2024. 3r^d Issue

Volume XVI, Number 3

Table of contents 

Full issue  

 

MESSAGE FROM THE EDITOR-IN-CHIEF

Pal Varga
AI and ML techniques in various fields of Infocommunications – in the autumn issue of ICJ  
WELCOME to the September 2024 issue of the Infocom munications Journal. Let’s have an brief overview of the papers. The paper by A. S. Jagmagji and his co-authors presents the application of machine learning models to predict network per formance metrics for the Self-Clocked Rate Adaptation for Mul timedia (SCReAM) congestion control algorithm. By employing a systematic approach, including regression models, hyperpa rameter tuning, and ensemble learning, the study achieved high accuracy in predicting key metrics such as network throughput, queue delay, and smoothed Round Trip Time (sRTT). The Light GBM and CatBoost models outperformed others in predicting these metrics, demonstrating the effectiveness of the applied techniques. The study also highlights areas for improvement, including more advanced hyperparameter tuning and ensemble methods, and calls for rigorous statistical testing to validate minor performance differences.

Download

 

PAPERS FROM OPEN CALL

Ahmed Samir Jagmagji, Haider Dhia Zubaydi, Sándor Molnár, and Mahmood Alzubaidi 
Utilizing Machine Learning as a Prediction Scheme for Network Performance Metrics of Self-Clocked Congestion Control Algorithm 
Congestion Control (CC) is a fundamental mecha- nism to achieve effective and equitable sharing of network fa cilities. As future networks evolve towards more complex para digms, traditional CC methods are required to become more powerful and reliable. On the other hand, Machine Learning (ML) has become increasingly popular for solving challeng ing and sophisticated problems, and scientists have started to turn their interest from rule-based approaches to ML-based methods. This paper employs machine learning models to con struct a performance evaluation scheme to predict network metrics for the Self-Clocked Rate Adaptation for Multimedia (SCReAM) algorithm. It uses a rigorous data preprocessing pipeline and a systematic application of ML methods to en hance the performance of the regression model for SCReAM’s performance metrics. Also, we constructed a dataset that pro vides SCReAM’s input parameters and output metrics, such as network queue delay, smoothed Round Trip Time (sRTT), and network throughput. Each prediction process has several phases: choosing the best initial regressor model, hyperparam eter tuning, ensemble learning, stacking regressors, and uti lizing the holdout data. Each model’s performance was evalu ated through various regression metrics; this study will mainly focus on the coefficient of determination (R2) score. The im provement between the initial best-selected model and the fi nal improved model determined that we were able to increase R2 up to 96.64% for network throughput, 99.4% for network queue delay, and 100% for sRTT.

Reference
DOI: 10.36244/ICJ.2024.3.1
Download 

 

Árpád Huszák, Vilmos Simon, László Bokor, László Tizedes, and Adrian Pekar
An AI-Driven Intelligent Transportation System: Functional Architecture and Implementation 
The surge in urbanization and the concomitant growth of the urban population have exacerbated issues such as traffic congestion and air pollution across cities globally. While Intelligent Transportation Systems (ITS) offer promise for im- proving urban mobility, existing solutions predominantly exhib it limitations in scalability and adaptability, thus falling short in delivering city-wide traffic management. This unaddressed gap necessitates the development of a robust, scalable, and adaptive system that can manage the intricacies of urban traffic. Our work introduces CityAI, an automated, AI-driven framework designed to operate on a city-wide scale. The system harvests data from diverse sensing infrastructures, employing machine learning algorithms to predict future traffic states and pat terns. Furthermore, it proposes real-time interventions, includ ing adaptive traffic light control and V2X-based solutions. The architecture and components of CityAI not only incorporate state-of-the-art techniques but are also applied in real-world en vironments. The CityAI framework was implemented in the city of Pécs, Hungary, as a proof-of-concept ITS system. The frame work enables city authorities to implement proactive measures, thus preventing traffic issues before they manifest. The paper fo cuses on practical development aspects of an ITS system under taking R&D on new technologies, applications, and techniques which may facilitate future product development.

Reference
DOI: 10.36244/ICJ.2024.3.2
Download 

 

 

Raghad Al-Shabandar, Ail Jaddoad, Taha A. Elwi, A. H. Mohammed, and Abir Jaafar Hussain 
A Systematic Review for the Implication of Generative AI in Higher Education 
The rapid advancement of genitive AI, like Chat GPT, has initiated a profound transformation in higher edu cation. It offers customized learning experiences, automates administrative tasks, and provides personalized support to stu dents and educators. Following PRISMA guidelines, this paper presents a systematic review that delves into the implications of genitive AI, a cutting-edge language model, in higher education. We adopted ChatGPT as an example of this study. It thoroughly examines the potential advantages and constraints of integrat ing ChatGPT into educational environments, assessing the quality of 35 selected articles and conducting a comprehensive meta-analysis of their findings. This study yields fresh insights into the multifaceted consequences of employing ChatGPT in higher education and underscores the intricate landscape asso ciated with AI integration in academic settings. It emphasizes the imperativeness of addressing ethical, legal, and pragmatic challenges while capitalizing on the potential benefits of AI tech nology in education. Our systematic review reveals a consistent reservation trend regarding generative AI integration within educational contexts. These concerns encompass many issues, emphasizing the necessity for judicious implementation and ro bust safeguards to mitigate potential challenges.

Reference
DOI: 10.36244/ICJ.2024.3.3
Download 

 

 

Mohd Aaqib Lone, Szilveszter Kovács, and Owais Mujtaba Khanday 
Implementation Guidelines for Ethologically Inspired Fuzzy Behaviour-Based Systems 
The adaptation of ethologically inspired behaviour models for human-machine interaction e.g. in Ethorobotics has become a challenging research topic in recent years. This paper presents a Fuzzy Behaviour Description Language (FBDL) approach for analyzing animal aggression behaviour. Fuzzy logic and fuzzy set theory approaches are used to analyze and classify the subjective impression of aggressive behaviour in a particular situation. This research aims to perform a meta analysis of aggression behaviour based on the fundamental values of animals and the possible ways of implementing animal aggressive behaviour in robots. Ultimately aiming to enhance the adaptability and effectiveness of human-robot interaction and performance in various real-world scenarios, e.g., by expressing disagreement in the direction of the human operator in case of unclear, or unsafe cooperative situations. In both industrial and everyday settings, mobile robots and robotic vehicles are becom- ing increasingly prevalent. Integrating aggressive behaviour into robotics is essential for boosting interactions between humans and robots, promoting safety in dynamic contexts, and getting a deeper understanding of animal behaviour. It aids robots in asserting their presence, maneuvering around barriers, and efficiently adjusting to dynamic surroundings. This guarantees more seamless operations in industrial and daily environments while also enhancing our comprehension of both robotics and ethology. We present graphical depictions of various animal behaviours, as well as trajectories, Gazebo simulations, and RViz visualizations of the animal robot, demonstrating the animal’s escape behaviour.

Reference
DOI: 10.36244/ICJ.2024.3.4
Download 

 

 

Nasraldeen Alnor Adam Khleel, and Károly Nehéz 
Optimizing LSTM for Code Smell Detection: The Role of Data Balancing 
Code smells are specific patterns or characteristics in software code that indicate potential design or implementation problems. Identifying code smells has gained significant attention in software engineering. It is essential to address code smells to maintain high-quality software systems. Machine learning (ML) models, such as Long Short-Term Memory (LSTM), have been to detect code smells automatically based on source code features. However, the imbalanced distribution of code smells within software projects poses a challenge to the accuracy of these models. This study explores the role of data balancing methods in optimizing the accuracy of the LSTM model for code smell detection. We investigate different techniques for addressing the class imbalance problem, including random oversampling and synthetic minority oversampling techniques (SMOTE). We evaluate the performance of the LSTM model with and without data balancing methods using accuracy, precision, recall, f-measure, Matthew’s correlation coefficient (MCC), and the area under a receiver operating characteristic curve (AUC). Our experimental results, conducted on four code smell datasets (God class, data class, feature envy, and long method) extracted from 74 open-source systems, demonstrate the effectiveness of data balancing methods in improving the accuracy of the LSTM model for code smell detection. The results indicate that the use of data balancing methods had a positive effect on the predictive accuracy of the LSTM model. In addition, we compared our proposed method with state-of-the-art code smell detection approaches. The findings from the comparison indicate that our proposed method performs notably better than existing state-of the-art approaches across the majority of datasets.

Reference
DOI: 10.36244/ICJ.2024.3.5
Download 

 

 

Adnane El Hanjri, Ikram Ben Abdel Ouahab and Abdelkrim Haqiq 
Survey on Handover Techniques for Heterogeneous Mobile Networks 
This paper presents an overview of existing tech- niques of Handover in Heterogeneous Mobile Networks. It gives an overview of the mobility management processes and mainly focuses on decision-making approaches. The literature has reported many problems with seamless support for mobil ity management techniques. Failures in the Handover operation are caused by frequent disconnections and ineffective seamless Handovers. Therefore, to provide customers with an accept able Quality of Service, Heterogeneous Mobile Networks must have an effective mobility management system that allows many wireless networks to collaborate. A single parameter, two or more extra factors, or a mix of both are used by several mobile controlled Handovers to assess the policy choice. In this paper, We have covered many Handover approaches, as well as ad vancements that have been achieved throughout time. Almost all of the Handover Techniques over the previous ten years have been covered. Based on the many Advantages and Limitations, we have tabulated all the Handover procedures. The paper will be beneficial to emerging specialists in the sector.

Reference
DOI: 10.36244/ICJ.2024.3.6
Download 

 

 

Lajos Nagy 
Dielectric Lens Antenna for Industrial Radar Applications
Industrial radar applications like tank level meas urement is an important research and application area in ra dar technology. Radar level measurement is a safe solution even under extreme process conditions (pressure, temperature) and vapors. Special antennas are required to meet electromagnet ic requirements such as high gain, low sidelobe level and high bandwidth. The small side-beam level and narrow main beam primarily minimize reflections from the side of the tank, while the bandwidth determines the distance resolution of the meas urement system. Another requirement is a small size and good manufacturability of the antenna. The main promising solutions are the use of microstrip, horn or dielectric lens antennas for tank level measurement systems. After several tests, we have concluded that the optimal choice for tank level radar measurement task, in terms of integrability and antenna parameters, is a dielectric antenna. The dielectric antenna has many other applications in modern mobile systems as 5 and 6 G systems where these antennas are elements of an tenna arrays of beamforming or MIMO systems. In this paper, a special dielectric lens antenna is presented, satisfying main requirements, namely a circular antenna cross section, high antenna aperture efficiency and low sidelobe level. The center frequency of the antenna is 26 GHz with a band width of 1 GHz. The paper presents the analytic investigation and design of the dielectric lens antenna and the circular wave guide transition in detail. The electromagnetic design of the an tenna was carried out using CST Microwave Studio 3D software.

Reference
DOI: 10.36244/ICJ.2024.3.7
Download 

 

 

Tamara Zuhair Fadhil Noor Asniza Murad, and Mohamad Rijal Hamid 
Broadside Gain Enhancement of Wideband Monopole Circular Shaped Antenna Using FSS for Sub-6 GHz Applications 
This paper introduces a wideband circular patch antenna designed with a frequency selective surface (FSS) for sub-6 GHz applications. The proposed antenna features a monopole circular-shaped patch with a partial ground plane, delivering an omnidirectional radiation pattern in the azimuth plane, resulting in relatively uniform gain in all directions. An FSS metamaterial enhances the antenna's gain and improves the broadside radiation pattern. The design incorporates three inner circular patches connected to the main patch. The FSS utilizes hybrid square/circle loop-based unit cells. The antenna and FSS are simulated using CST software and subsequently fabricated on an FR-4 substrate. The measured results demon strate an impedance bandwidth of 1.6 GHz with a peak gain of 5.4 dB at 3.5 GHz. The omnidirectional radiation pattern is converted into a directional one by placing a reflector FSS as a bottom substrate layer. The overall structure size is compact, measuring (0.34λ0 × 0.27λ0 × 0.016λ0), where λ0 is the free space wavelength corresponding to the lowest resonant fre quency within the operational bandwidth. This design achieves significant antenna size reduction and is well-suited for future sub-6 GHz applications.

Reference
DOI: 10.36244/ICJ.2024.3.8
Download 

 

 

Tariq Emad Ali, Faten Imad Ali, Mohammed A. Abdala, Ameer Hussein Morad, Győző Gódor, and Alwahab Dhulfiqar Zoltá́n 
Blockchain-Based Deep Reinforcement Learning System for Optimizing Healthcare 
The Industrial Internet of Things (IIoT) has become a transformative force in various healthcare applications, provid ing integrated services for daily life. The app healthcare based on the IIoT framework is broadly used to remotely monitor clients health using advanced biomedical sensors with wireless technolo gies, managing activities such as monitoring blood pressure, heart rate, and vital signs. Despite its widespread use, IIoT in health care faces challenges such as security concerns, inefficient work scheduling, and associated costs. To address these issues, this paper proposes and evaluates the Blockchain-Based Deep Rein forcement Learning System for Optimizing Healthcare (BDRL) framework. BDRL aims to enhance security protocols and maxi mize makespan efficiency in scheduling medical applications. It facilitates the sharing of legitimate and secure data among linked network nodes beyond the initial stages of data validation and as signment. This study presents the design, implementation, and statistical evaluation of BDRL using a new dataset and varying platform resources. The evaluation shows that BDRL is versatile and successfully addresses the security, privacy, and makespan needs of healthcare applications on distributed networks, while also delivering excellent performance. However, the framework utilizes high resources as the size of inserted data increases.

Reference
DOI: 10.36244/ICJ.2024.3.9
Download 

 

Call for papers