Special Issue on Multidisciplinary Sciences and Advanced Technology (MSAT 2024)
Download Complete Issue
ABC
Front Cover
Publication Month: January 2024, Page(s): A1 – A1
Editorial Board
Publication Month: January 2024, Page(s): B1 – B1
Editorial
Publication Month: January 2024, Page(s): C1 – C1
Table of Contents
Publication Month: January 2024, Page(s): D1 – D1
SimulatorBridger: System for Monitoring Energy Efficiency of Electric Vehicles in Real-World Traffic Simulations
Reham Almutairi, Giacomo Bergami, Graham Morgan
J. Engg. Res. & Sci. 3(6), 33-40 (2024);
The increasing popularity and attention in Vehicular Ad-hoc Networks (VANETs) have prompted researchers to develop accurate and realistic simulation tools. Realistic simulation for VANETs is challenging due to the high mobility of vehicles and the need to integrate various communication modalities such as Vehicle-to-Infrastructure (V2I) and Vehicle-to-Vehicle (V2V) interactions. Existing simulators lack the capability to simulate VANET environments based on IoT infrastructure. In this work, we propose SimulatorBridger, a novel simulator that bridges IoTSim-OsmosisRES with SUMO, a traffic simulator, to simulate VANET environments with integrated IoT infrastructure. Our study focuses on analyzing the generated dataflows from V2I and V2V interactions and their impact on vehicle energy efficiency. Even though On-Board Units (OBUs) appear to have insignificant energy demands compared to other vehicle energy consumptions such as electric motors or auxiliary systems (HVAC, lights, comfort facilities), we found a near-perfect correlation between the intensity of communication dataflows and the battery consumption. This correlation indicates that increased communication activity can contribute to an increase in overall energy consumption. Furthermore, we propose future research directions, including traffic rerouting based on battery consumption optimization, which can be efficiently tested using our simulation platform. By including communication energy costs in the design of energy-efficient vehicular networks, these insights contribute to a deeper understanding of energy management in VANETs.
A Thorough Examination of the Importance of Machine Learning and Deep Learning Methodologies in the Realm of Cybersecurity: An Exhaustive Analysis
Ramsha Khalid , Muhammad Naqi Raza
J. Engg. Res. & Sci. 3(7), 11-22 (2024);
In today’s digital age, individuals extensively engage with virtual environments hosting a plethora of public and private services alongside social platforms. As a consequence, safeguarding these environments from potential cyber threats such as data breaches and system disruptions becomes paramount. Cybersecurity encompasses a suite of technical, organizational, and managerial measures aimed at thwarting unauthorized access or misuse of electronic information and communication systems. Its objectives include ensuring operational continuity, safeguarding the confidentiality and integrity of sensitive data, and shielding consumers from various forms of cyber intrusions. This paper delves into the realm of cybersecurity practices devised to fortify computer systems against diverse threats including hacking and data breaches. It examines the pivotal role of artificial intelligence within this domain, offering insights into the utilization of machine learning and deep learning techniques. Moreover, it synthesizes key findings from relevant literature exploring the efficacy and impact of these advanced methodologies in cybersecurity. Findings underscore the substantial contributions of machine learning and deep learning techniques in fortifying computer systems against unauthorized access and mitigating the risks posed by malicious software. These methodologies facilitate proactive measures by predicting and comprehending the behavioral patterns and traffic associated with potential cyber threats.
A Flourished Approach for Recognizing Text in Digital and Natural Frames
Mithun Dutta, Dhonita Tripura, Jugal Krishna Das
J. Engg. Res. & Sci. 3(7), 53-58 (2024);
Acquiring tenable text detection and recognition outcomes for natural scene images as well as for digital frames is very challenging emulating task. This research approaches a method of text identification for the English language which has advanced significantly, there are particular difficulties when applying these methods to languages such as Bengali because of variations in script, morphology. Text identification and recognition is accomplished on multifarious distinct steps. Firstly, a photo is taken with the help of a device and then, Connected Component Analysis (CCA) and Conditional Random Field (CRF) model are introduced for localization of text components. Secondly, a merged model (region-based Convolutional Neural Network (Mask-R-CNN) and Feature Pyramid Network (FPN)) are used to detect and classify text from images into computerized form. Further, we introduce a combined method of Convolutional Recurrent Neural Network (CRNN), Connectionist Temporal Classification (CTC) with K-Nearest Neighbors (KNN) Algorithm for extracting text from images/ frames. As the goal of this research is to detect and recognize the text using a machine learning-based model a new Fast Iterative Nearest Neighbor (Fast INN) algorithm is now proposed based on patterns and shapes of text components. Our research focuses on a bilingual issue (Bengali and English) as well as it producing satisfactory image experimental outcome with better accuracy and it gives around 98% accuracy for our proposed text recognition methods which is better than the previous studies.
Dynamic and Partial Grading of SQL Queries
Benard Wanjiru, Patrick van Bommel, Djoerd Hiemstra
J. Engg. Res. & Sci. 3(8), 1-14 (2024);
Automated grading systems can help save a lot of time when evaluating students’ assignments. In this paper we present our ongoing work for a model for generating correctness levels. We utilize this model to demonstrate how we can grade students SQL queries employing partial grading in order to allocate points to parts of the queries well written and to enable provision of feedback for the missing parts. Furthermore, we show how we can grade the queries taking into account the skill level of students at different stages of SQL learning process. We divide the stages into introductory, intermediary, and advanced stages and in each apply different type of grading that takes account the students’ knowledge at that stage. We implemented this model in our class and graded 5 quizzes containing more than 25 different questions for 309 students. We discuss 3 examples for each stage and offer comprehensive examples of the model in action.
MCNN+: Gemstone Image Classification Algorithm with Deep Multi-feature Fusion CNNs
Haoyuan Huang, Rongcheng Cui
J. Engg. Res. & Sci. 3(8), 15-20 (2024);
Accurate gemstone classification is critical to the gemstone and jewelry industry, and the good performance of convolutional neural networks in image processing has received wide attention in recent years. In order to better extract image content information and improve image classification accuracy, a CNNs gemstone image classification algorithm based on deep multi-feature fusion is proposed. The algorithm effectively deeply integrates a variety of features of the image, namely the main color features extracted by the k-means++ clustering algorithm and the spatial position features extracted by the denoising convolutional neural network. Experimental results show that the proposed method provides competitive results in gemstone image classification, and the classification accuracy is nearly 9% higher than that of CNN. By deeply integrating multiple features of the image, the algorithm can provide more comprehensive and significant useful information for subsequent image processing.
Fingerprint Bio-metric: Confronting Challenges, Embracing Evolution, and Extending Utility – A Review
Diptadip Maiti, Madhuchhanda Basak, Debashis Das
J. Engg. Res. & Sci. 3(9), 26-60 (2024);
As documented in recent research, this reviewoffers a thorough examination of the intricate subject of fingerprint authentication, including a wide range of issues and applications. Addressing problems like non-linear deformations and enhancing picture quality, which are frequently reduced by sophisticated improvement and alignment techniques are important components of fingerprint image authentication. Countering security concerns such as spoofing is a major focus of Automated Fingerprint Identification Systems and necessitates the use of sophisticated cryptographic techniques and liveness detection. In order to accomplish speedier identification processes, the paper emphasizes the advancements made in fingerprint indexing and retrieval, with a focus on deep learning technologies and minutiae-based methodologies. Furthermore, fingerprint authentication is used for a variety of age groups, including neonates, where it is essential for identification verification and the management of medical records. The paper also highlights the wider uses of fingerprint technology, such as improved crime detection skills, insights into age-related features, and contributions to medical diagnostics. This review provides a thorough overview of the latest developments and potential future directions in fingerprint authentication by combining state-of-the-art methodologies and analysing technical details, implementation challenges, and security issues. This captures the dynamic and important role of this biometric technology.
HivePool: An Exploratory Visualization System for Honey Beehive Data
Tinghao Feng, Sophie Columbia, Christopher Campell, Rahman Tashakkori
J. Engg. Res. & Sci. 3(9), 61-74 (2024);
Honey bee health is crucial for global ecosystems, but traditional data analysis methods often struggle to capture the complex interplay between bee behavior and environmental factors. To bridge this gap, we developed HivePool, a novel data visualization and analysis tool designed to empower beekeepers and researchers with deeper insights into these interactions. This paper explores HivePool’s functionalities, focusing on its interactive visualizations and innovative time-oriented pattern recognition for event prediction. By leveraging time series visualization techniques, HivePool allows users to explore not only static relationships between environmental variables but also how these variables change dynamically leading up to specific events within the hive. The paper showcases HivePool’s effectiveness through two use cases: data-driven event exploration and example-driven event prediction. Overall, HivePool equips beekeepers and researchers with a powerful set of tools, facilitating a deeper understanding of bee behavior and environmental influences, ultimately leading to improved beehive health and management strategies.
An Integrated Approach to Manage Imbalanced Datasets using PCA with Neural Networks
Swarup Kumar Mondal and Anindya Sen
J. Engg. Res. & Sci. 3(10), 1-14 (2024);
Imbalanced dataset handling in real time is one of the most challenging tasks in predictive modelling. This work handles the critical issues arising in imbalanced dataset with implementation of artificial neural network and deep neural network architecture. The usual machine learning algorithms fails to achieve desired throughput with certain input circumstances due to mismatched class ratios in the sample dataset. Dealing with imbalanced dataset leads to performance degradation and interpretability issue in traditional ML architectures. For regression tasks, where the target variable is continuous, the skewed data distribution is major issue. In this study, we have investigated a detailed comparison of traditional ML algorithms and neural networks with dimensionality reduction method to overcome this problem. Principle component analysis has been used for feature selection and analysis on real time satellite-based air pollution dataset. Five regression algorithms Multilinear, Ridge, Lasso, Elastic Net and SVM regression is combined with PCA and non PCA to interpret the outcome. To address unbalanced datasets in real-time, deep neural networks and artificial neural network architectures have been developed. Each model’s experiments and mathematical modelling is done independently. The Deep neural network is superior compared to other conventional models for performance measures of target variable in imbalanced datasets
Conceptual Business Model Framework for AI-based Private 5G-IoT Networks
Haoyuan Huang and Rongcheng Cui
J. Engg. Res. & Sci. 3(10), 15-20 (2024);
Accurate gemstone classification is critical to the gemstone and jewelry industry, and the good performance of convolutional neural networks in image processing has received wide attention in recent years. In order to better extract image content information and improve image classification accuracy, a CNNs gemstone image classification algorithm based on deep multi-feature fusion is proposed.
Navigating the Autonomous Era: A Detailed Survey of Driverless Cars
Vaibhavi Tiwari
J. Engg. Res. & Sci. 3(10), 21-36 (2024);
The incorporation of cutting-edge technologies like sensor networks, artificial intelligence (AI), and vehicle-to-everything (V2X) communication has hastened the rollout of autonomous vehicles (AVs), offering significant possibilities for the future of transportation. This document offers an extensive overview of AV technology, covering essential elements such as technological infrastructure, degrees of automation, cybersecurity threats, societal impacts, regulatory structures, and emerging trends. This analysis emphasizes the existing obstacles and progress within the industry by examining the activities of key entities like Tesla, Waymo, and General Motors. Additionally, a comparative examination of autonomous vehicles and drones is performed, providing distinct perspectives on possible cybersecurity vulnerabilities shared by both technologies, including GPS spoofing, jamming, and unauthorized data interception. This multifaceted approach highlights not only the existing vulnerabilities but also proposes proactive measures that can be implemented to reduce comparable risks across various AV
platforms. The results highlight the necessity of establishing strong cybersecurity measures, overcoming regulatory challenges, and building public confidence to realize the complete promise of autonomous vehicles as secure, effective, and eco-friendly transportation options. This analysis provides an essential resource for comprehending the complex aspects of AV technology and its consequences, offering readers a comprehensive perspective on the challenges and opportunities within the autonomous vehicle sector.
Evaluation of equivalent acceleration factors of repairable systems in a fleet: a process-average-based approach
Renyan Jiang, Kunpeng Zhang, Xia Xua and Yu Cao
J. Engg. Res. & Sci. 3(10), 44-54 (2024);
Research on repairable systems in a fleet is mainly concerned with modelling of the failure times using point processes. One important issue is to quantitatively evaluate the heterogeneity among systems, which is usually analyzed using frailty models. Recently, a fleet heterogeneity evaluation method is proposed in the literature. This method describes the heterogeneity with the relative dispersion of equivalent acceleration factors (EAFs) of systems, which is defined as the ratio of the mean times between failures (MTBFs) of a system and a reference system. A main drawback of this method is that the MTBFs of a specific system and the reference system are estimated at different times while the MTBF estimated at different time can be different. This paper aims to address this issue by proposing an improved method. The proposed method uses an “average process” as the reference process and estimates the MTBFs of systems and the reference system at a common time point. This leads to more robust MTBF estimates. Three datasets are analyzed to illustrate the proposed method and its superiority.
Biclustering Results Visualization of Gene Expression Data: A Review
Haithem Aouabed, Mourad Elloumi and Fahad Algarni
J. Engg. Res. & Sci. 3(10), 55-68 (2024);
Biclustering is a non-supervised data mining method used to analyze gene expression data by identifying groups of genes that exhibit similar patterns across specific groups of conditions. Discovering these co-expressed genes (called biclusters) can aid in understanding gene interactions in various biological contexts. Biclustering is characterized by its bi-dimensional nature, grouping both genes and conditions in the same bicluster and its overlapping property, allowing genes to belong to multiple biclusters. Biclustering algorithms often produce a large number of overlapping biclusters. Visualizing these results is not a straightforward task due to the specific characteristics of biclusters. In fact, biclustering results visualization is a crucial process to infer patterns from the expression data. In this paper, we explore the various techniques for visualizing multiple biclusters simultaneously and we evaluate them in order to help biologists to better choose their appropriate visualization techniques
Enhancing Mental Health Support in Engineering Education with Machine Learning and Eye-Tracking
Yuexin Liu, Amir Tofighi Zavareh and Ben Zoghi
J. Engg. Res. & Sci. 3(10), 69-75 (2024);
Mental health concerns are increasingly prevalent among university students, particularly in engineering programs where academic demands are high. This study builds upon previous work aimed at improving mental health support for engineering students through the use of machine learning (ML) and eye-tracking technology. A framework was developed to monitor mental health by analyzing eye movements and physiological data to provide personalized support based on student behavior. In this extended study, baseline data were analyzed to explore the correlations between emotions and physiological biomarkers. Key findings indicate that emotions such as Anger and Fear are positively correlated with increased physical activity, while Sadness is associated with elevated respiratory rates. A strong positive correlation between Electrodermal Activity (EDA) and Happiness was also identified, indicating physiological markers linked to positive emotional states. Temporal patterns were observed, with heightened emotional tagging occurring more frequently in the evening. These findings deepen the understanding of how emotional states manifest through physiological changes, providing a foundation for enhancing real-time, personalized mental health interventions. The results contribute to a more comprehensive framework for supporting student well-being and academic performance within engineering education.
Secure Anonymous Acknowledgments in a Delay-Tolerant Network
Edoardo Biagioni
J. Engg. Res. & Sci. 3(11), 24-30 (2024);
ATCP and many other protocols use acknowledgments to provide reliable transmission of data over unreliable media. Secure acknowledgments offer a cryptographic guarantee that valid acknowledgments for a given message can only be issued by the intended receiver. In the context of an ad-hoc network, anonymous acknowledgments make it hard for an attacker to determine which device issued a particular acknowledgment. And unlike TCP, the acknowledgments described here work well even for connectionless communications. This acknowledgment mechanism assumes that message data is protected by secure encryption. The sender of a data message includes in the encrypted part of the message a randomly-generated acknowledgment. Only the intended receiver can decrypt the message and issue the acknowledgment. The acknowledgment is issued by sending it out to its peers, who will forward it until it reaches the sender of the data being acknowledged. Such randomly-generated acknowledgments in no way identify senders and receivers, providing a degree of anonymity. This paper describes the use of such acknowledgments in both ad-hoc networks and Delay-Tolerant Networks. In such networks every peer participates in forwarding data, including both the routing and the end-host functionalities of more conventional networks. In a Delay-Tolerant Network, peers may cache messages and deliver them to other peers at a later time, supporting end-to-end delivery even when peers are only connected intermittently. Caches have limited size, so peers must selectively remove cached messages when the cache is full. As an additional aid to selecting messages to be removed from a cache, peers can remove messages for which they have received a matching ack. This can be done while preserving both security and anonymity, by including in every message, unencrypted, a message ID computed as the hash of the message ack sent encrypted with the message. A peer seeing a new ack can then hash it and discard any cached message whose message ID matches the hash of the ack.
Fuzzy-Based Approach for Classifying Road Traffic Conditions: A Case Study on the Padua-Venice Motorway
Gizem Erdinc, Chiara Colombaroni and Gaetano Fusco
J. Engg. Res. & Sci. 3(11), 31-40 (2024);
This study offers a fuzzy-based method for determining the variety of traffic conditions on roads. The fuzzy approach appears more appropriate than the deterministic technique for giving drivers qualitative information about the present traffic condition, as drivers have a shaky understanding of the traffic status. It was used in an analysis that included flow, occupancy, and speed measurements from the Italian freeway that runs between Padua and Venice. MATLAB is used in the application’s development. The empirical findings demonstrate how effectively the suggested study performs in classification. The experiment can offer a straightforward and distinctive viewpoint for induction and traffic control on motorways.
A Comparative Analysis of Interior Gateway Protocols in Large-Scale Enterprise Topologies
Saleh Hussein Al-Awami, Emad Awadh Ben Srity and Ali Tahir Abu Raas
J. Engg. Res. & Sci. 3(11), 60-73 (2024);
Interior gateway protocols (IGPs) have gained popularity in networking technologies due to their capacity to enable standardized and flexible communication among these algorithms. In autonomous systems (AS), network devices communicate with one another via IGPs. This work presents a fresh investigation into the performance of inner gateway protocols in large-scale enterprise topologies. Also, the experiment lab using GNS3 simulation has been conducted to evaluate and examine the performance of RIP, EIGRP, OSPF, and IS-IS, taking into account convergence time, latency, and jitter in large-scale network topology. The work has used a tri-connected architecture, with ten (10) routers connected via three serial connections and fifteen (15) network subnets, resulting in thirty (30) different paths for routing data packets for each tested routing technique. End-to-end delay, jitter, and convergence time are three measurement measures used to investigate network topologies. The experiment’s outcomes have revealed that EIGRP has superior delay and convergence time performance. Furthermore, the results have been showed that IS-IS outperformed OSPF in terms of convergence time. Overall, this work improves the field by giving a grand average computation approach for measuring the jitter metric, which has been compared to a standard method. The method has been thoroughly explored utilizing derivate statistical equations and associated pseudo code.
