Construction of Improved Device-to-Device Communication in 5G Networks based on Deep Learning Techniques

Sajad Ali Zearah; Ahmed R. Hassan; Aqeel Ali; Saad Qasim Abbas; Tamarah Alaa Diame; Ahmed Mollah Khan; Mariok Jojoal

doi:https://doi.org/10.54216/FPA.130116

Fusion: Practice and Applications | 2023

Anomaly Detection in IoT Networks: Machine Learning Approaches for Intrusion Detection

The proliferation of Internet of Things (IoT) devices has ushered in an era of unprecedented connectivity and innovation. However, this interconnected landscape also presents unique security challenges, necessitating robust intrusion detection mechanisms. In this research, we present a comprehensive study of anomaly detection in IoT networks, leveraging advanced machine learning techniques. Specifically, we employ the Gated Recurrent Unit (GRU) architecture as the backbone network to capture temporal dependencies within IoT traffic. Furthermore, our approach embraces hierarchical federated training to ensure scalability and privacy preservation across distributed IoT devices. Our experimental design encompasses public IoT datasets, facilitating rigorous evaluation of the model's performance and adaptability. Results indicate that our GRU-based model excels in identifying a spectrum of attacks, from Distributed Denial of Service (DDoS) incursions to SQL injection attempts. Visualizations of learning curves, Receiver Operating Characteristic (ROC) curves, and confusion matrices offer insights into the model's learning process, discriminatory power, and classification performance. Our findings contribute to the evolving landscape of IoT security, offering a roadmap for enhancing the resilience of interconnected systems in an era of increasing connectivity.

groups

Reem Atassi mail

link https://doi.org/10.54216/FPA.130110

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications | 2023

Test Design Optimisation of factors and levels by Covering double and triple mode Combinations using Orthogonal Array test strategies and Random Forest Algorithm

Testing is a process of trying to find out every believable fault or weakness in a project. In today’s world, software products and components play a vital part in our life. Software testing is a world, it contains its own life cycle consists of the following stages – Requirements, Test Plan, Test Design, Test Execution, Defect reporting/tracking. The core of software testing lies in writing test cases based on specifications. Software testers play a vital role writing the test cases during test design phase of software testing life cycle. Research have proved that writing test cases is the most time killing and challenging activity among other testing life cycle phases. It is very crucial to sequence and write optimized test cases to increase the rate of fault identification during test design phase as early as possible. There are various proven test design techniques available which focuses on optimizing test cases in different test stages. Our key focus in this paper is to identify the optimized test cases minimizing the actual number of test cases with minimal effort using OATS (Orthogonal Array Test Strategy) techniques covering double mode and triple mode test combinations and Random Forest algorithm.

groups

S. Malathi mail -

M. Sangeetha mail -

Faiyaz Ahmad mail -

Saravanan M. S. mail -

T. Kalachelvi mail

link https://doi.org/10.54216/FPA.130111

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2023

The Role of Internet of Things in Smart City Environmental Monitoring: A Pollution Detection System

Smart cities represent a transformative vision of urban living, where technology seamlessly integrates with the urban landscape to enhance sustainability and quality of life. Central to this vision is the effective management of environmental factors, particularly air quality and temperature. This paper presents a comprehensive study on real-time environmental pollution detection within smart cities, utilizing Internet of Things (IoT) sensors. We explore the intricate relationships between air pollutant indicators (o3_AQI, no2_AQI, co_AQI, and pm25_AQI) and temperature, shedding light on the dynamic interactions that underlie urban atmospheric conditions. Our research employs a robust dataset and employs statistical analysis, including Ordinary Least Squares (OLS) regression, to uncover significant correlations between key environmental variables. These insights not only contribute to a deeper understanding of urban pollution dynamics but also enable the development of predictive models for temperature fluctuations based on pollutant levels. Such models hold promise for proactive environmental management and public health interventions. Furthermore, our study highlights the pivotal role of IoT sensors in revolutionizing smart city governance, offering real-time data-driven solutions for sustainable urban living. As cities worldwide strive to enhance their environmental resilience, this research provides valuable insights and tools for harnessing the potential of IoT technologies in the pursuit of cleaner and more livable urban environments.

groups

Rasha Almajed mail -

Abedallah Zaid Abualkishik mail -

Laiali Almazaydeh mail -

Sameh Ghwanmeh mail

link https://doi.org/10.54216/JISIoT.090218

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new

International Journal of Neutrosophic Science | 2023

Multi-criteria group decision making method in Pythagorean interval-valued neutrosophic fuzzy soft soft using VIKOR approach

In contrast to the Pythagorean interval valued fuzzy soft set and the neutrosophic interval valued fuzzy soft set, the Pythagorean neutrosophic interval valued fuzzy soft set is a generalization of these sets. We discuss aggregating PyNIVFS decision matrixes by using aggregated operations. The VIKOR method, which is an extension of neutrosophic fuzzy soft sets, is a powerful method for evaluating multi-criteria group decision making. The score function in this approach is based on the aggregation of the VIKOR method to a PyNIVFSpositive and negative solution. Optimal alternatives are introduced under closeness. An investment company plans to purchase some shares of the top five companies in the stock exchange to invest some money on. In order to minimize this factor, they decided to invest some of their cash in percentages of 30 dollars, 25 dollars, 20 dollars, 15 dollars and

groups

M. Palanikumar mail -

Aiyared Iampan mail -

Said Broumi mail -

Lejo J Manavalan mail -

K. Sundareswari mail

link https://doi.org/10.54216/IJNS.220108

Volume & Issue

Vol. Volume 22 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2023

Generalization of neutrosophic interval-valued soft sets with different aggregating operators using multi-criteria group decision-making

In this paper, we present the Pythagorean neutrosophic interval valued fuzzy soft set. This is a generalization of the Pythagorean interval valued fuzzy soft set as well as the neutrosophic interval valued fuzzy soft set. It is discussed in this paper how an aggregated operation is used to aggregate the decision matrix of PNIVS. There are a number of extensions to the normosophic fuzzy soft sets that involve the use of multi-criteria decisionmaking. The aim of this study is to develop a score function based on aggregating TOPSIS methods in order to find ideal solutions for PNIVS that have both positive and negative values. The purpose of this study is to identify the optimal alternative under closeness conditions. It gives us the opportunity to interact with two real life problems, such as the production of ten different types of motorbikes by an automobile company. According to this set of parameters, a motorbike is determined by the fuel tank capacity, better styling, a better price, more mileage, durable, and other factors that determine how a customer can choose which bike to buy.

groups

M. Palanikumar mail -

Aiyared Iampan mail -

Said Broumi mail -

G.Balaji mail

link https://doi.org/10.54216/IJNS.220109

Volume & Issue

Vol. Volume 22 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications |

A Study on Artificial Intelligence-based Security Techniques for IoT-based Systems

In a recent scenario, the Internet of Things (IoT) enables the Integration of disparate home automation systems into a unified network that can be managed from a single device, such as a smartphone. Connections to the Internet that aren't secure: A lack of security standards may make the Internet of Things devices vulnerable to assault, including hacking. Though current designs may address some security concerns inherent to the Internet of Things, most solutions suffer from two significant flaws. This only addresses a single threat at the level of IoT-edge architecture and cannot be expanded to deal with new threats as misunderstood obstacles. Second, its core operations are trustworthy and seldom require additional hardware to implement the advised security measures. The AI-SM-IoT framework is a three-tiered system incorporating security components based on AI motors into every IoT stack that communicates with the network's edge. AI motors were also added as a new transmission layer. This study suggests an AI-based security method for IoT environments (AI-SM-IoT system). This concept was based on the periphery of a network of AI-enabled security components for IoT disaster preparedness. The architecture recommends three main modules: cyber threat searching, intelligent firewalls for online applications, and cybercrime information. Based on the idea of the "cyberspace killing chain," the modules given detect, identify, and continue to identify the stage of an assault life cycle. It describes each long-term security in the suggested framework and demonstrates its usefulness in applications facing various risks. A distinct layer of AI-SM-IoT services is used to deliver artificial intelligence (AI) safety modules to address each risk in the boundaries layer. The architectural freedom from the project's essential regions and comparatively low latency, which offers safety as a service rather than an embedded network edge on the Internet of Things design, contrasted with the system framework's earlier designs. Based on the administration score of the IoT platform, throughput, security, and working time, it evaluated the proposed method

groups

Mustafa Al-Tahee mail -

Marwa S. Mahdi Hussin mail -

Mohammed Jameel Alsalhy mail -

Hussein Alaa Diame mail -

Noor Hanoon Haroon mail -

Salem Saleh Bafjaish mail -

Mohammed Nasser Al-Mhiqani mail

link https://doi.org/10.54216/FPA.130112

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications |

Optimizing Resource Management in Physical Education through Intelligent 5G-Enabled Robotic Systems

Resource Management in Physical Education (RMPE) is the term used to describe the management of the curriculum, materials, and human resources needed for Physical Education (PE). Due to increased sports and physical activity participation, student performance in PE classes across all schools and universities has decreased. According to the analysis, it is hard for the available PE educators and managers to establish a relationship between all the resources. This study uses a robotic system with 5G capability for RMPE. The Big Data Analytics-based Artificial Neural Network method (BDA-ANNA) handles all PE resources in this computerized system. The BDA-ANNA can efficiently increase RMPE work quality and efficiency, enabling managers to obtain and save appropriate information accurately and quickly. With assistance from the robotic system, the material stock may be maintained. With the aid of BDA-ANNA, the mechanical system can keep the material stored. Automated systems with 5G capabilities can provide PE instructors with complete remote-control access with a 2-millisecond latency. These two clauses mandate that the RMPE supervise athletic events and physical activity. The suggested 5 G-enabled robotic systems for RMPE can manage all the resources effectively and efficiently with a low error rate. The advanced system and BDA-ANNA were put through a simulation exercise, demonstrating their independence in classifying and managing resources while reducing processing time. The experimental result improves a prediction ratio of 95.5 %, a learning ratio of 90.5%, an error rate of 92.3%, an Efficiency ratio of 96.6%, an Accuracy ratio of 92.5%, and performance ratio of 96.7%, a Movement Detection ratio of 90.7% compared to other methods.

groups

Maryam Ghassan Majeed mail -

Waleed Hameed mail -

Noor Hanoon Haroon mail -

Sahar R. Abdul Kadeem mail -

Hayder Mahmood Salman mail -

Seifedine Kadry mail

link https://doi.org/10.54216/FPA.130113

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications | 2023

A Framework Based on "One Belt, All Road" Strategy to Evaluate Regional Industry's Cluster Innovation Capacity

Expanding the industrial component through investment in R&D is a crucial objective of the region's current industrial strategy. Significant research and investment opportunities must complement the effectiveness of the region's industrial policy. Few studies have attempted to understand the interactions between inter-organizational clusters and the capacity to sustain those clusters; most studies on innovation capacity focus on the business level. This article suggests using the One Belt All Road (OBAR) strategic framework to assess regional industry's cluster innovation capacity (CIC) and international trade and investment. The cluster innovation capability was developed using a theoretical framework through qualitative textual assessment. As a result, information management, diffusion, and acquisition capacity are the three primary abilities that make up the cluster innovation capacity. The degree of investment effort in the region's industrial sectors and the factors influencing corporate innovation have been found to be correlated. The research highlighted obstacles and potential remedies for encouraging creative thinking and financial backing among regional manufacturers. Compared to the current system, the suggested system (OBAR) achieves superior results in accuracy (87.6%), system dependability (94.8%), the F-1 measure (87.1%), and error rate (8.1%).

groups

Sajad Ali Zearah mail -

Maryam Ghassan Majeed mail -

Mohammed Brayyich mail -

Nabaa R. Wasmi Zaydan mail -

Aqeel Ali mail -

Marwan Qaid Mohammed6 mail -

Venkatesan Rajinikanth mail

link https://doi.org/10.54216/FPA.130114

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications | 2023

Anticipating Student Engagement in Classroom through IoT-Enabled Intelligent Teaching Model Enhanced by Machine Learning

Machine learning provides several advantages for the usage of physical teaching technology. Machine learning is one of the major paths with connected technology and is part of a powerful frontier discipline that develops and influences overall education growth. To enhance student connection and assess student involvement in physical education, the Machine Learning assisted Computerized Physical Teaching Model (MLCPTM) has been developed in this work. The proposed MLCPTM intends to investigate and address contemporary technical physical education to create the ideal theoretical foundation for the growth of technology and current physical activity. Virtual reality (VR) technologies are used in the proposed MLCPTM to create a system for correcting physical education activity. The theory and category of machine learning were covered in this essay, along with a thorough analysis and examination of modern technological advancements in physical education. The challenges with machine learning in contemporary sports instructional technologies are also explained. Then, athletes should accelerate their knowledge of the movement techniques and heighten the training effect. According to the results of the experiments, the suggested MLCPTM model outperforms other existing models in terms of an effective learning ratio of 82.5 per cent, feedback ratio of 96 per cent, response ratio of 98.6 per cent, decision-making ratio of 96.3 per cent, and movement detection ratio of 79.84 per cent, the precision ratio of 97.8 per cent.

groups

Raaid Alubady mail -

Tamarah Alaa Diame mail -

Hawraa Sabah mail -

Hasan H. Jameel Mahdi mail -

Munqith Saleem mail -

Korhan Cengiz mail -

Sahar Yassine mail

link https://doi.org/10.54216/FPA.130115

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Fusion: Practice and Applications | 2023

Construction of Improved Device-to-Device Communication in 5G Networks based on Deep Learning Techniques

Device-to-Device (D2D) Communication promises outstanding data speeds, overall system capacity, and spectrum and energy efficiency without base stations and conventional network infrastructures, and these improvements in network performance sparked a lot of D2D research that exposed substantial challenges before being used to their fullest extent in 5G networks. This study suggests using Deep Learning-based Improved D2D communication (DLID2DC) in 5G networks to address these issues. Reprocessing resources between Cellular User Equipment (CUE) and D2D User Equipment (DUE) can increase system capacity without endangering the CUEs. The D2D resource allocation method allows for a flexible distribution of available resources across CUEs. In addition, several CUEs can consume the same pool of resources simultaneously. Researchers utilize various deep learning techniques to handle the difficulty of constructing D2D links and addressing their interference, mainly when using millimeter-wave (mmWave), to improve the performance of D2D networks. This research aims to increase system capacity by optimizing resource allocation using the suggested DLID2DC paradigm. The model uses Deep Learning methods to overcome interference issues and make D2D link building more efficient, especially in mmWave communication. The model uses Convolutional Neural Networks (CNNs) to learn and adapt to complicated D2D communication patterns, improving performance and dependability. The experimental findings show that, compared to other conventional approaches, the proposed DLID2DC model improves connection with lower end-to-end delay, energy efficiency, throughput, and efficient convergence time.

groups

Sajad Ali Zearah mail -

Ahmed R. Hassan mail -

Aqeel Ali mail -

Saad Qasim Abbas mail -

Tamarah Alaa Diame mail -

Ahmed Mollah Khan mail -

Mariok Jojoal mail

link https://doi.org/10.54216/FPA.130116

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new