IRAQI JOURNAL OF COMPUTERS, COMMUNICATIONS, CONTROL AND SYSTEMS ENGINEERING

Model uncertainties and exogenous disturbances are inherent
phenomena in chemical plants that degrade performance and lead
to poor-quality products. To mitigate these issues, this paper
addresses disturbances by exploiting the differential flatness
property (DFP) of the nonlinear continuous stirred tank reactor
(NCSTR). A two-degree-of-freedom (2DOF) control design was
implemented, incorporating flatness-based feedforward and
feedback controllers. This design leverages the differential
flatness method for effective stabilization and tracking
performance. A robust feedback tracking control strategy
integrating a disturbance observer (DO) and a simple feedback
controller was implemented to form a flatness active disturbance
rejection control (FADRC) scheme. The B-spline technique was
used for trajectory generation to enhance the performance of the
proposed control strategy. Additionally, constraints were handled
symbolically rather than numerically by designing B-spline curves
for the system's flat output. This approach ensures a feasible
open-loop reference trajectory that satisfies the feedback
controller's states and input constraints. A comparison was
conducted between the switching function-based (SF) trajectory
and the B-spline (Bs) designed reference trajectory regarding
their impact on system tracking performance, smoothness, and
robustness to disturbances. A detailed analysis evaluated tracking errors, control effort, and disturbance rejection capabilities under
both trajectory designs. The proposed flatness-based ADRC
methodology efficiently managed exogenous disturbances and
constraints on the input of a nonlinear plant.

The k-nearest neighbors (kNN) algorithm is widely
adopted for classification due to its simplicity and
effectiveness. However, its computational cost remains a
significant challenge, particularly for resource-
constrained environments with limited processing power
and memory. This issue is addressed by proposing the
1
Group-Based Sample Partitioning (????????
−kNN) Algorithm,
which introduces a two-phase approach to reduce
computational complexity while maintaining classification
accuracy. In the first phase, the algorithm pre-groups
training samples by iteratively selecting anchor points and
partitioning their k-nearest neighbors, thereby reducing
redundancy in the dataset. In the second phase, the test
sample dynamically selects local anchor points,
constructing a smaller, more relevant neighborhood for
efficient classification. Experimental results using the
Breast Cancer Dataset from the UCI repository (WBC)
1
demonstrate that ????????
−kNN significantly reduces training
and testing iterations while preserving high classification
accuracy (95.78%), with a recall of 100%. Compared to
exhaustive kNN, our approach achieves a substantial reduction in distance computations approximately 78%
lower without requiring additional memory. While the
algorithm was tested on a relatively small dataset,
1
????????
−kNN shows promise for scalable implementation in
embedded systems. Also, the proposed approach shows the
computational cost can be reduced by over 75% for larger
datasets when different datasets ranging from 100 to
30,000 samples were tested. This work specifically targets
tabular datasets suitable for resource-constrained
embedded systems. Therefore, the comparison primarily
emphasizes exhaustive kNN, which serves as a clear
baseline for computational complexity evaluation.
Nevertheless, we have also provided comparisons with
related works, highlighting methodological distinctions
and similarities explicitly. Future work will explore an
extended ????????
????
−kNN framework, introducing multiple k-
parameters for adaptive scaling to high-dimensional
datasets while maintaining computational efficiency.
https://github.com/AyadMDalloo/kg-kNN.

Deep fake videos are a threat to information integrity by
using advanced deep learning techniques to manipulate
visual content and make it hard to distinguish from real
videos. This includes videos where a person’s image is
changed or replaced using deep learning techniques. In this
paper, we propose a hybrid approach for deep fake video
detection that combines spatial and temporal features. We
use a pre-trained VGG16 to extract deep spatial features
from individual frames, and then global average pooling at
both spatial and temporal level to convert variable length
video sequences into fixed dimensional representations. A
fully connected classifier with dropout regularization is then
used to classify the video as real or fake. We evaluated our
method on the DFDC dataset and got 95% accuracy and
0.81 AUC. Our framework not only uses transfer learning
to reduce the need for large amount of training data but also
computational efficiency, so it’s suitable for large scale
deep fake detection.

The emergence of quantum computers, coupled with
relentless advancements in technology, calls for the
immediate transition from classical to modern cryptography
for efficient data transfer across optic fibers. This paper
assesses different proposed post-quantum signature schemes
pertaining to their plausibility of being useful against
quantum computers’ capabilities. The ability of quantum
computers to solve mathematical equations at speeds
significantly faster than those of classical computers is
remarkable. Existing cryptographic frameworks that depend
on discrete logarithms and integer factorization are
seriously threatened by this exceptional efficiency. In order
to protect sensitive data in a post-quantum world, it may be
necessary to reevaluate existing encryption techniques as the
introduction of quantum computing has the potential to
seriously compromise the security of many popular
cryptographic systems. Hence, during the post-quantum
epoch, the implementation of cryptographic measures to
safeguard the confidentiality, integrity, and authenticity of
all data amid the challenges instigated by quantum
computers is an undertaking that must be addressed
tactically and strategically. This document endeavors to
depict the state-of-the-art methodologies in post-quantum
cryptography outlining lattice, hash, code, and multivariate
polynomial lattice-based schemes. By discussing their
advantages, disadvantages, and implementation hurdles, this
work seeks to educate academics and industry experts about
the steps that must be taken to safeguard digital
communications in the age of quantum technology.

Emotion recognition through body movement presents a
compelling alternative to traditional facial and vocal analysis in
the field of emotion recognition . This study introduces a deep
learning-based framework that utilizes upper-body movements
features to classify emotional states using Convolutional Neural
Networks (CNNs). The system is trained and evaluated on the Body
Language Dataset (BoLD) dataset, targeting seven primary
emotions: happiness, sadness, anger, fear, surprise, joy, and
disgust. The proposed model achieves a classification accuracy of
95.72%, significantly outperforming existing methods in body-
based emotion recognition. The result highlights the potential of
body posture as a reliable and standalone modality for emotion
detection, especially in contexts where facial information is
ambiguous or unavailable. The presented work contributes to the
advancement of non-intrusive, vision-based affective computing
systems, with promising applications in human-computer
interaction, behavioral analysis, security, and assistive
technologies. Future research may explore multimodal integration,
real-time deployment, and cross-cultural adaptability to further
enhance the robustness and versatility of body-driven emotion
recognition systems.

An ad hoc network in a mobile system is known as a vehicular
network. This application is useful in communication between the
roadside and vehicle infrastructures. Nowadays, the deficiency
and quality of transport classifications of smart transport systems
are playing a vital and excessive role. This paper presents the
examination of the traffic stream under VANET protocol methods.
In the execution of luxury and safe vehicular ad hoc networks, the
automobile must resolve inquiry information from other vehicles
through multi-loop organizations. In case of the data move in
vehicular ad hoc networks, it should have to aspect numerous
disruptions because of a lot of related networks and flexibility.
This paper's subsequent protocol is compared, and a study is
performed on DSR, DSDV, and AODV. In both urban and rural
areas, the preformation exists and is analyzed. The ID
preformation is based on the basis of drop of data, density of
vehicles, the delay of end-to-end loops, and throughput. The
testing and analysis of obtained results from high-throughput DSR
and low packet drop, which provide better results compared with
DSDV and AODV in rural places. Additionally. The AODV
provides better performance compared with the DSR environment
of low densities

A cable driven parallel robot (CDPR) is a specific kind of parallel
robot that utilizes cables as opposed to traditional rigid links to
perform their functions. This gives it the advantages of being
lightweight, having a wide operational workspace, and being
easier and less expensive to implement than alternatives since it
only requires cables operated by motors and fixed to a structure
suitable for a certain application, allowing them to move and
rotate an end-effector. The concept of this robot is not new though
using a fully automated version of it was only started in the mid-
1980s and witnessed extensive research in the past two decades.
This paper gives a review of this robot, its types, applications,
modelling and control which is helpful in shortening the time for
researchers in this area. CDPRs are still a fruitful area for
research like exploiting it in new applications or improving and
designing new control strategies.

Over the past four decades, the communication sector has
experienced significant economic growth and widespread
adoption. Some emerging applications require high data rates, low
latency, and must operate efficiently in environments where
wireless devices are moving at high speeds. Noise and high device
mobility have a considerable impact on received signal frequency,
ultimately affecting system performance. So, accurate frequency
estimation for received signals is therefore crucial, prompting
extensive research. In this paper, we propose a three-step
frequency estimation technique. The first stage, termed the coarse
stage, employs a few sampling sizes to perform a fast Fourier
transform. The second stage utilizes the Candan algorithm to refine
the coarse output, while the third stage applies the Generalized
Goertzel algorithm for further accuracy. The proposed method
offers more precise frequency estimation compared to several well-
known techniques, while significantly reducing latency and
computational power requirements. Our proposed result enhances
the real-time transfer of vital data, such as telemedical information
collected by wearable devices, as well as remote surgeries that
require extremely low latency. It offers a viable solution for
enhancing the performance of the telecommunications sector,
particularly in applications like autonomous vehicles and 5G and
beyond, which demand stringent latency requirements

This review systematically outlines the progression and
implementation of artificial intelligence (AI) techniques, with a
focus on deep learning (DL), for the analysis of
electroencephalogram (EEG) signals. We cover the whole EEG
processing pipeline from signal acquisition and preprocessing
to feature extraction and classification. The survey begins with
traditional machine learning techniques such as Common
Spatial Patterns (CSP) and Support Vector Machines (SVMs),
then transitions to more recent DL architectures—including
Convolutional Neural Networks (CNNs), Recurrent Neural
Networks (RNNs), and their hybrid configurations—that have
significantly expanded the analytical capabilities of
neurophysiological systems. These methods have been widely
applied in fields ranging from clinical diagnostics and brain-
computer interaction to affective computing and cognitive
assessment. Rather than relying on a singular methodology,
research efforts emphasize tailored strategies to overcome
persistent obstacles like signal denoising and the selection of
meaningful features. We discuss applications in a broad array
of areas like clinical diagnosis, brain-computer interfaces,
emotion recognition, and cognitive tests. We compare
performance metrics across methods and note existing
limitations like interpretability issues and computational
complexity. Finally, we mention future directions and trends,
including multimodal integration and explainable AI. This
review provides researchers and practitioners with a
comprehensive overview of state-of-the-art AI techniques for
EEG analysis and indicates some promising avenues for future
advances in this rapidly evolving field.