Research

Jump to: Review

18 pages, 1676 KiB

Open AccessArticle

A Machine Learning Approach as a Surrogate for a Finite Element Analysis: Status of Research and Application to One Dimensional Systems

by Poojitha Vurtur Badarinath, Maria Chierichetti and Fatemeh Davoudi Kakhki

Sensors 2021, 21(5), 1654; https://doi.org/10.3390/s21051654 - 27 Feb 2021

Cited by 37 | Viewed by 8137

Abstract

Current maintenance intervals of mechanical systems are scheduled a priori based on the life of the system, resulting in expensive maintenance scheduling, and often undermining the safety of passengers. Going forward, the actual usage of a vehicle will be used to predict stresses [...] Read more.

Current maintenance intervals of mechanical systems are scheduled a priori based on the life of the system, resulting in expensive maintenance scheduling, and often undermining the safety of passengers. Going forward, the actual usage of a vehicle will be used to predict stresses in its structure, and therefore, to define a specific maintenance scheduling. Machine learning (ML) algorithms can be used to map a reduced set of data coming from real-time measurements of a structure into a detailed/high-fidelity finite element analysis (FEA) model of the same system. As a result, the FEA-based ML approach will directly estimate the stress distribution over the entire system during operations, thus improving the ability to define ad-hoc, safe, and efficient maintenance procedures. The paper initially presents a review of the current state-of-the-art of ML methods applied to finite elements. A surrogate finite element approach based on ML algorithms is also proposed to estimate the time-varying response of a one-dimensional beam. Several ML regression models, such as decision trees and artificial neural networks, have been developed, and their performance is compared for direct estimation of the stress distribution over a beam structure. The surrogate finite element models based on ML algorithms are able to estimate the response of the beam accurately, with artificial neural networks providing more accurate results. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

18 pages, 1139 KiB

Open AccessArticle

A Smart Biometric Identity Management Framework for Personalised IoT and Cloud Computing-Based Healthcare Services

by Farnaz Farid, Mahmoud Elkhodr, Fariza Sabrina, Farhad Ahamed and Ergun Gide

Sensors 2021, 21(2), 552; https://doi.org/10.3390/s21020552 - 14 Jan 2021

Cited by 39 | Viewed by 4497

Abstract

This paper proposes a novel identity management framework for Internet of Things (IoT) and cloud computing-based personalized healthcare systems. The proposed framework uses multimodal encrypted biometric traits to perform authentication. It employs a combination of centralized and federated identity access techniques along with [...] Read more.

This paper proposes a novel identity management framework for Internet of Things (IoT) and cloud computing-based personalized healthcare systems. The proposed framework uses multimodal encrypted biometric traits to perform authentication. It employs a combination of centralized and federated identity access techniques along with biometric based continuous authentication. The framework uses a fusion of electrocardiogram (ECG) and photoplethysmogram (PPG) signals when performing authentication. In addition to relying on the unique identification characteristics of the users’ biometric traits, the security of the framework is empowered by the use of Homomorphic Encryption (HE). The use of HE allows patients’ data to stay encrypted when being processed or analyzed in the cloud. Thus, providing not only a fast and reliable authentication mechanism, but also closing the door to many traditional security attacks. The framework’s performance was evaluated and validated using a machine learning (ML) model that tested the framework using a dataset of 25 users in seating positions. Compared to using just ECG or PPG signals, the results of using the proposed fused-based biometric framework showed that it was successful in identifying and authenticating all 25 users with 100% accuracy. Hence, offering some significant improvements to the overall security and privacy of personalized healthcare systems. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

20 pages, 5126 KiB

Open AccessArticle

Intelligent Tensioning Method for Prestressed Cables Based on Digital Twins and Artificial Intelligence

by Zhansheng Liu, Guoliang Shi, Anshan Zhang and Chun Huang

Sensors 2020, 20(24), 7006; https://doi.org/10.3390/s20247006 - 08 Dec 2020

Cited by 33 | Viewed by 3499

Abstract

In this study, to address the problems of multiple dimensions, large scales, complex tension resource scheduling, and strict quality control requirements in the tensioning process of cables in prestressed steel structures, the technical characteristics of digital twins (DTs) and artificial intelligence (AI) are [...] Read more.

In this study, to address the problems of multiple dimensions, large scales, complex tension resource scheduling, and strict quality control requirements in the tensioning process of cables in prestressed steel structures, the technical characteristics of digital twins (DTs) and artificial intelligence (AI) are analyzed. An intelligent tensioning of prestressed cables method driven by the integration of DTs and AI is proposed. Based on the current research status of cable tensioning and DTs, combined with the goal of intelligent tensioning, a fusion mechanism for DTs and AI is established and their integration to drive intelligent tensioning of prestressed cables technology is analyzed. In addition, the key issues involved in the construction of an intelligent control center driven by the integration of DTs and AI are discussed. By considering the construction elements of space and time dimensions, the tensioning process is controlled at multiple levels, thereby realizing the intelligent tensioning of prestressed cables. Driven by intelligent tensioning methods, the safety performance evaluation of the intelligent tensioning process is analyzed. Combined with sensing equipment and intelligent algorithms, a high-fidelity twin model and three-dimensional integrated data model are constructed to realize closed-loop control of the intelligent tensioning safety evaluation. Through the study of digital twins and artificial intelligence fusion to drive the intelligent tensioning method for prestressed cables, this study focuses on the analysis of the intelligent evaluation of safety performance. This study provides a reference for fusion applications with DTs and AI in intelligent tensioning of prestressed cables. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

20 pages, 9332 KiB

Open AccessEditor’s ChoiceArticle

A Framework for an Indoor Safety Management System Based on Digital Twin

by Zhansheng Liu, Anshan Zhang and Wensi Wang

Sensors 2020, 20(20), 5771; https://doi.org/10.3390/s20205771 - 12 Oct 2020

Cited by 62 | Viewed by 6659

Abstract

With the development of the next generation of information technology, an increasing amount of attention is being paid to smart residential spaces, including smart cities, smart buildings, and smart homes. Building indoor safety intelligence is an important research topic. However, current indoor safety [...] Read more.

With the development of the next generation of information technology, an increasing amount of attention is being paid to smart residential spaces, including smart cities, smart buildings, and smart homes. Building indoor safety intelligence is an important research topic. However, current indoor safety management methods cannot comprehensively analyse safety data, owing to a poor combination of safety management and building information. Additionally, the judgement of danger depends significantly on the experience of the safety management staff. In this study, digital twins (DTs) are introduced to building indoor safety management. A framework for an indoor safety management system based on DT is proposed which exploits the Internet of Things (IoT), building information modelling (BIM), the Internet, and support vector machines (SVMs) to improve the level of intelligence for building indoor safety management. A DT model (DTM) is developed using BIM integrated with operation information collected by IoT sensors. The trained SVM model is used to automatically obtain the types and levels of danger by processing the data in the DTM. The Internet is a medium for interactions between people and systems. A building in the bobsleigh and sled stadium for the Beijing Winter Olympics is considered as an example; the proposed system realises the functions of the scene display of the operation status, danger warning and positioning, danger classification and level assessment, and danger handling suggestions. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

25 pages, 469 KiB

Open AccessArticle

A Generalized Threat Model for Visual Sensor Networks

by Jennifer Simonjan, Sebastian Taurer and Bernhard Dieber

Sensors 2020, 20(13), 3629; https://doi.org/10.3390/s20133629 - 28 Jun 2020

Cited by 8 | Viewed by 4740

Abstract

Today, visual sensor networks (VSNs) are pervasively used in smart environments such as intelligent homes, industrial automation or surveillance. A major concern in the use of sensor networks in general is their reliability in the presence of security threats and cyberattacks. Compared to [...] Read more.

Today, visual sensor networks (VSNs) are pervasively used in smart environments such as intelligent homes, industrial automation or surveillance. A major concern in the use of sensor networks in general is their reliability in the presence of security threats and cyberattacks. Compared to traditional networks, sensor networks typically face numerous additional vulnerabilities due to the dynamic and distributed network topology, the resource constrained nodes, the potentially large network scale and the lack of global network knowledge. These vulnerabilities allow attackers to launch more severe and complicated attacks. Since the state-of-the-art is lacking studies on vulnerabilities in VSNs, a thorough investigation of attacks that can be launched against VSNs is required. This paper presents a general threat model for the attack surfaces of visual sensor network applications and their components. The outlined threats are classified by the STRIDE taxonomy and their weaknesses are classified using CWE, a common taxonomy for security weaknesses. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

15 pages, 499 KiB

Open AccessArticle

Covert Timing Channel Analysis Either as Cyber Attacks or Confidential Applications

by Shorouq Al-Eidi, Omar Darwish and Yuanzhu Chen

Sensors 2020, 20(8), 2417; https://doi.org/10.3390/s20082417 - 24 Apr 2020

Cited by 11 | Viewed by 4410

Abstract

Covert timing channels are an important alternative for transmitting information in the world of the Internet of Things (IoT). In covert timing channels data are encoded in inter-arrival times between consecutive packets based on modifying the transmission time of legitimate traffic. Typically, the [...] Read more.

Covert timing channels are an important alternative for transmitting information in the world of the Internet of Things (IoT). In covert timing channels data are encoded in inter-arrival times between consecutive packets based on modifying the transmission time of legitimate traffic. Typically, the modification of time takes place by delaying the transmitted packets on the sender side. A key aspect in covert timing channels is to find the threshold of packet delay that can accurately distinguish covert traffic from legitimate traffic. Based on that we can assess the level of dangerous of security threats or the quality of transferred sensitive information secretly. In this paper, we study the inter-arrival time behavior of covert timing channels in two different network configurations based on statistical metrics, in addition we investigate the packet delaying threshold value. Our experiments show that the threshold is approximately equal to or greater than double the mean of legitimate inter-arrival times. In this case covert timing channels become detectable as strong anomalies. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

21 pages, 840 KiB

Open AccessArticle

Determining the Optimal Restricted Driving Zone Using Genetic Algorithm in a Smart City

by Tony Jan, Pegah Azami, Saeid Iranmanesh, Omid Ameri Sianaki and Shiva Hajiebrahimi

Sensors 2020, 20(8), 2276; https://doi.org/10.3390/s20082276 - 16 Apr 2020

Cited by 12 | Viewed by 2847

Abstract

Traffic control is one of the most challenging issues in metropolitan cities with growing populations and increased travel demands. Poor traffic control can result in traffic congestion and air pollution that can lead to health issues such as respiratory problems, asthma, allergies, anxiety, [...] Read more.

Traffic control is one of the most challenging issues in metropolitan cities with growing populations and increased travel demands. Poor traffic control can result in traffic congestion and air pollution that can lead to health issues such as respiratory problems, asthma, allergies, anxiety, and stress. The traffic congestion can also result in travel delays and potential obstruction of emergency services. One of the most well-known traffic control methods is to restrict and control the access of private vehicles in predetermined regions of the city. The aim is to control the traffic load in order to maximize the citizen satisfaction given limited resources. The selection of restricted traffic regions remains a challenge because a large restricted area can reduce traffic load but with reduced citizen satisfaction as their mobility will be limited. On the other hand, a small restricted area may improve citizen satisfaction but with a reduced impact on traffic congestion or air pollution. The optimization of the restricted zone is a dynamic multi-regression problem that may require an intelligent trade-off. This paper proposes Optimal Restricted Driving Zone (ORDZ) using the Genetic Algorithm to select appropriate restricted traffic zones that can optimally control the traffic congestion and air pollution that will result in improved citizen satisfaction. ORDZ uses an augmented genetic algorithm and determinant theory to randomly generate different foursquare zones. This fitness function considers a trade-off between traffic load and citizen satisfaction. Our simulation studies show that ORDZ outperforms the current well-known methods in terms of a combined metric that considers the least traffic load and the most enhanced citizen satisfaction with over 30.6% improvements to some of the comparable methods. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

15 pages, 3446 KiB

Open AccessArticle

Federated Reinforcement Learning for Training Control Policies on Multiple IoT Devices

by Hyun-Kyo Lim, Ju-Bong Kim, Joo-Seong Heo and Youn-Hee Han

Sensors 2020, 20(5), 1359; https://doi.org/10.3390/s20051359 - 02 Mar 2020

Cited by 41 | Viewed by 7564

Abstract

Reinforcement learning has recently been studied in various fields and also used to optimally control IoT devices supporting the expansion of Internet connection beyond the usual standard devices. In this paper, we try to allow multiple reinforcement learning agents to learn optimal control [...] Read more.

Reinforcement learning has recently been studied in various fields and also used to optimally control IoT devices supporting the expansion of Internet connection beyond the usual standard devices. In this paper, we try to allow multiple reinforcement learning agents to learn optimal control policy on their own IoT devices of the same type but with slightly different dynamics. For such multiple IoT devices, there is no guarantee that an agent who interacts only with one IoT device and learns the optimal control policy will also control another IoT device well. Therefore, we may need to apply independent reinforcement learning to each IoT device individually, which requires a costly or time-consuming effort. To solve this problem, we propose a new federated reinforcement learning architecture where each agent working on its independent IoT device shares their learning experience (i.e., the gradient of loss function) with each other, and transfers a mature policy model parameters into other agents. They accelerate its learning process by using mature parameters. We incorporate the actor–critic proximal policy optimization (Actor–Critic PPO) algorithm into each agent in the proposed collaborative architecture and propose an efficient procedure for the gradient sharing and the model transfer. Using multiple rotary inverted pendulum devices interconnected via a network switch, we demonstrate that the proposed federated reinforcement learning scheme can effectively facilitate the learning process for multiple IoT devices and that the learning speed can be faster if more agents are involved. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

Review

Jump to: Research

21 pages, 1487 KiB

Open AccessReview

Digital Twin Coaching for Physical Activities: A Survey

by Rogelio Gámez Díaz, Qingtian Yu, Yezhe Ding, Fedwa Laamarti and Abdulmotaleb El Saddik

Sensors 2020, 20(20), 5936; https://doi.org/10.3390/s20205936 - 21 Oct 2020

Cited by 35 | Viewed by 7841

Abstract

Digital Twin technology has been rising in popularity thanks to the popularity of machine learning in the last decade. As the life expectancy of people around the world is increasing, so is the focus on physical activity to remain healthy especially in the [...] Read more.

Digital Twin technology has been rising in popularity thanks to the popularity of machine learning in the last decade. As the life expectancy of people around the world is increasing, so is the focus on physical activity to remain healthy especially in the current times where people are staying sedentary while in quarantine. This article aims to provide a survey on the field of Digital Twin technology focusing on machine learning and coaching techniques as they have not been explored yet. We also define what Digital Twin Coaching is and categorize the work done so far in terms of the objective of the physical activity. We also list common Digital Twin Coaching characteristics found in the articles reviewed in terms of concepts such as interactivity, privacy and security and also detail future perspectives in multimodal interaction and standardization, to name a few, that can guide researchers if they choose to work in this field. Finally, we provide a diagram for the Digital Twin Ecosystem showing the interaction between relevant entities and the information flow as well as an idealization of an ideal Digital Twin Ecosystem for team sports’ athlete tracking. Full article

(This article belongs to the Special Issue Machine Learning for IoT Applications and Digital Twins)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Machine Learning for IoT Applications and Digital Twins

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (9 papers)

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI