 40ms for image and
audio files. At node density of 60, the latency of DES and 3DES for the text and image files increased,
and that of audio files of AES decreased for text, image, and audio files at 70ms, 75ms, and 75ms
respectively. Hence, within the number of nodes used in this experiment, it can be concluded that the
network size does not negatively affect the latency when the AES algorithm is used.
Some conclusions could be drawn from the results presented in Table 1. An evaluation of the
three cryptographic algorithms revealed that the latency of AES was better than DES and 3DES at
the higher node densities for all three file formats. However, DES and 3DES were better than AES at
the lower node densities for all three file formats. Thus, the performance of different cryptographic
algorithms may vary depending on the input data (Marwaha et al, 2013) and the node densities. This
experiment provides an insight into the relationship between cryptographic algorithms and latency
and how AES, DES, and 3DES algorithms perform under latency with various node densities. While
previous research has focused on the latency of the final nodes (Kashani & Mahriyar, 2014), this
study also considered latency at the initial nodes. AES can perform better than DES and 3DES with
an increasing number of nodes, particularly in distributed systems where the number of nodes is
generally large.
Generally, at the lower node densities, AES shows the highest throughput followed by DES and 3DES
respectively (see Table 2). In addition, cryptographic algorithms have a higher throughput for smaller
node densities and lower throughput for higher node densities. For DES, the average throughput
increases and drops as the number of nodes increases. Throughputs for 3DES are constant as the
node densities increase. From Table 2, 3DES shows the highest throughput with a value of 148950,
140050, and 145550 for text, image, and audio files respectively, followed by AES and DES at a node
Table 1. Latency for text, image, and audio files
Latency for Text File Latency for Image File Latency for Audio File
Node Density DES AES 3DES DES AES 3DES DES AES 3DES
20 30 30 30 30 40 30 30 40 30
30 34 35 32 35 40 34 39 40 34
40 38 40 35 40 45 37 40 45 35
50 40 60 50 44 69 55 48 70 50
60 80 70 80 90 75 90 85 75 85
Volume 12 • Issue 1
7
density of 60. In addition, all three cryptographic algorithms have a higher throughput for smaller
node densities and lower throughput for higher node densities.
At nodes 30, 40, and 50, AES performed equally with DES and 3DES in terms of throughput
for the different file formats. As a whole, AES has a better throughput than DES and 3DES. Norouzi
and Mirzakuchaki (2016) found AES as having the highest throughput. However, DES performed
better than AES and 3DES at node 20 and node 60 (see Table 2). In this case, 3DES showed poor
performance results perhaps because it required high processing power (Ahmad et al., 2015; Kushwaha
et al., 2018).
Table 3 and Figure 3 show the end-to-end delay created by different node densities for AES, DES, and
3DES cryptographic algorithms for text, image, and audio files. At node density of 20, the end-to-end
delay started with an increased value of 0.015, 0.018, and 0.047 respectively for DES, AES, and 3DES
but dropped at node density of 30 and 40 and increased again at node density of 50 and 60 for text,
image, and audio files. From Table 3, AES and DES had the better end-to-end delay from nodes 20
to 60. However, 3DES had the worst end-to-end delay from nodes 20 to 60. Thus, for real-time data
transmission end-to-end delay can be affected by cryptographic algorithms (Ronaldo et al., 2022).
Figure 4 shows how the packet delivery ratio varies with network size for AES, DES, and 3DES for
text, image, and audio files. Observably, AES has the highest packet delivery ratio for text, image,
Figure 1. Latency for text, image, and audio files
Table 2. Throughput for text, image, and audio files
Latency for Text File Latency for Image File Latency for Audio File
Node Density DES AES 3DES DES AES 3DES DES AES 3DES
20 30289 49743 42362 30389 50043 40062 31389 50743 41362
30 62075 62340 63485 60075 61000 61000 60475 61660 61485
40 91065 89000 92000 90065 87440 82330 90065 87040 82000
50 120060 129000 130000 127060 122300 120300 127060 126600 120000
60 130000 142700 148950 130880 140000 140050 130880 140000 145550
Volume 12 • Issue 1
8
and audio files at node densities of 20, 30, 40, and 50. The results also show that the packet delivery
ratio of DES is significantly better than that of DES and 3DES due to AES’s strong architecture
and modular design. In comparison, Riman and Abi-Char (2015) found that for low input file sizes
a high packet delivery ratio is achieved for AES and the input delivery ratio normally decreases
rapidly as the input file size increases. In this study, for each input size, the packet delivery ratio is
highest for AES. The results of the study indicate a better packet delivery ratio for AES for all the
file formats from nodes 20 to 60. The low packet delivery ratio for DES and 3DES may be due to
Figure 2. Throughput for text, image, and audio files
Table 3. End-to-end delay for text, image, and audio files
Latency for Text File Latency for Image File Latency for Audio File
Node Density DES AES 3DES DES AES 3DES DES AES 3DES
20 0.015 0.018 0.047 0.015 0.018 0.047 0.015 0.018 0.047
30 0.010 0.012 0.015 0.013 0.014 0.017 0.013 0.014 0.017
40 0.023 0.010 0.013 0.026 0.018 0.014 0.026 0.018 0.014
50 0.012 0.029 0.047 0.019 0.029 0.049 0.019 0.029 0.049
60 0.014 0.025 0.070 0.018 0.029 0.076 0.018 0.029 0.076
Figure 3. End-to-End Delay for text, image, and audio files
Volume 12 • Issue 1
9
network congestion. An earlier study found that the packet delivery ratio for AES was the highest
among encryption algorithms studied (Riman & Abi-Char, 2015).
The packet loss of the three cryptographic algorithms increases with node densities, probably due
to the processes associated with the encapsulation and de-encapsulation. Table 5 shows the packet
loss for AES, DES, and 3DES. For different node densities of 20 to 60, the packet loss for all three
cryptographic algorithms was the same at a node density of 20 to 40 with a zero value. The value
varied at a node density of 50 to 60 with a constant value of 15, 18, and 20 for text, image, and audio
files respectively for DES, whereas AES had a constant value of 25, 23, and 25 for text, image, and
audio files respectively. However, 3DES had a constant value of 20, 21, and 20 for text, image, and
audio files respectively. This observation was a result of each module requiring its buffer stack, which
leads to a higher packet loss for the simulation program. Meanwhile, it was observed at nodes 50 and
60 that, AES had a constant value of the three cryptographic algorithms.
The lower the packet loss, the better the network performance. The results, as presented in Table
5, shows that AES cryptographic algorithm gave the highest packet loss for the various file formats at
nodes 50 and 60, perhaps due to its complex mathematical computation leading to network congestion.
At nodes 20 to 40 all the cryptographic algorithms had no packet loss. For audio files, DES and 3DES
had the same packet loss but DES had a lower value than 3DES for packet loss at nodes 50 and 60.
This is contrary to an earlier study that found DES to have high packet loss and therefore can easily
be hacked by brute force, making it the least secured algorithm (Marwaha et al, 2013).
Table 4. Packet delivery ratio for text, image, and audio files
Packet Delivery Ratio for Text File Packet Delivery Ratio for
Image File
Packet Delivery Ratio for Audio
File
Node
Density
DES AES 3DES DES AES 3DES DES AES 3DES
20 90.309493 100.065728 101.472186 90.30949 100.0657 90.30949 90.30949 100.0657 90.30949
30 88.627517 100.807024 100.733028 89.44752 100.8070 89.44752 89.50752 100.8066 89.33752
40 85.931008 100.875719 100.114439 82.88601 100.8757 84.11444 81.83401 100.8857 85.32444
50 72.427575 60.273858 45.578424 82.42756 60.27386 65.57846 82.42756 60.27446 67.56646
60 61.641176 54.428822 40.854015 61.77118 54.42882 45.46402 61.31117 64.42882 55.46402
Figure 4. Packet delivery ratio for text, image, and audio files
Volume 12 • Issue 1
10
This study evaluated the impact of three cryptographic algorithms on network performance. The NS-3
simulator was used to evaluate the latency, throughput, end-to-end delay, packet delivery ratio, and
packet loss for DES, 3DES, and AES. Several experiments were performed on the three file formats
(text, image, and audio). The study generated six test cases for AES, DES, and 3DES by modifying
the values of the node densities to examine the impact on the latency, throughput, end-to-end delay,
packet delivery ratio, and packet loss. An encrypted data of different file types was transmitted across
the different node densities and the output was analyzed with Gnuplot.
The results showed that the performance of AES was better than DES and 3DES for a large
number of node densities for latency, data throughput, end-to-end delay, and packet delivery ratio.
However, DES had the lowest packet loss as AES cryptographic algorithm gave the highest packet
loss. This study revealed how cryptographic algorithms could impact network performance. However,
the study examined only three algorithms, future work will analyze other cryptographic algorithms to
ascertain the effect they might have on network performance. Moreover, although the node densities
were limited (20 to 60 nodes), the results of the study could provide meaningful findings and insights
that could be generalized to large node densities. Future studies would examine increasing the number
of nodes to ascertain the overhead costs of several cryptographic algorithms.
Table 5. Packet loss for data privacy
Volume 12 • Issue 1
11
Abood, O. G., & Guirguis, S. K. (2018). A survey on cryptography algorithms. International Journal of Scientific
and Research Publications, 8(7), 410–415.
Agrawal, M., & Mishra, P. (2012). A comparative survey on symmetric key encryption techniques. International
Journal on Computer Science and Engineering, 4(5), 877.
Ahmad, M., Khan, I. R., & Alam, S. (2015). Cryptanalysis of image encryption algorithm based on the
fractional-order Lorenz-like chaotic system. Emerging ICT for Bridging the Future-Proceedings of the 49th
Annual Convention of the Computer Society of India CSI, 2, 381–388.
Ali, B., Sher, A., Javaid, N., Islam, S. U., Aurangzeb, K., & Haider, S. I. (2018). Retransmission avoidance for
reliable data delivery in underwater WSNs. Sensors (Basel), 18(1), 149. doi:10.3390/s18010149 PMID:29316664
Bondorf, S., & Schmitt, J. (2016). Calculating accurate end-to-end delay bounds-you better know your cross-
traffic. Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and
Tools, 17–24. doi:10.4108/eai.14-12-2015.2262565
Braeken, A. (2022). Public key versus symmetric key cryptography in client-server authentication protocols.
International Journal of Information Security, 21(1), 103–114. doi:10.1007/s10207-021-00543-w
Cambareri, V., Mangia, M., Pareschi, F., Rovatti, R., & Setti, G. (2015). Low-Complexity Multiclass
Encryption by Compressed Sensing. IEEE Transactions on Signal Processing, 63(9), 2183–2195. doi:10.1109/
TSP.2015.2407315
Elhoseny, M., Shankar, K., Lakshmanaprabu, S. K., Maseleno, A., & Arunkumar, N. (2020). Hybrid optimization
with cryptography encryption for medical image security in Internet of Things. Neural Computing & Applications,
32(15), 10979–10993. doi:10.1007/s00521-018-3801-x
Fazzat, A., Khatoun, R., Labiod, H., & Dubois, R. (2020). A comparative performance study of cryptographic
algorithms for connected vehicles. 2020 4th Cyber Security in Networking Conference (CSNet). doi:10.1109/
CSNet50428.2020.9265529
Gueron, S. (2016). A memory encryption engine suitable for general-purpose processors. IACR Cryptol. EPrint
Arch., 2016, 204.
Jing, Q., Vasilakos, A. V., Wan, J., Lu, J., & Qiu, D. (2014). Security of the Internet of Things: Perspectives and
challenges. Wireless Networks, 20(8), 2481–2501. doi:10.1007/s11276-014-0761-7
Jintcharadze, E., Sarajishvili, T., Surmanidze, A., & Khojava, D. (2021). Implementation and Comparative
Analysis of Symmetric Encryption Model Based on Substitution Cipher Techniques. 2021 IEEE East-West
Design & Test Symposium (EWDTS), 1-6. doi:10.1109/EWDTS52692.2021.9580978
Kales, D., & Zaverucha, G. (2020). Improving the Performance of the Picnic Signature Scheme. IACR Cryptol.
EPrint Arch., 2020, 427.
Kashani, M. A. A., & Mahriyar, H. (2014). A new method for preventing wormhole attacks in wireless sensor
networks. Advances in Environmental Biology, 8(10), 1339–1346.
Kumar, Y., Munjal, R., & Sharma, H. (2011). Comparison of symmetric and asymmetric cryptography with
existing vulnerabilities and countermeasures. International Journal of Computer Science and Management
Studies, 11(03), 60–63.
Kumari, S. (2017). A research paper on Cryptography Encryption and Compression Techniques. International
Journal of Engineering and Computer Science, 6(4). Advance online publication. http://103.53.42.157/index.
php/ijecs/article/view/3630. doi:10.18535/ijecs/v6i4.20
Kushwaha, A., Sharma, H. R., & Ambhaikar, A. (2018). Selective encryption using natural language processing
for text data in a mobile ad hoc network. In Modeling, Simulation, and Optimization (pp. 15–26). Springer.
doi:10.1007/978-3-319-70542-2_2
Lagendijk, R. L., Erkin, Z., & Barni, M. (2012). Encrypted signal processing for privacy protection: Conveying
the utility of homomorphic encryption and multiparty computation. IEEE Signal Processing Magazine, 30(1),
82–105. doi:10.1109/MSP.2012.2219653
Volume 12 • Issue 1
12
Laszka, A., Vorobeychik, Y., & Koutsoukos, X. (2018). A game-theoretic approach for integrity assurance
in resource-bounded systems. International Journal of Information Security, 17(2), 221–242. doi:10.1007/
s10207-017-0364-2
Makarenko, I., Semushin, S., Suhai, S., Ahsan Kazmi, S. M., Oracevic, A., & Hussain, R. (2020). A Comparative
Analysis of Cryptographic Algorithms in the Internet of Things. 2020 International Scientific and Technical
Conference Modern Computer Network Technologies (MoNeTeC). doi:10.1109/MoNeTeC49726.2020.9258156
Marwaha, M., Bedi, R., Singh, A., & Singh, T. (2013). Comparative analysis of cryptographic algorithms. Int
J Adv Engg Tech, 16, 18.
Mohindra, A. R., & Gandhi, C. (2021). A Secure Cryptography Based Clustering Mechanism for Improving
the Data Transmission in MANET. Walailak Journal of Science and Technology, 18(6), 1–18. doi:10.48048/
wjst.2021.8987
Mushtaq, M. F., Jamel, S., Disina, A. H., Pindar, Z. A., Shakir, N. S. A., & Deris, M. M. (2017). A survey on the
cryptographic encryption algorithms. International Journal of Advanced Computer Science and Applications,
8(11), 333–344.
Norouzi, B., & Mirzakuchaki, S. (2016). Breaking an image encryption algorithm based on the new substitution
stage with chaotic functions. Optik (Stuttgart), 127(14), 5695–5701. doi:10.1016/j.ijleo.2016.03.076
Pal, S. K., Datta, B., & Karmakar, A. (2022). An Artificial Neural Network Technique of Modern Cryptography.
Journal of Scientific Research, 14(2), 471–481. doi:10.3329/jsr.v14i2.55669
Patil, P., Narayankar, P., Narayan, D. G., & Meena, S. M. (2016). A comprehensive evaluation of cryptographic
algorithms: DES, 3DES, AES, RSA, and Blowfish. Procedia Computer Science, 78, 617–624. doi:10.1016/j.
procs.2016.02.108
Ramesh, A., & Suruliandi, A. (2013). Performance analysis of encryption algorithms for Information Security.
2013 International Conference on Circuits, Power and Computing Technologies (ICCPCT), 840–844. doi:10.1109/
ICCPCT.2013.6528957
Riman, C., & Abi-Char, P. E. (2015). Comparative Analysis of Block Cipher-Based Encryption Algorithms: A
Survey. Information Security and Computer Fraud, 3(1), 1–7.
Rojasree, V., & Gnanajayanthi, J. (2020). Cryptographic Algorithms to Secure Networks - A Technical Survey
on Research Perspectives. 2020 Third International Conference on Smart Systems and Inventive Technology
(ICSSIT). doi:10.1109/ICSSIT48917.2020.9214259
Ronaldo, F., Sudarsono, A., & Pramadihanto, D. (2022). Secure Real-time Data Transmission for Drone
Delivery Services using Forward Prediction Scheduling SCTP. EMITTER International Journal of Engineering
Technology, 120–135.
Seth, S. M., & Mishra, R. (2011). Comparative analysis of encryption algorithms for data communication.
Academic Press.
Shaktawat, R., Shaktawat, R. S., Lakshmi, N., Panwar, A., & Vaishnav, A. (2020). A Hybrid Technique of
Combining AES Algorithm with Block Permutation for Image Encryption. Reliability. Theory & Applications,
15(1), 51–56.
Sharma, A., & Kushwaha, G. (2019). Comparative Analysis of Different Encryption Techniques in Mobile Ad-
Hoc Networks (MANETs). IITM Journal of Management and IT, 10(1), 55–64.
Singh, G. (2013). A study of encryption algorithms (RSA, DES, 3DES, and AES) for information security.
International Journal of Computers and Applications, 67(19).
Sirajuddin, M., Rupa, C., Bhatia, S., Thakur, R. N., & Mashat, A. (2022). Hybrid Cryptographic Scheme for
Secure Communication in Mobile Ad Hoc Network-Based E-Healthcare System. Wireless Communications and
Mobile Computing, 2022, 2022. doi:10.1155/2022/9134036
Subash, K. (2021). Energy-Aware Path Selectio.n to Improve Packet Delivery Ratio and Throughput in RPL
Networks. Turkish Journal of Computer and Mathematics Education, 12(9), 2301-2307.
Volume 12 • Issue 1
13
Taha, A. A., Elminaam, D. S. A., & Hosny, K. M. (2018). An improved security schema for mobile cloud
computing using hybrid cryptographic algorithms. Far East Journal of Electronics and Communications, 18(4),
521–546. doi:10.17654/EC018040521
Unal, D., Al-Ali, A., Catak, F. O., & Hammoudeh, M. (2021). A secure and efficient internet of things cloud
encryption scheme with forensics investigation compatibility based on identity-based encryption. Future
Generation Computer Systems, 125, 433–445. doi:10.1016/j.future.2021.06.050
Yan, S. Y. (2012). Computational Number Theory and Modern Cryptography. John Wiley & Sons.
doi:10.1002/9781118188606
Zhang, S., & Liu, L. (2021). A novel image encryption algorithm based on SPWLCM and DNA coding.
Mathematics and Computers in Simulation, 190, 723–744. doi:10.1016/j.matcom.2021.06.012
Zhu, T., Berger, D. S., & Harchol-Balter, M. (2016). SNC-Meister: Admitting more tenants with tail latency
SLOs. Proceedings of the Seventh ACM Symposium on Cloud Computing, 374–387.
Volume 12 • Issue 1
14
3DES Triple Data Encryption Standard
ACK Acknowledgement
AES Advanced Encryption Standard
CA Certification Authority
CPU Central Processing Unit
Delaypg Propagation Delay
Delaypr Processing Delay
Delayqu The time a packet waits in input or output queues in a router
Delaytd Total Delay
Delaytr Transmission Delay
DES Data Encryption Standard
DH Diffie-Hellman
DHKE Diffie-Hellman Key Exchange
DSA Digital Signature Algorithm
DSS Digital Signature Standard
EAACK Enhanced Adaptive Acknowledgment
ECC Elliptic Curve Cryptography
ECDH Enhanced Elliptic Curve Diffie-Hellman
IDEA International Data Encryption Algorithm
IDS Intrusion Detection System
IP Internet Protocol
MANET Mobile Ad-Hoc Network
MATLAB Matrix Laboratory
MD5 Message-Digest Algorithm 5
MRA Misconduct Report Authentication
ms Milliseconds
NS2 Network Simulator 2
NS3 Network Simulator 3
OLSR Optimized Link State Routing
OS Operating System
PMAP Performance Management Appraisal Program
RAM Random Access Memory
RC2 Ron’s Code or Rivest Cipher 2
RSA Rivest, Shamir, Adleman
RTT Round Trip Time
SACK Secure Acknowledgement
TCP Transmission Control Protocol
TP Time taken to destination
TR Time taken to the source
Volume 12 • Issue 1
15
Samuel Asare is a final-year M.Sc. Computer Science student at the University of Ghana, specializing in Computer
Science. Like many other Computer Science students, he uses technology to solve problems and make plans
for the future. However, he is primarily concerned with validating and developing models for human-computer or
software-device interaction. Samuel graduated from West End University College with a B.Sc. Computer Science
and went on to serve as an IT officer before becoming the University College’s IT manager. Samuel obtained
CCNA and CCNP certification in 2009 before he decided to join West End University College. He also spent four
years at 1st October Network Academy as a Cisco instructor and network technician. He has, however, trained
over 500 students for the CCNA certification, and they have all passed.
Winfred Yaokumah is a researcher and senior faculty at the Department of Computer Science of the University
of Ghana. He has published several articles in highly rated journals including Information and Computer Security,
Information Resources Management Journal, IEEE Xplore, International Journal of Human Capital and Information
Technology, International Journal of Human Capital and Information Technology Professionals, International Journal
of Technology and Human Interaction, International Journal of e-Business Research, International Journal of
Enterprise Information Systems, Journal of Information Technology Research, International Journal of Information
Systems in the Service Sector, and Education and Information Technologies. His research interest includes cyber
security, cyber ethics, network security, and information systems security and governance. He serves as a member
of the International Review Board for the International Journal of Technology Diffusion.
Ernest Barfo Boadi Gyebi was born in Ghana in 1960. He received BSc degree in Engineering from Kwame Nkrumah
University of Science and Technology, Kumasi in 1987, MSc degree in Information Systems Engineering from
South Bank University, London in 1997 and PhD degree in Educational Robotics from University of Lincoln, UK
in 2018. He joined the University of Ghana as a lecturer at the Department of Computer science in 2003 and has
been teaching at the premier University in Ghana ever since. His main areas of research interest are educational
robotics, artificial intelligence and computer systems.
Jamal-Deen Abdulai is senior lecturer of Computer Science, at Dept. of Computer Science, University of Ghana
with over 12 years of working experience in both industry and academia. He graduated with a PhD in Computer
Science in 2009 from University of Glasgow, UK. Prior to that, he received BSc Computer Science in 2002 from the
Kwame Nkrumah University of Science and Technology (KNUST). Prior to his appointment at University of Ghana,
Dr Abdulai had worked with the School of Technology, GIMPA as a lecturer from 2009 to 2013 and worked at the
University for Development Studies (UDS) as Senior Research Assistant from 2002 to 2004. Dr Abdulai current
research interest includes Performance modelling and evaluation of Mobile Wireless Ad hoc and Sensor Networks,
Network Security and Management, Embedded Systems, Parallel and Distributed Systems, Artificial Intelligence
and its applications. Dr Abdulai has been providing consultancy services to public and private organisations in
Ghana over the last 10 years. Among them are the Ministry of Finance and the Ministry of Roads and Highways.
