IRAQI JOURNAL OF COMPUTERS, COMMUNICATIONS, CONTROL AND SYSTEMS ENGINEERING

Providing security for each online consumer over the internet is a
critical issue that may cause a time consuming problem that may cause big
load on the website server especially for the large websites at the rush time.
This process may generate a variety of issues, including response time delays,
client orders being lost, and system crash or deadlock, all of which decrease
system performance. This work intends to present a new multi-agent system
prototype structure that solves the challenge of security while avoiding issues
that might degrade system performance. This is accomplished by installing a
software agent on the client's device that handles the purchase and encryption
processes without the need for the user to intervene. The suggested agent
evades the problems of stalemate (i.e., failure) and request loss, ensuring that
information exchanged between all entities is protected. The use of a software
agent to manage buying and encrypting operations improves system
performance by 10% and increases the reaction time of the system by 30.5
percent (response time, page loading time, transaction processing speed,
orders per second) according to test results.

Wireless Sensor Networks (WSNs) have become the most cost-
effective monitoring solution due to their low cost, despite their major drawback
of limited power due to dependence on batteries. Each Sensor Node (SN) is
clustered in a particular location and forms a network by self-organizing. They
often operate in some of the world's most unusual or dangerous conditions.
Networking errors, memory and processor limitations, and energy constraints
all pose problems for WSN developers. Many problems in WSNs are expressed
as multivariate optimization problems that are solved using biologically
inspired techniques. Particle swarm optimization (PSO) is an easy,
algorithmically sound, and robust optimization technique. It has been used to
address problems like Clustering, data routing, Cluster Head (CH) collection,
and data collecting in WSNs. This paper presents a brief analysis of WSN
studies in which the PSO algorithm was used as the primary or secondary
algorithm for enhancing lifespan of WSNs, focusing on results that show energy
efficiency in the sensors, extending the network's life.

Communication and computing systems have made it easier for the
world to transfer data and information from the sender to the recipient at the
lowest cost and most efficiency. The transmission process may cause data
corruption or error for many reasons, including the environment, the large
volume of transmitted data, heat, and noise. for these Reasons, There is a need
to correct and treat these errors. Individual errors can be easily corrected by
hamming code, while burst errors cannot be corrected easily and need a
hardware device called the interleaver used to correct the burst error. In this
research, the different types of interleaver are studied and compared to find the
best interleaver in order to increase the efficiency and accuracy of the systems.
The issue is that interleaving takes a long time, which increases the turbo code's
overall execution time. Our goal is to create an interleaver that is more
sensitive and efficient than other varieties.

Permanent Magnet Synchronous Motors (PMSM) are extensively used in the
industry owing to their excellent efficiency, low weight/power ratio, and smooth torque
with no or minimal ripple. Field Oriented Control (FOC) is a modern and effective
approach for closed-loop controlling the speed of PMSM. In this paper, three-level Space
Vector Pulse Width Modulation (SVPWM) is proposed for minimizing harmonics in the
output voltage inverter. sensorless approaches are performed by using Model Reference
Adaptive System (MRAS) which eliminates mechanical uncertainties. Because
mechanical sensors increase the cost, size, weight, and wiring complexity, employing
PMSM with them is extremely difficult Tuning of Proportional Integral (PI) controller
gains is performed by using the Whale Optimization Algorithm (WOA). The results show
that the proposed controller enhances the system's performance. In the application of
felid-oriented control to a PMSM, with simulation data to back it up the entire system is
simulated using the MATLAB/Simulink tool.

In most applications, electric drives are actuated using on/off
devices due to their low cost and also due to the relatively high power
consumption of the electric drives which make applying linear power amplifiers
very costly. In this paper, the operation of PMDC motors under discontinuous
control action is analyzed. In addition, to reduce chattering, boundary layer
solution has been addressed. Both suggested control techniques have been
applied to a PMDC motor model in a software simulation using MATLAB. The
results show better performance of boundary layer technique due to the reduced
chattering.

A mobile robot's major purpose is to get to its destination by
traveling over an optimum path defined by various parameters such as time,
distance, and the robot's safety from any impediments in its path. As a result,
the backbone of the autonomous mobile robot is path planning and obstacle
avoidance. Several algorithms for path planning and obstacle avoidance have
been presented by various researchers, each with its own set of benefits and
drawbacks. This paper focuses on two parts; the first part finds the short and
smooth collision-free path for a mobile robot to navigate in a static environment
based on two proposed hybrid algorithms. The first hybrid is between Firefly
Algorithm (FA) and Modify Chaotic Particle Swarm Optimization (MCPSO),
namely (HFACPSO), while the other hybrid is between Genetic Algorithm (GA)
and MCPSO, namely (HGACPSO). The second part suggests an algorithm
planner for improving the efficiency of the route-planning algorithm with
moving obstacle avoidance by adjusting the velocity or re-planning the path for
the mobile robot. To demonstrate the effectiveness of the proposed algorithms in
terms of the shortest path length and collision-free, as well as obtaining optimal
or near-optimal wheel velocities with the minimum number of iterations. The
proposed hybrid (FAMCPSO) algorithm provides enhancement on the path
length equal to (0.82%) compared to the firefly algorithm (FA). Moreover, the
hybrid (GAMCPSO) algorithm enhancement on the path length equals (0.67%)
compared to the genetic algorithm (GA). All methods are simulated in a static
and dynamic obstacle environment using MATLAB 2018b.

A new technique called cognitive radio seeks to utilize the available
spectrum.. Spectrum sensing is the fundamental cognitive radio component.
There are many types of sensing spectrums, one of which is The Two Thresholds
Based on Covariance Absolute Values (TTCAV) method. This method's
confused region, which is unsure whether it is a signal or noise, is one of its
drawbacks. To solve this problem proposed two techniques, a 2-bit quantization
technique and a 3-bit quantization technique. Both techniques use an adaptive
threshold. The experimental results conducted using python version 3.7 and
Jetson Nano kit show an improvement in the detection probability values after
using the techniques. where the value of ???????? = 80.2% for a 2-bit technique,
????????=90.3% for a 3-bit quantization technique, at SNR=-20 under smoothing
factor L=3. The Monte Carlo was used to determine the effectiveness of the
proposed techniques, and Binary Phase Shift Keying modulation BPSK.
According to the experiment, the results after using the proposed techniques are
better than before.

As the Internet of Things (IoT) is growing in popularity globally, which has
resulted in a rise in cyber threats, experts are focusing more on its security. The majority of
IoT security research to date has concentrated on huge devices, while small IoT devices
have received comparably little attention. Our primary purpose is, therefore, to research
how to ensure the operation of IoT devices that are small. The security gateway is a
Security Settings on the Gateway for RaspberryPi built gateway that may link Internet of
Things devices to their private network, safeguarding IoT devices from exposure to external
networks. In addition, a variety of Security Settings on the Gateway for RaspberryPi
security settings are installed, including fiel2ban and a Security Settings on the Gateway
for RaspberryPi firewall, in order to avoid brute force and dictionary attacks. This article
also studies the communication between Internet of Things (IoT) devices utilizing various
secure communications, including Secure Shell (SSH), and analyzes their performance in a
variety of circumstances. The gateway's experimental evaluation reveals that the proposed
framework can secure tiny IoT devices.

Sign language (SL) is Non-verbal communication and a way for the
deaf and mute to communicate without words. A deaf and mute person's hands,
face, and body shows what they want to say. Since the number of deaf and dumb
people is increasing, there must be other ways to learn sign language or
communicate with deaf and dumb people. One of these ways is using advanced
technology to produce systems that help the deaf/dumb, such as creating
recognition and sign language translators. This paper presents an application
that works on the computer for machine translation of Iraqi sign language in
two directions from sign language to Arabic language (text/speech) and from
Arabic language(text) to Iraqi sign language. The proposed system uses a
Convolution Neural Network (CNN) to classify sign language based on its
features to predicate the sign meaning. The sign language to Arabic
language(text/speech) part of the proposed system has an accuracy of 99.3% for
letters.

Physiotherapeutic exoskeleton devices have recently been developed to help
people rehabilitate impaired limb mobility and replace the use of physiotherapists. Such
systems are characterized by high nonlinear and time-varying coefficients. In order to
cope with such difficult control challenges, a need arose for reliable nonlinear
controllers. While in this study the Sliding Mode Control (SMC) was used to track the
trajectory of the knee exoskeleton-system (KES) while having parameter uncertainty. In
addition, the whale optimization algorithm (WOA) was introduced and developed to
adjust the thickness design parameters for further optimization of its performance. The
simulation was performed on a calculator using the MATLAB-Simulink program to
conduct a comparative study between the optimal and Classical SMC where the results
of comparison with the test parameters used by the SMC showed, the results of the
proposed optimal SMC revealed that the positioning inaccuracy of the knee increased by
31.8807% and it follows from this result that the controller could successfully perform
tracking the track well. Also, the control system created at the optimal thickness has a
better dynamic performance than the classical thickness.

Target detection, one of the key functions of computer vision, has
grown in importance as a study area over the past two decades and is currently
often employed. In a certain video, it seeks to rapidly and precisely detect and
locate a huge amount of the objects according to redetermined categories. The
two forms of deep learning (DL) algorithms that are used in the model training
algorithm are single-stage and 2-stage algorithms of detection. The representative
algorithms for every level have been thoroughly discussed in this work. The
analysis and comparison of numerous representative algorithms in this subject is
after that explained. Last but not least, potential obstacles to target detection are
anticipated.

From big data analytics to computer vision and human-level control,
deep learning has been effectively applied to a wide range of complicated
challenges. However, these same deep learning advancements have also been
used to develop malicious software that threatens individuals\' personal data,
democratic processes, and even national security. Apps backed by deep
learning have lately appeared, with deepfake being one of the most notable.
Deepfake algorithms can create fake images and videos that humans cannot
distinguish them from authentic ones. One of the fields that deep learning
accomplished major success is face synthesis and animation generation. On the
other hand, it can create unethical software called deepfake that presents a
severe privacy threat or even a huge security risk that can affect innocent
people. This work introduces the most recent algorithms and methods used in
deepfake. In addition, it provides a brief explanation of the principles that
underpin these technologies and facilitates the development of this field by
identifying the challenges and scopes that require further investigation in the
future.

Software-Defined Network (SDN) is one of the most predominant
technologies for networking in the existing and next-generation networks. SDN
can conFig. , control, protect, and optimize network resources through
software. The fundamental benefit of SDN is enabling the application of
dynamic management. In addition, the literature shows that partitioning a
Software-Defined Wide Area Network (SD-WAN) into several logical networks
efficiently will optimize its performance. The main aim of paper is to design an
algorithm to slice SD-WAN dynamically into several virtual networks according
to the server-clients’ correlation using the Virtual Local Area Network (VLAN).
The several virtual networks improve QoS of SD-WAN and reduce its
broadcasting domain. The proposed framework consists of two parts. The first
part is the management algorithm that finds the best server for each client; then
it groups this server with their client in a dedicated logical network. The second
part includes creating a VLAN for each logical network in an SD-WAN. The
application of the POX controller calculates and maintains the dynamic VLAN,
which will be applied by the control plan to slice the topology in the data plan.
SD-WAN topology is tested before and after applying VLANs. The results show
enhancement in latency by 42.85%, throughput by 4.61%, loss packet by 72%
and jitter by 47.86% after applying VLAN. Finally, the greatest gain is reducing
the broadcasting ratio by 77.77%.

Upper limb amputation is a condition that severely limits the amputee’s
movement. Patients who have lost the use of one or more of their upper extremities have
difficulty performing activities of daily living. To help improve the control of upper limb
prosthesis with pattern recognition, non-invasive approaches (EEG and EMG signals) is
proposed in this paper and are integrated with machine learning techniques to recognize
the upper-limb motions of subjects. EMG and EEG signals are combined, and five features
are utilized to classify seven hand movements such as (wrist flexion (WF), outward part of
the wrist (WE), hand open (HO), hand close (HC), pronation (PRO), supination (SUP),
and rest (RST)). Experiments demonstrate that using mean absolute value (MAV),
waveform length (WL), Wilson Amplitude (WAMP), Sine Slope Changes (SSC), and
Cardinality features of the proposed algorithm achieves a classification accuracy of 89.6%
when classifying seven distinct types of hand and wrist movement.

Deep Dream (DD) is a new technology that works as a creative
image-editing approach by employing the representations of CNN to produce
dreams-like images by taking the benefits of both Deep CNN and Inception to
build the dream through layer-by-layer implementation. As the days go by, the
DD becomes widely used in the artificial intelligence (AI) fields. This paper is
the first systematic review of DD. We focused on the definition, importance,
background, and applications of DD. Natural language processing (NLP),
images, videos, and audio are the main fields in which DD is applied. We also
discussed the main concepts of the DD, like transfer learning and Inception. We
addressed the contributions, databases, and techniques that have been used to
build the models, the limitations, and evaluation metrics for each one of the
included research papers. Finally, some interesting recommendations have
been listed to serve the researchers in the future.

The use of video surveillance systems has increased due to security
concerns and their relatively low cost. Researchers are working to create
intelligent Closed Circuit Television (CCTV) cameras that can automatically
analyze behavior in real-time to detect anomalous behaviors and prevent
approaches, particularly
dangerous accidents.
(DL)
Learning
Deep
Convolutional Neural Networks (CNNs), have shown outstanding results in
video analysis and anomaly detection. This research paper focused on using
Inception-v3 transfer learning approaches to improve the accuracy and
efficiency of abnormal behavior detection in video surveillance. The Inception-
v3 network is used to classify keyframes of a video as normal or abnormal
behaviors by utilizing both pre-training and fine-tuning transfer learning
approaches to extract features from the input data and develop a new classifier.
The UCF-Crime dataset is used to train and evaluate the proposed models. The
performance of both models was evaluated using accuracy, recall, precision,
and F1 score. The fine-tuned model achieved 88.0%, 89.24%, 85.83%, and
87.50% for these measures, respectively. In contrast, the pre-trained model
obtained 86.2%, 86.43%, 84.62%, and 85.52%, respectively. These results
demonstrate that transfer learning using Inception-v3 architecture can
effectively classify normal and abnormal behaviors in videos, and fine-tuning
the weights of the layers can further improve the model's performance.