• Issue 1 (2023)
• Issue 2 (2023)

Enhancement of the TSCH-Sim Simulator via Web Service Interface to Support Co-simulation Optimization

JUSPN, volume-18 , Issue 2 (2023), PP 69 - 76

Published: 02 Apr 2023

DOI: 10.5383/JUSPN.18.02.003

by Tarana Ara, Aida Vatankhah, Ramiro Liscano from Ontario Tech University, Oshawa, Canada, L1G 0C5

Abstract: Co-simulation is an important concept in the optimization of computer networks because a typical optimization scenario integrates an optimization algorithm with a network simulator. In many cases optimization algorithms are implemented in the MATLAB environment while network simulators are implemented as stand alone applications. In this paper we present enhancements to the TSCH-Sim network simulator in order to facilitate its integration with an optimization algorithm. The core enhancement is the definitions and implementation of a set of REST APIs for TSCH-Sim that allows a remote optimization algorithm to set the network configuration, routes, and 802.15.4e TSCH schedule of a sensor network. The significance of the REST API is demonstrated through the integration of a Differential Evolution based TSCH scheduling optimizer executing in MATLAB leveraging the TSCH-Sim simulator through the REST APIs in order to find a TSCH schedule that maximizes throughput. read more... read less...

Keywords: 802.15.4e TSCH, TSCH-Sim, REST-API, Co-simulation, DE Optimization, MATLAB Simulation

Combining the Internet of Things (IoT) and the Internet of Behavior (IoB) to create a robust educational environment

JUSPN, volume-18 , Issue 2 (2023), PP 61 - 68

Published: 02 Apr 2023

DOI: 10.5383/JUSPN.18.02.002

by Ossama H. Embarak from Higher Colleges of Technology Dept. of Computer Sciences, Fujairah, UAE

Abstract: The ongoing COVID-19 pandemic has accelerated the adoption of e-learning, remote learning, and hybrid models in education. These models have become essential in meeting the demands of smart cities and addressing the limitations of traditional distance learning. However, to truly achieve academic success, education must be adaptive and tailored to the individual needs of students. This study presents a novel concept for intelligent educational systems that integrate Explainable Artificial Intelligence (XAI) and Internet of Behavior (IoB) technologies. The integration of these technologies aims to revolutionize intelligent educational systems by providing a more personalized and effective learning experience. By collecting and analyzing student behavior data, the system can provide real-time feedback and adjust to meet the needs of each individual student. The results of this study demonstrate the significant impact of IoB technology on student performance. The integration of IoB led to a substantial increase in student response from 40% to 79%. These findings highlight the potential for IoB to enhance learner assistance and improve system modifications to better meet the expectations of students for increased performance. The proposed concept of integrating XAI and IoB technologies in intelligent educational systems can pave the way for a more personalized and effective learning experience in the future. read more... read less...

Keywords: Tailored education, Hybrid models, Smart cities, XAI (Explainable Artificial Intelligence), IoB (Internet of Behavior) Personalized education Adaptive learning

Predicting Cardiovascular Events with Machine Learning Models and Heart Rate Variability

JUSPN, volume-18 , Issue 2 (2023), PP 49 - 59

Published: 02 Apr 2023

DOI: 10.5383/JUSPN.18.02.001

by Mohammad Moshawrab, Mehdi Adda, Abdenour Bouzouane, Hussein Ibrahim, Ali Raad from Département de Mathématiques, Informatique et Génie, Université du Québec à Rimouski, 300 Allée des Ursulines, Rimouski, G5L 3A1, Québec, Canada, Département D’informatique et de Mathématique, Université du Québec à Chicoutimi, Chicoutimi, 555 boulevard de l’Université, Chicoutimi, QC G7H 2B1, Canada, Institut Technologique de Maintenance Industrielle, 175 Rue de la Vérendrye, Sept-Îles, G4R 5B7, Québec, Canada and Dean of the Faculty of Science and Arts, Islamic University of Lebanon, Wardaniyeh, Lebanon

Abstract: Artificial Intelligence (AI) is increasingly becoming a potential answer to many of science’s most challenging problems. In this context, healthcare is using this technology and its advancement to improve the quality of services provided, including cardiac healthcare services. According to studies, Cardiovascular Diseases (CVDs) are among the most common and deadly diseases in the world. However, Artificial Intelligence and its branches such as Machine Learning (ML) and Deep Learning (DL) offer tremendous potential to improve disease diagnosis and even predict its occurrence. In this study, eight Machine Learning and Deep Learning models are created and trained with "PhsyioNet Smart Health for Assessing the Risk of Events via ECG Database" to analyze the characteristics of Heart Rate Variability and predict the occurrence of heart disease and cerebrovascular events. The results support the use of Artificial Intelligence in cardiology, with five of the proposed models outperforming previous implementations. Specifically, Support Vector Machines, TabTransformers, Deep Neural Networks, AdaBoost, and XGBoost achieved accuracy rates of 91.80%, 90.38%, 90.19%, 89.50%, and 89.10%, respectively. Further performance metrics are presented throught the article such as precision, recall and others. read more... read less...

Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Cardiovascular Diseases, Heart Rate Variability

Hybrid-MiGrror: An Extension to the Hybrid Live Migration to Support Mobility in Edge Computing

JUSPN, volume-18 , Issue 1 (2023), PP 39 - 48

Published: 22 Jan 2023

DOI: 10.5383/JUSPN.18.01.006

by Arshin Rezazadeh, Davood Abednezhad, Hanan Lutfiyya from Computer Science Department, Western University, London ON N6A 3K7 Canada, Information and Communications Technology, Khouzestan Oxin Steel Company, Ahvaz, Iran

Abstract: User-Equipments (UEs) capable of working with cloud computing have grown exponentially in recent years, leading to a significant increase in the amount of data production. Moreover, upcoming Internet-of-Things (IoT) applications such as virtual and augmented reality, video streaming, intelligent transportation, and healthcare will require low latency, communications, and processing. Edge computing is a revolutionary criterion in which dispersed edge nodes supply resources near end devices because of the limited resources available on UEs. Rather than transmitting massive amounts of data to the cloud, edge nodes could filter, analyze, and process the data they receive using local resources. Mobile Edge Computing (MEC), in particular, when user mobility is considered, has the potential to significantly reduce processing delays and network traffic between UEs and servers. This research demonstrated a novel technique for migration that minimizes delay and downtime by utilizing edge computing. Our proposed method syncs more frequently than the pre-copy method which is the most used migration method that synchronizes (sync) the source and destination only based on multiple rounds. When compared to established migration methodologies, our results indicate that our mechanism has less latency, downtime, migration time, and packet loss. These results allow delay-sensitive applications that require ultra-low latency to function smoothly during migration. read more... read less...

Keywords: delay (latency), mobile edge computing (MEC), downtime, hand-off (handover), live migration, fog computing

5G-enhanced Positioning Accuracy in Smart City

JUSPN, volume-18 , Issue 1 (2023), PP 31 - 38

Published: 17 Jan 2023

DOI: 10.5383/JUSPN.18.01.005

by Salam Traboulsi, Dieter Uckelmann from Stuttgart University of Applied Sciences, D-70174 Stuttgart, Germany

Abstract: The new 5G mobile network promises to enhance existing services or include new ones to address key challenges presented by smart city stakeholders (citizens, municipalities, politics, industries, architects, etc.) to improve system implementations. These challenges cover various smart city fields such as transportation, environmental monitoring, healthcare, industrial automation, smart grid, etc. Thus, the main objective of 5G functionalities is to provide solutions to the various identified needs, which are defined as constraints and requirements. Therefore, three categories of 5G-based use cases have been defined: Enhanced Mobile Broadband (eMBB), Massive Machine Type Communications (mMTC), and Ultra-reliable and Low Latency Communications (uRLLC). Each group involves a set of use cases and characterized by specific technical features that address the corresponding needs. However, accurate and real-time positioning information is a vital requirement common to all three categories, but the degree of performance varies across scenarios and descriptions. Therefore, this work presents a summary of existing positioning technologies crossed with wireless technologies and smart city use cases to highlight the potential that will add accurate and real-time positioning to 5G capabilities. 5G promises decimeter accuracy in some critical use cases. read more... read less...

Keywords: 5G, Positioning methods, Smart city

A Study of the Ambient Noise in the Public Space on Campus and the Correlation Between the Campus Crowds’ Ambient Noise and the WiFi Log

JUSPN, volume-18 , Issue 1 (2023), PP 23 - 30

Published: 16 Jan 2023

DOI: 10.5383/JUSPN.18.01.004

by Yun Jie Lim, Seanglidet Yean, Bu Sung Lee, Peter Edwards from School of Computer Science and Engineering, Nanyang Technological University, 50 Nanyang Ave, 639798, Singapore and Manaaki Whenua – Landcare Research, Box 10345, The Terrace, 6143, Wellington, New Zealand

Abstract: Urban noise is becoming more serious and increasingly concerning environmental problems. This has led to numerous studies on traffic noise. However, not many studies have been done on noise from a human perspective as they go about their daily life. In another aspect, using of the crowd-sourcing platform is on the rise as the usage of personal devices (smartphones) and the deployment of Internet-of-thing increases. Thus, a large pool of data collected via mobile applications enables users to measure the environmental factor directly and provide immediate feedback for and community’s greater good. In this paper, we utilize the crowd-sourcing platform to collect noise data by volunteers to study the noise level in a campus environment, in open common areas which are frequented by students. We are able to map out the noise across the campus from the perspective of the students. The noise level increase through the day as the student gather around popular open spaces. Our study shows that the sound level on campus is due mainly to human and mechanical noise. By combining the noise data with WiFi log data, we were able to show a good correlation between sound level and human density in an area. read more... read less...

Keywords: sound level, ambient noise, environmental noise monitoring, environmental noise analysis, urban noise

Segment Routing v6 - Security Issues and Experimental Results

JUSPN, volume-18 , Issue 1 (2023), PP 15 - 21

Published: 16 Jan 2023

DOI: 10.5383/JUSPN.18.01.003

by David Lo Bascio, Flavio Lombardi from Dept. of Information Engineering, Electronics and Telecommunications (DIET) “Sapienza” University of Rome, Italy, Istituto per le Applicazioni del Calcolo, Consiglio Nazionale delle Ricerche (IAC-CNR), Rome, Italy, Member of the INdAM-GNCS research group

Abstract: SRv6 can provide hybrid cooperation between a centralized network controller and network nodes. IPv6 routers maintain multi-hop ECMP-aware segments, whereas the controller establishes a source-routed path through the network. Since the state of the flow is defined at the ingress to the network and then is contained in a specific packet header, called Segment Routing Header (SRH), the importance of such a header itself is vital. Motivated by the need to study and investigate this technology, this paper discusses some security-related issues of Segment Routing. A SRv6 capable experimental testbed is built and detailed. Finally, an experimental test campaign is performed and results are evaluated and discussed. read more... read less...

Keywords: Segment Routing, Networking, Security

Control of autonomous UAV using an onboard LSTM neural network

JUSPN, volume-18 , Issue 1 (2023), PP 09 - 14

Published: 16 Jan 2023

DOI: 10.5383/JUSPN.18.01.002

by Bill Karakostas from Independent Researcher, Building 94700, PO Box 4336, Manchester, UK M61 0BW

Abstract: Towards realising autonomous UAVs, this paper investigates one of the fundamental autonomous flying research problems, i.e., the ability of a vehicle to control its flying behaviour autonomously, without reliance on external infrastructure like Instrument Landing Systems or GPS. In this paper we experiment with a physical UAV prototype with embedded intelligent control capabilities, utilising a Long Short Term Memory (LSTM) neural network, in order to learn lift-off control sequences using self-training. The initial results are promising and show potential for embedding LSTMs in the control systems of autonomous UAVs. read more... read less...

Keywords: Autonomous UAV, Intelligent Control, LSTM neural networks

Empowering Reality: A New Injury Prevention Education System to Promote the Empowerment of Child Caregivers

JUSPN, volume-18 , Issue 1 (2023), PP 01 - 08

Published: 16 Jan 2023

DOI: 10.5383/JUSPN.18.01.001

by Mikiko Oono, Thassu Srinivasan Shreesh Babu, Yoshifumi Nishida, Tatsuhiro Yamanaka from National Institute of Advanced Industrial Science and Technology, Koto-ku, Tokyo 135-0064, Japan , Safe Kids Japan, Setagaya-ku, Tokyo 157-0074, Japan , Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8550, Japan , Ryokuen Children’s Clinic, Yokohama, Kanagawa 245-0002, Japan

Abstract: Although awareness about the importance of injury prevention has been increasing among Japanese people, preventable injuries remain the third leading cause of death in children aged 0–14 years, and prevention of these injuries is critically important in terms of childhood health. To identify dangerous situations for children and provide preventive measures to avoid such situations, this paper proposes an effective method, called “Empowering Reality (ER)”, that integrates knowledge graphs with object detection to enable lecturers to educate caregivers on preventing unintentional childhood injuries while communicating with caregivers using augmented reality technology. The proposed ER system consists of knowledge graphs for explaining dangerous situations, an online video capture part, and a situation recognition part. This paper describes the major advantages of knowledge graphs that consider not only the relationship between objects and injuries, but also dangerous layouts with the help of “inclusion” and “collocation” features. The feasibility and effectiveness of the system were evaluated through tests among caregivers, including 11 parents and six teachers from three nursery schools. This system allows lecturers to conduct in-situ suggestions about specific preventive measures adapted to the home or nursery school environment via online learning read more... read less...

Keywords: Empowerment, Injury Prevention, Online Education, Augmented Reality, Knowledge Graphs, Situation Recognition