Nothing Special   »   [go: up one dir, main page]
Next Issue
Volume 13, December
Previous Issue
Volume 13, August
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
7.9
3.3
Journals
JSAN
Volume 13
Issue 5
share announcement

J. Sens. Actuator Netw., Volume 13, Issue 5 (October 2024) – 22 articles

Cover Story (view full-size image): Underwater optical wireless communication (UOWC) has gained importance with the rise of unmanned underwater vehicles, particularly in offshore industries focused on sustainable food production and energy security. UOWC supports high data rates and low latency for video transmission; however, links are influenced by inhomogeneous light attenuation and sunlight. This study models the underwater spectral light distribution across depths in eight stratified oceanic water types, exploring optimal transmission wavelengths for maximum signal-to-noise ratio (SNR). We challenge the assumption that the blue-green spectrum is always ideal, revealing a unique relationship between wavelength, SNR, and depth. The results have implications for solar discrimination and routing strategies in optical wireless networks in the photic zone. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
19 pages, 985 KiB  
Article
On the Energy Behaviors of the Bellman–Ford and Dijkstra Algorithms: A Detailed Empirical Study
by Othman Alamoudi and Muhammad Al-Hashimi
J. Sens. Actuator Netw. 2024, 13(5), 67; https://doi.org/10.3390/jsan13050067 - 12 Oct 2024
Viewed by 787
Abstract
The Single-Source Shortest Paths (SSSP) graph problem is a fundamental computation. This study attempted to characterize concretely the energy behaviors of the two primary methods to solve it, the Bellman–Ford and Dijkstra algorithms. The very different interactions of the algorithms with the hardware [...] Read more.
The Single-Source Shortest Paths (SSSP) graph problem is a fundamental computation. This study attempted to characterize concretely the energy behaviors of the two primary methods to solve it, the Bellman–Ford and Dijkstra algorithms. The very different interactions of the algorithms with the hardware may have significant implications for energy. The study was motivated by the multidisciplinary nature of the problem. Gaining better insights should help vital applications in many domains. The work used reliable embedded sensors in an HPC-class CPU to collect empirical data for a wide range of sizes for two graph cases: complete as an upper-bound case and moderately dense. The findings confirmed that Dijkstra’s algorithm is drastically more energy efficient, as expected from its decisive time complexity advantage. In terms of power draw, however, Bellman–Ford had an advantage for sizes that fit in the upper parts of the memory hierarchy (up to 2.36 W on average), with a region of near parity in both power draw and total energy budgets. This result correlated with the interaction of lighter logic and graph footprint in memory with the Level 2 cache. It should be significant for applications that rely on solving a lot of small instances since Bellman–Ford is more general and is easier to implement. It also suggests implications for the design and parallelization of the algorithms when efficiency in power draw is in mind. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Average energy consumption (in logarithmic scale), with a view on small cases (linear scale). A region where the trend reverses is highlighted. (<b>a</b>) Fully dense (complete) graphs; (<b>b</b>) Moderately dense graphs.</p> Full article ">Figure 2
<p>Average execution time (logarithmic), with a linear scale view on small cases. Regions identified from the energy figures are highlighted. (<b>a</b>) Fully dense (complete) graphs; (<b>b</b>) Moderately dense graphs.</p> Full article ">Figure 3
<p>Average power consumption in watts. (<b>a</b>) Fully dense (complete graph) cases; (<b>b</b>) Moderately dense cases.</p> Full article ">Figure 4
<p>Graph data structure size (KiB) in memory of a sample of random graphs (base-2 logarithmic). Shaded areas mark the cases that fit in the 256 KiB L2 cache. The inner view (in linear scale) focuses on cases that fit in 512 KiB for perspective.</p> Full article ">Figure 5
<p>Average cache misses by the fully dense (complete) graphs. In (<b>b</b>), the first apparent jump is insignificant and was exaggerated by the logarithmic scale. (<b>a</b>) Level 2 cache miss counts; (<b>b</b>) Level 3 cache miss counts.</p> Full article ">Figure 6
<p>Moderately dense graphs (in lighter color) compared against fully dense ones. (<b>a</b>) L2 cache miss behavior; (<b>b</b>) power consumption patterns.</p> Full article ">
22 pages, 4093 KiB  
Article
Helicopters Turboshaft Engines Neural Network Modeling under Sensor Failure
by Serhii Vladov, Anatoliy Sachenko, Valerii Sokurenko, Oleksandr Muzychuk and Victoria Vysotska
J. Sens. Actuator Netw. 2024, 13(5), 66; https://doi.org/10.3390/jsan13050066 - 10 Oct 2024
Viewed by 684
Abstract
This article discusses the development of an enhanced monitoring and control system for helicopter turboshaft engines during flight operations, leveraging advanced neural network techniques. The research involves a comprehensive mathematical model that effectively simulates various failure scenarios, including single and cascading failure, such [...] Read more.
This article discusses the development of an enhanced monitoring and control system for helicopter turboshaft engines during flight operations, leveraging advanced neural network techniques. The research involves a comprehensive mathematical model that effectively simulates various failure scenarios, including single and cascading failure, such as disconnections of gas-generator rotor sensors. The model employs differential equations to incorporate time-varying coefficients and account for external disturbances, ensuring accurate representation of engine behavior under different operational conditions. This study validates the NARX neural network architecture with a backpropagation training algorithm, achieving 99.3% accuracy in fault detection. A comparative analysis of the genetic algorithms indicates that the proposed algorithm outperforms others by 4.19% in accuracy and exhibits superior performance metrics, including a lower loss. Hardware-in-the-loop simulations in Matlab Simulink confirm the effectiveness of the model, showing average errors of 1.04% and 2.58% at 15 °C and 24 °C, respectively, with high precision (0.987), recall (1.0), F1-score (0.993), and an AUC of 0.874. However, the model’s accuracy is sensitive to environmental conditions, and further optimization is needed to improve computational efficiency and generalizability. Future research should focus on enhancing model adaptability and validating performance in real-world scenarios. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>The proposed nonlinear autoregression neural network with exogenous inputs (NARX).</p> Full article ">Figure 2
<p>The TV3-117 turboshaft engine parameters dynamics time series using digitized oscillograms: (<b>black curve</b>): Gas-generator rotor r.p.m; (<b>green curve</b>) Free turbine rotor speed; (<b>red curve</b>) Gas temperature in front of the compressor turbine (author’s research).</p> Full article ">Figure 3
<p>Cluster analysis results: (<b>a</b>) Training dataset, (<b>b</b>) Test dataset (author’s research).</p> Full article ">Figure 4
<p>Scheme of the helicopter turboshaft engine model with the semi-physical modeling stand interaction (author’s research).</p> Full article ">Figure 5
<p>Overall view of the NARX neural network interaction with the semi-physical modeling stand implementation within the Matlab Simulink environment (author’s research).</p> Full article ">Figure 6
<p>Resulting diagrams: (<b>a</b>–<b>c</b>) are the simulated engine thermogas-dynamic parameters taking into account sensor break; (<b>d</b>–<b>f</b>) are the discrete signals during the engine model reconfiguration (author’s research).</p> Full article ">Figure 6 Cont.
<p>Resulting diagrams: (<b>a</b>–<b>c</b>) are the simulated engine thermogas-dynamic parameters taking into account sensor break; (<b>d</b>–<b>f</b>) are the discrete signals during the engine model reconfiguration (author’s research).</p> Full article ">Figure 7
<p>Diagram of the model error magnitude over time: (<b>a</b>) On the first run, (<b>b</b>) On the second run (author’s research).</p> Full article ">Figure 8
<p>Accuracy metric diagram (author’s research).</p> Full article ">Figure 9
<p>The obtained AUC-ROC curve (author’s research).</p> Full article ">
16 pages, 4527 KiB  
Article
High-Transparency Linear Actuator Using an Electromagnetic Brake for Damping Modulation in Physical Human–Robot Interaction
by Zahid Ullah, Thachapan Sermsrisuwan, Khemwutta Pornpipatsakul, Ronnapee Chaichaowarat and Witaya Wannasuphoprasit
J. Sens. Actuator Netw. 2024, 13(5), 65; https://doi.org/10.3390/jsan13050065 - 10 Oct 2024
Viewed by 599
Abstract
Enhancing the transparency of high-transmission-ratio linear actuators is crucial for improving the safety and capability of high-force robotic systems having physical contact with humans in unstructured environments. However, realizing such enhancement is challenging. A proposed solution for active body weight support systems involves [...] Read more.
Enhancing the transparency of high-transmission-ratio linear actuators is crucial for improving the safety and capability of high-force robotic systems having physical contact with humans in unstructured environments. However, realizing such enhancement is challenging. A proposed solution for active body weight support systems involves employing a macro–mini linear actuator incorporating an electrorheological-fluid brake to connect a high-force unit with an agile, highly back-drivable unit. This paper introduces the use of an electromagnetic (EM) brake with reduced rotor inertia to address this challenge. The increased torque capacity of the EM brake enables integration with a low-gear-ratio linear transmission. The agile translation of the endpoint is propelled by a low-inertia motor (referred to as the “mini”) via a pulley-belt mechanism to achieve high transparency. The rotor of the EM brake is linked to the pulley. Damping modulation under high driving force is achieved through the adjustment of the brake torque relative to the rotational speed of the pulley. When the brake is engaged, it prevents any relative motion between the endpoint and the moving carrier. The endpoint is fully controlled by the ball screw of the high-force unit, referred to as the “macro”. A scaled prototype was constructed to experimentally characterize the damping force generated by the mini motor and the EM brake. The macro–mini linear actuator, equipped with an intrinsic failsafe feature, can be utilized for active body weight support systems that demand high antigravity force. Full article
(This article belongs to the Section Actuators, Sensors and Devices)
►▼ Show Figures

Figure 1

Figure 1
<p>System network flow diagram illustrating the configuration of a control system that uses an Arduino Mega 2560 microcontroller to process the data. The orange dashed lines indicate the network comprising the EM brake and the mini actuator, responsible for dynamic damping adjustments. The blue dashed lines denote the network involving the mini and the macro actuators along with their respective sensors (encoders), highlighting the systems’ integrated sensor-network configuration.</p> Full article ">Figure 2
<p>Conceptual design of a high transparency linear actuator utilizing the EM brake to engage the high-force unit (macro) and the low-impedance unit (mini).</p> Full article ">Figure 3
<p>Dynamic model of the macro–mini linear actuator using an EM brake for damping modulation.</p> Full article ">Figure 4
<p>CAD rendering of the macro–mini linear actuator using an EM brake.</p> Full article ">Figure 5
<p>Alpha prototype of the macro–mini linear actuator using an EM brake.</p> Full article ">Figure 6
<p>Damping estimated from the vibration test. At zero damping command, the estimated damping represents an intrinsic property of the system attributed to the viscous components. The total damping exhibits a linear increase with the commanded damping of the mini motor.</p> Full article ">Figure 7
<p>Inertia estimated from the vibration test. The estimated inertia is another intrinsic property of the system, which is constant against the commanded damping of the mini motor.</p> Full article ">Figure 8
<p>Driving torque required from the mini motor to move the end effector at constant speeds. Positive angular velocity corresponds to upward motion and vice versa. The slope of the velocity–torque relationship reflects the intrinsic damping. Static friction is estimated from half of the offset between the intersections of the trendlines with the vertical axis. The influence of gravity is evident in the vertical shift of the starting torque.</p> Full article ">Figure 9
<p>Experimental setup with a fixed end effector to enable the macro motor to generate motion against the EM brake torque. The root of the load cell is positioned on the end effector, while the tip of the load cell is anchored to the stationary base.</p> Full article ">Figure 10
<p>Response to the constant velocity reference at two different speeds. (<b>a</b>) Recorded angular velocity of the mini motor. (<b>b</b>) PWM command for the EM brake. (<b>c</b>) Interaction force is measured by the load cell.</p> Full article ">Figure 11
<p>Response to the sinusoidal velocity reference of frequency 0.1 Hz. (<b>a</b>) Recorded angular velocity of the mini motor. (<b>b</b>) PWM command for the EM brake. (<b>c</b>) Interaction force measured by the load cell. The cyan plot shows the ideal force directly computed from the rotational speed of mini the motor.</p> Full article ">Figure 12
<p>Response to the sinusoidal velocity reference of frequency 0.5 Hz. (<b>a</b>) Recorded angular velocity of the mini motor. (<b>b</b>) PWM command for the EM brake. (<b>c</b>) Interaction force measured by the load cell. The cyan plot shows the ideal force directly computed from the rotational speed of mini the motor.</p> Full article ">
20 pages, 3912 KiB  
Article
A Spectral-Based Blade Fault Detection in Shot Blast Machines with XGBoost and Feature Importance
by Joon-Hyuk Lee, Chibuzo Nwabufo Okwuosa, Baek Cheon Shin and Jang-Wook Hur
J. Sens. Actuator Netw. 2024, 13(5), 64; https://doi.org/10.3390/jsan13050064 - 9 Oct 2024
Viewed by 756
Abstract
The optimal functionality and dependability of mechanical systems are important for the sustained productivity and operational reliability of industrial machinery, and have a direct impact on its longevity and profitability. Therefore, the failure of a mechanical system or any of its components would [...] Read more.
The optimal functionality and dependability of mechanical systems are important for the sustained productivity and operational reliability of industrial machinery, and have a direct impact on its longevity and profitability. Therefore, the failure of a mechanical system or any of its components would be detrimental to production continuity and availability. Consequently, this study proposes a robust diagnostic framework for analyzing the blade conditions of shot blast industrial machinery. The framework explores the spectral characteristics of the vibration signals generated by the industrial shot blast for discriminative feature excitement. Furthermore, a peak detection algorithm is introduced to identify and extract the unique features present in the peak magnitudes of each signal spectrum. A feature importance algorithm is then deployed as the feature selection tool, and these selected features are fed into ten machine learning classifiers (MLCs), with extreme gradient boosting (XGBoost (version 2.1.1)) as the core classifier. The results show that the XGBoost classifier achieved the best accuracy of 98.05%, with a cost-efficient computational cost of 0.83 s. Other global assessment metrics were also implemented in the study to further validate the model. Full article
(This article belongs to the Special Issue Fault Diagnosis in the Internet of Things Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Shot Blast Working Mechanism.</p> Full article ">Figure 2
<p>The proposed diagnostic flowchart.</p> Full article ">Figure 3
<p>Detailed diagnostic model.</p> Full article ">Figure 4
<p>Data collection setup.</p> Full article ">Figure 5
<p>Shot blast product finishing: (<b>a</b>) Raw input; and (<b>b</b>) Finished product.</p> Full article ">Figure 6
<p>Blade conditions: (<b>a</b>) Healthy blade; and (<b>b</b>) Faulty blade.</p> Full article ">Figure 7
<p>Current signals collected from the shot blast machine: (<b>a</b>) Raw healthy signal; (<b>b</b>) Raw faulty signal; (<b>c</b>) FFT of the healthy signal; and (<b>d</b>) FFT of the faulty signal.</p> Full article ">Figure 8
<p>Peak detection feature extraction: (<b>a</b>) Peak detection extract from healthy signal’s FFT; (<b>b</b>) Peak detection extract from faulty signal’s FFT.</p> Full article ">Figure 9
<p>Signal processing: (<b>a</b>) Correlation plot of all extracted features; (<b>b</b>) Feature importance plot of all extracted features; (<b>c</b>) Selected feature plot; and (<b>d</b>) Correlation plot of the selected features.</p> Full article ">Figure 10
<p>Performance metrics plot of all ML classifier models.</p> Full article ">Figure 11
<p>Test data confusion matrix of all ML classifier algorithms: (<b>a</b>) RBF SVM; (<b>b</b>) Linear SVM; (<b>c</b>) KNN; (<b>d</b>) SGD; (<b>e</b>) RF; (<b>f</b>) DT; (<b>g</b>) NBC; (<b>h</b>) ABC; (<b>i</b>) XGBoost; and (<b>j</b>) GBC.</p> Full article ">
17 pages, 4996 KiB  
Article
Safeguarding Personal Identifiable Information (PII) after Smartphone Pairing with a Connected Vehicle
by Jason Carlton and Hafiz Malik
J. Sens. Actuator Netw. 2024, 13(5), 63; https://doi.org/10.3390/jsan13050063 - 6 Oct 2024
Viewed by 746
Abstract
The integration of connected autonomous vehicles (CAVs) has significantly enhanced driving convenience, but it has also raised serious privacy concerns, particularly regarding the personal identifiable information (PII) stored on infotainment systems. Recent advances in connected and autonomous vehicle control, such as multi-agent system [...] Read more.
The integration of connected autonomous vehicles (CAVs) has significantly enhanced driving convenience, but it has also raised serious privacy concerns, particularly regarding the personal identifiable information (PII) stored on infotainment systems. Recent advances in connected and autonomous vehicle control, such as multi-agent system (MAS)-based hierarchical architectures and privacy-preserving strategies for mixed-autonomy platoon control, underscore the increasing complexity of privacy management within these environments. Rental cars with infotainment systems pose substantial challenges, as renters often fail to delete their data, leaving it accessible to subsequent renters. This study investigates the risks associated with PII in connected vehicles and emphasizes the necessity of automated solutions to ensure data privacy. We introduce the Vehicle Inactive Profile Remover (VIPR), an innovative automated solution designed to identify and delete PII left on infotainment systems. The efficacy of VIPR is evaluated through surveys, hands-on experiments with rental vehicles, and a controlled laboratory environment. VIPR achieved a 99.5% success rate in removing user profiles, with an average deletion time of 4.8 s or less, demonstrating its effectiveness in mitigating privacy risks. This solution highlights VIPR as a critical tool for enhancing privacy in connected vehicle environments, promoting a safer, more responsible use of connected vehicle technology in society. Full article
(This article belongs to the Special Issue Feature Papers in the Section of Network Security and Privacy)
►▼ Show Figures

Figure 1

Figure 1
<p>Illustration of the Vehicle-to-Everything (V2X) communications model.</p> Full article ">Figure 2
<p>Illustration of modern in-vehicle network architecture.</p> Full article ">Figure 3
<p>High-level illustration PII leakage through Bluetooth pairing and removing user profiles using VIPR.</p> Full article ">Figure 4
<p>VIPR state diagram for rental vehicle depot return and subsequent rentals.</p> Full article ">Figure 5
<p>VIPR state diagram for ride-sharing.</p> Full article ">Figure 6
<p>Replicated vehicle infotainment system.</p> Full article ">Figure 7
<p>Infotainment system display showing current and previous paired devices (active or inactive).</p> Full article ">Figure 8
<p>Menu illustration of current (green) and previous (red) paired devices.</p> Full article ">Figure 9
<p>VIPR automatic removal of inactive profiles.</p> Full article ">Figure 10
<p>Menu illustration of active profiles after the VIPR executes.</p> Full article ">
30 pages, 3145 KiB  
Article
Create a Realistic IoT Dataset Using Conditional Generative Adversarial Network
by Miada Almasre and Alanoud Subahi
J. Sens. Actuator Netw. 2024, 13(5), 62; https://doi.org/10.3390/jsan13050062 - 3 Oct 2024
Viewed by 693
Abstract
The increased use of Internet of Things (IoT) devices has led to greater threats to privacy and security. This has created a need for more effective cybersecurity applications. However, the effectiveness of these systems is often limited by the lack of comprehensive and [...] Read more.
The increased use of Internet of Things (IoT) devices has led to greater threats to privacy and security. This has created a need for more effective cybersecurity applications. However, the effectiveness of these systems is often limited by the lack of comprehensive and balanced datasets. This research contributes to IoT security by tackling the challenges in dataset generation and providing a valuable resource for IoT security research. Our method involves creating a testbed, building the ‘Joint Dataset’, and developing an innovative tool. The tool consists of two modules: an Exploratory Data Analysis (EDA) module, and a Generator module. The Generator module uses a Conditional Generative Adversarial Network (CGAN) to address data imbalance and generate high-quality synthetic data that accurately represent real-world network traffic. To showcase the effectiveness of the tool, the proportion of imbalance reduction in the generated dataset was computed and benchmarked to the BOT-IOT dataset. The results demonstrated the robustness of synthetic data generation in creating balanced datasets. Full article
(This article belongs to the Special Issue Security and Smart Applications in IoT and Wireless Sensor and Actuator Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>The IoT testbed for collecting attack traffic.</p> Full article ">Figure 2
<p>Reconnaissance, TCP-DOS, UDP-DOS attack.</p> Full article ">Figure 3
<p>Man-in-the-Middle attack.</p> Full article ">Figure 4
<p>Deauthentication attack.</p> Full article ">Figure 5
<p>The overview of the tool prototype.</p> Full article ">Figure 6
<p>The main page of the CGAN-based tool.</p> Full article ">Figure 7
<p>EDA of the dataset analysis.</p> Full article ">Figure 8
<p>Select the dataset as CSV file.</p> Full article ">Figure 9
<p>Example of user selection.</p> Full article ">Figure 10
<p>ANOVA F-test with the threshold to select the best number of features.</p> Full article ">Figure 11
<p>The generator and discriminator networks.</p> Full article ">Figure 12
<p>Full page view of CGAN processing the user dataset.</p> Full article ">Figure 13
<p>Synthetic and balanced IoT dataset ready to downloaded on the user device using CGAN.</p> Full article ">Figure 14
<p>The imbalance reduction proportion of Joint Dataset.</p> Full article ">Figure 15
<p>The imbalance reduction proportion of BoT-IoT.</p> Full article ">Figure 16
<p>The comparison of synthetic and original in BOT dataset.</p> Full article ">Figure 17
<p>The comparison of synthetic and original in Joint Dataset.</p> Full article ">Figure 18
<p>The Joint Dataset categorical features cumulative difference.</p> Full article ">
17 pages, 491 KiB  
Article
Dynamic Event-Triggered Control for Sensor–Controller–Actuator Networked Control Systems
by Mahmoud Abdelrahim and Dhafer Almakhles
J. Sens. Actuator Netw. 2024, 13(5), 61; https://doi.org/10.3390/jsan13050061 - 1 Oct 2024
Viewed by 703
Abstract
We consider the problem of output feedback stabilization of LTI systems under event-triggering implementation. In particular, we assume that both the plant output and the control input are both transmitted over the network in an asynchronous manner. To that end, two independent event-triggering [...] Read more.
We consider the problem of output feedback stabilization of LTI systems under event-triggering implementation. In particular, we assume that both the plant output and the control input are both transmitted over the network in an asynchronous manner. To that end, two independent event-triggering rules are constructed to generate the transmission instants of the submitted signals. The proposed approach is dynamic in the sense that the triggering rules involve internal dynamical variables to allow for further reduction in the communication load. Moreover, the inter-transmission times for both sides of the channel are lower bound by enforced dwell times to prevent the occurrence of Zeno phenomena. The problem is challenging due to mutual interactions between the sampling errors of the plant output and the control input, which requires careful handling to ensure closed-loop stability. The triggering mechanisms are designed by emulation as we first ignore the effect of the network and stabilize the plant in continuous-time. Then, the communication constraints are taken into account to derive the triggering conditions such that the stability of the networked control system is preserved. The required conditions are formulated in terms of a linear matrix inequality. The effectiveness of the technique is demonstrated by numerical simulations. Full article
(This article belongs to the Section Communications and Networking)
►▼ Show Figures

Figure 1

Figure 1
<p>Event-triggered control schematic.</p> Full article ">Figure 2
<p>Layout of DFIG-based wind turbine control.</p> Full article ">Figure 3
<p>State trajectory for the plant and the observer.</p> Full article ">Figure 4
<p>Sampling-induced errors.</p> Full article ">Figure 5
<p>Inter-transmission times for the output.</p> Full article ">Figure 6
<p>Inter-transmission times for the input.</p> Full article ">
30 pages, 4047 KiB  
Article
Advanced Data Augmentation Techniques for Enhanced Fault Diagnosis in Industrial Centrifugal Pumps
by Dong-Yun Kim, Akeem Bayo Kareem, Daryl Domingo, Baek-Cheon Shin and Jang-Wook Hur
J. Sens. Actuator Netw. 2024, 13(5), 60; https://doi.org/10.3390/jsan13050060 - 25 Sep 2024
Viewed by 2190
Abstract
This study presents an advanced data augmentation framework to enhance fault diagnostics in industrial centrifugal pumps using vibration data. The proposed framework addresses the challenge of insufficient defect data in industrial settings by integrating traditional augmentation techniques, such as Gaussian noise (GN) and [...] Read more.
This study presents an advanced data augmentation framework to enhance fault diagnostics in industrial centrifugal pumps using vibration data. The proposed framework addresses the challenge of insufficient defect data in industrial settings by integrating traditional augmentation techniques, such as Gaussian noise (GN) and signal stretching (SS), with advanced models, including Long Short-Term Memory (LSTM) networks, Autoencoders (AE), and Generative Adversarial Networks (GANs). Our approach significantly improves the robustness and accuracy of machine learning (ML) models for fault detection and classification. Key findings demonstrate a marked reduction in false positives and a substantial increase in fault detection rates, particularly in complex operational scenarios where traditional statistical methods may fall short. The experimental results underscore the effectiveness of combining these augmentation techniques, achieving up to a 30% improvement in fault detection accuracy and a 25% reduction in false positives compared to baseline models. These improvements highlight the practical value of the proposed framework in ensuring reliable operation and the predictive maintenance of centrifugal pumps in diverse industrial environments. Full article
(This article belongs to the Special Issue Fault Diagnosis in the Internet of Things Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Framework for using data augmentation using Gaussian noise and signal stretching.</p> Full article ">Figure 2
<p>Framework for the integration of LSTMAEGAN architecture.</p> Full article ">Figure 3
<p>Vibration data collection setup for the water pump system.</p> Full article ">Figure 4
<p>Time–frequency analysis of vibration signals from different conditions. The top row shows the STFT of vibration data: (<b>a</b>) Normal, (<b>b</b>) Crack, and (<b>c</b>) Wear. The bottom row displays the CWT of the same signals: (<b>d</b>) Normal, (<b>e</b>) Crack, and (<b>f</b>) Wear.</p> Full article ">Figure 5
<p>Analysis of vibration signals using the HT. For each dataset, (<b>a</b>) Normal, (<b>b</b>) Crack, and (<b>c</b>) Wear, the plot shows the following: (1) The original signal, (2) The amplitude envelope computed from the Hilbert Transform, and (3) The instantaneous frequency derived from the phase of the analytic signal.</p> Full article ">Figure 6
<p>Before augmentation: (<b>a</b>) Correlation plot of all statistical features, showing the degree of the linear relationship between each pair of features. This plot helps identify multi-collinearity among the features. (<b>b</b>) Selected features after applying a PCC threshold of &lt;0.9, highlighting the features with lower inter-correlation, thus reducing redundancy and improving the robustness of the model.</p> Full article ">Figure 7
<p>After augmentation: (<b>a</b>) Correlation plot of all statistical features, showing the degree of linear relationship between each pair of features. This plot helps identify multi-collinearity among the features. (<b>b</b>) Selected features after applying a PCC threshold of &lt;0.9, highlighting the features with lower inter-correlation, thus reducing redundancy and improving the robustness of the model.</p> Full article ">Figure 8
<p>Plot of the normalized weighted average of Gaussian noise and signal stretching under different fault conditions: normal, wear, and crack.</p> Full article ">Figure 9
<p>Confusion matrices for the classification performance of three models before augmentation: (<b>a</b>) SVM, (<b>b</b>) RF, and (<b>c</b>) GB. Each matrix shows the classification of vibration signal labels: “Normal”, “Wear”, and “Crack Fault”.</p> Full article ">Figure 10
<p>Confusion matrices for the classification performance of three models after augmentation: (<b>a</b>) SVM, (<b>b</b>) RF, and (<b>c</b>) GB. Each matrix shows the classification of vibration signal labels: “Normal”, “Wear”, and “Crack Fault”.</p> Full article ">Figure 11
<p>(<b>a</b>) Distribution of reconstruction errors for the dataset, with the red dashed line indicating the chosen threshold for anomaly detection. The histogram shows the frequency of Mean Squared Errors (MSEs), helping to visualize the separation between normal and abnormal data points. (<b>b</b>) Time series plot of the original data, with anomalies highlighted in red. The anomalies, identified based on the reconstruction error threshold, are marked against the backdrop of the normal data, illustrating the model’s ability to detect deviations over time. (<b>c</b>) Reconstruction error for anomaly data over different sample indices. The blue line represents the reconstruction error for each sample, and the red dashed line represents the threshold for detecting anomalies, showing which data points exceed the anomaly detection threshold.</p> Full article ">
20 pages, 3288 KiB  
Article
Task Scheduling Algorithm for Power Minimization in Low-Cost Disaster Monitoring System: A Heuristic Approach
by Chanankorn Jandaeng , Jongsuk Kongsen , Peeravit Koad, May Thu and Sirirat Somchuea
J. Sens. Actuator Netw. 2024, 13(5), 59; https://doi.org/10.3390/jsan13050059 - 24 Sep 2024
Viewed by 708
Abstract
This study investigates the optimization of a low-cost IoT-based weather station designed for disaster monitoring, focusing on minimizing power consumption. The system architecture includes application, middleware, communication, and sensor layers, with solar power as the primary energy source. A novel task scheduling algorithm [...] Read more.
This study investigates the optimization of a low-cost IoT-based weather station designed for disaster monitoring, focusing on minimizing power consumption. The system architecture includes application, middleware, communication, and sensor layers, with solar power as the primary energy source. A novel task scheduling algorithm was developed to reduce power usage by efficiently managing the sensing and data transmission periods. Experiments compared the energy consumption of polling and deep sleep techniques, revealing that deep sleep is more energy-efficient (4.73% at 15 s time intervals and 16.45% at 150 s time intervals). Current consumption was analyzed across different test scenarios, confirming that efficient task scheduling significantly reduces power consumption. The energy consumption models were developed to quantify power usage during the sensing and transmission phases. This study concludes that the proposed system, utilizing affordable hardware and solar power, is an effective and sustainable solution for disaster monitoring. Despite using non-low-power devices, the results demonstrate the importance of adaptive task scheduling in extending the operational life of IoT devices. Future work will focus on implementing dynamic scheduling and low-power routing algorithms to enhance system functionality in resource-constrained environments. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>The power consumption of the node.</p> Full article ">Figure 2
<p>The system architecture of the IoT weather station. The data line is represented with the dotted line to turn the network module on/off via relay.</p> Full article ">Figure 3
<p>The low-cost disaster monitoring system and its peripheral sensor.</p> Full article ">Figure 4
<p>The sequence diagram of the software architecture.</p> Full article ">Figure 5
<p>The sequence diagram of the software architecture.</p> Full article ">Figure 6
<p>Current consumption of all test cases.</p> Full article ">Figure 7
<p>The average current consumption (A) for 40 rounds per interval.</p> Full article ">Figure 8
<p>The power source estimation of the proposed system.</p> Full article ">
27 pages, 22292 KiB  
Article
RFSoC Softwarisation of a 2.45 GHz Doppler Microwave Radar Motion Sensor
by Peter Hobden, Edmond Nurellari and Saket Srivastava
J. Sens. Actuator Netw. 2024, 13(5), 58; https://doi.org/10.3390/jsan13050058 - 23 Sep 2024
Viewed by 1098
Abstract
Microwave Doppler sensors are used extensively in motion detection as they are energy-efficient, small-size and relatively low-cost sensors. Common applications of microwave Doppler sensors are for detecting intrusion behind a car roof liner inside an automotive vehicle and to detect moving objects. These [...] Read more.
Microwave Doppler sensors are used extensively in motion detection as they are energy-efficient, small-size and relatively low-cost sensors. Common applications of microwave Doppler sensors are for detecting intrusion behind a car roof liner inside an automotive vehicle and to detect moving objects. These applications require a millisecond response from the target for effective detection. A Doppler microwave sensor is ideally suited to the task, as we are only interested in movement of a large water-based mass (i.e., a person) (FMCW Radar also detect static objects). Although microwave components at 2.45 GHz are now relatively cheap due to mass production of other Industrial Scientific and Medical application (ISM) devices, they do require tuning for temperature compensation, dielectric, and manufacturing variability. A digital solution would be ideal, as chip solutions are known to be more repeatable, but Application-Specific Integrated Circuits (ASICs) are expensive to initially prototype. This paper presents the first completely digital Doppler motion sensor solution at 2.45 GHz, implemented on the new RFSoC from Xilinx without the need to up/downconvert the frequency externally. Our proposed system uses a completely digital approach bringing the benefits of product repeatability, better overtemperature performance and softwarisation, without compromising any performance metric associated with a comparable analogue motion sensor. The RFSoC shows to give superior distance versus false detection, as the Signal-to-Noise Ratio (SNR) is better than a typical analogue system. This is mainly due to the high gain amplification requirement of an analogue system, making it susceptible to electrical noise appearing in the intermediate-frequency (IF) baseband. The proposed RFSoC-based Doppler sensor shows how digital technology can replace traditional analogue radio frequency (RF). A case study is presented showing how we can use a novel method of using multiple Doppler channels to provide range discrimination, which can be performed in both analogue and in a digital implementation (RFSoC). Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Phase vector—used to provide a high resolution of the Doppler return by combining the frequency and phase shift of a Doppler return.</p> Full article ">Figure 2
<p>Spectrum analyser (RFSoC) PYNQ software framework (frequencies up to 4.096 GHz using higher-order Nyquist zone techniques) [<a href="#B13-jsan-13-00058" class="html-bibr">13</a>], showing (<b>A</b>) <math display="inline"><semantics> <mrow> <mn>2.45</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math> carrier and (<b>B</b>) <math display="inline"><semantics> <mrow> <mn>1.6</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math> fold of the <math display="inline"><semantics> <mrow> <mn>2.45</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math> carrier into the 1st Nyquist zone.</p> Full article ">Figure 3
<p>The Xilinx RFSoC Frequency planner setup for the ‘TX carrier signal’ (in blue) at <math display="inline"><semantics> <mrow> <mn>2.45</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math>, showing an image in the 1st Nyquist at <math display="inline"><semantics> <mrow> <mn>1.645</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math>. This is confirmed in <a href="#jsan-13-00058-f002" class="html-fig">Figure 2</a>, where we see the carrier and fold on the proposed sensor’s RF output (HD = harmonic distortion; dotted lines denote Nyquist zones (NYQ)).</p> Full article ">Figure 4
<p>Xilinx RF data converter’s IP block—DDS-generated <math display="inline"><semantics> <mrow> <mn>2.45</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math>, from the onboard 6 Gasps DAC.</p> Full article ">Figure 5
<p>Patch antenna design for the transmit and receive modes. (<math display="inline"><semantics> <msub> <mi>ε</mi> <mi>r</mi> </msub> </semantics></math> = dielectric constant of about 4.5). The width W of the antenna controls the input impedance (300 <math display="inline"><semantics> <mo>Ω</mo> </semantics></math>). By increasing the width, the impedance can be reduced. However, to decrease the input impedance to 50 <math display="inline"><semantics> <mo>Ω</mo> </semantics></math> requires a very wide patch [<a href="#B16-jsan-13-00058" class="html-bibr">16</a>] (90 mm for RF4 at <math display="inline"><semantics> <mrow> <mn>2.45</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math>).</p> Full article ">Figure 6
<p>Hemispherical radiation pattern plot taken from the proposed patch antenna.</p> Full article ">Figure 7
<p>High-level diagram of the RF DACs (top) and ADCs (bottom).</p> Full article ">Figure 8
<p>(<b>A</b>) Metal plate moving towards the antenna at walking pace, from a distance of 3 m, (<b>B</b>) and away from RFSoC antenna. (<b>C</b>) The frequency remains constant when there is no motion detected (taken in UoL’s anechoic chamber).</p> Full article ">Figure 9
<p>Top Level diagram of the RFSoC-based Doppler motion sensor: implementation in Matlab Simulink using Sysgen. Note that the design is synthesised under Xilinx Vivado for the XCZU28DR-2E RFSoC.</p> Full article ">Figure 10
<p>Recorded phase noise for <math display="inline"><semantics> <mrow> <mn>2.45</mn> </mrow> </semantics></math> <math display="inline"><semantics> <mi mathvariant="normal">G</mi> </semantics></math><math display="inline"><semantics> <mi>Hz</mi> </semantics></math>, taken on an R&amp;S©FSWP phase-noise analyser and Voltage-Controlled Oscillator (VCO) tester [<a href="#B23-jsan-13-00058" class="html-bibr">23</a>] Note that the non-linearity at the frequency offset of <math display="inline"><semantics> <msup> <mn>10</mn> <mn>5</mn> </msup> </semantics></math> is due to the effect of the PLL charge pump [<a href="#B24-jsan-13-00058" class="html-bibr">24</a>].</p> Full article ">Figure 11
<p>RFSoC 2 × 2 development board from GlobalTech, connected to a patch antenna.</p> Full article ">Figure 12
<p>(<b>Left</b>): Proposed RFSoC-based Doppler sensor. (<b>Right</b>): Typical analogue Doppler sensor. The proposed RFSoC Doppler sensor’s main components consist of a crystal/PLL, RFSoC FPGA, and BALUN. A typical analogue Doppler sensor consists of a Voltage-Controlled Oscillator (VCO), mixer, power splitter, amplifier, sample and hold (or ADC), and a Microcontroller Unit (MCU). Both designs have an antenna on the bottom side of the PCB and an optional CANbus interface. From a PCB design prospective, the analogue sensor is more complex. The typical analogue sensor does have more functional chips in the chip list, but compared with the digital RFSoC, the circuit complexity of microprocessors and others does not seem to be as high as that of RFSoC. For example, analogue sensors using the bandpass IF sampling principle only need a crystal oscillator, PLL with integrated Voltage-Controlled Oscillator (VCO), ADC with a high enough analogue bandwidth, and a low-cost Microcontroller Unit (MCU). The RFSoC presents the concept of a single-chip solution, bringing the benefit of a digital system.</p> Full article ">Figure 13
<p>After decimation, the phase information can be preserved then used to re-calculate the frequency shift. Here, we can see a reconstructed frequency shift from the preserved phase information, giving us a finer resolution otherwise achieved from just straight decimation.</p> Full article ">Figure 14
<p>Boyer technique applied to two baseband signals, from the dual-antenna arrangement. Here we see the receding target: Doppler from lags <math display="inline"><semantics> <mfenced separators="" open="(" close=")"> <msub> <mi>ω</mi> <mn>0</mn> </msub> <mo>−</mo> <msub> <mi>ω</mi> <mn>1</mn> </msub> </mfenced> </semantics></math>, approaching target: Doppler from leads <math display="inline"><semantics> <mfenced separators="" open="(" close=")"> <msub> <mi>ω</mi> <mn>0</mn> </msub> <mo>+</mo> <msub> <mi>ω</mi> <mn>1</mn> </msub> </mfenced> </semantics></math>.</p> Full article ">Figure 15
<p>Angle verses range factor—ideal conditions over the range from <math display="inline"><semantics> <mrow> <mo>+</mo> <msup> <mn>90</mn> <mi mathvariant="normal">o</mi> </msup> </mrow> </semantics></math> to <math display="inline"><semantics> <mrow> <mo>−</mo> <msup> <mn>90</mn> <mi mathvariant="normal">o</mi> </msup> </mrow> </semantics></math>.</p> Full article ">Figure 16
<p>Ground plane distortion caused by the roof of the car being a metal body. In <span style="color: #0000FF">blue</span> is antenna one and in <span style="color: #FF0000">red</span> is antenna two.</p> Full article ">Figure 17
<p>Angle versus range factor—real world conditions.</p> Full article ">Figure 18
<p>Multiple charge pumps.</p> Full article ">Figure 19
<p>Charge pump on the extracted signal as shown in <span style="color: #FFFF00">yellow</span>.</p> Full article ">Figure 20
<p>Sensitivity contours for a single beam sensor. Reduced sensitivity shown in <span style="color: #00FF00">green</span> contour and high-sensitivity contour shown in <span style="color: #FF0000">red</span>.</p> Full article ">Figure 21
<p>Sensitivity contours when isolating three sensitive areas in the car. <span style="color: #FF00FF">Magenta</span> zone covering the rear and back seats. <span style="color: #00FF00">Green</span> is the front and <span style="color: #ff9900">orange</span> is for the boot.</p> Full article ">Figure 22
<p>Distribution of phase error with changing acceleration.</p> Full article ">Figure 23
<p>Phase error against increasing range.</p> Full article ">Figure 24
<p>Measured frequency phase error against Doppler frequency.</p> Full article ">Figure 25
<p>Captured Boyer results. The <span style="color: #0000FF">blue</span> channel is the Doppler return from antenna one and the <span style="color: #FF00FF">magenta</span> is the Doppler return from channel two. The <span style="color: #FFFF00">yellow</span> trace is the processed ‘Boyer’ channel.</p> Full article ">Figure 26
<p>Boyer scheme tested with a 1 m<sup>2</sup> metal plate manually moved towards and away from the sensor. In <span style="color: #0000FF">blue</span> is the Doppler return from antenna one and in <span style="color: #FF00FF">magenta</span> is antenna two. Both channels are plotted against distance and speed (where the amplitude is used for the distance, and frequency/phase change relates to the speed of travel).</p> Full article ">Figure 27
<p>Error distribution for 500 range measurements (10–480 cm).</p> Full article ">
20 pages, 566 KiB  
Article
Predictive Maintenance in IoT-Monitored Systems for Fault Prevention
by Enrico Zero, Mohamed Sallak and Roberto Sacile
J. Sens. Actuator Netw. 2024, 13(5), 57; https://doi.org/10.3390/jsan13050057 - 19 Sep 2024
Viewed by 4378
Abstract
This paper focuses on predictive maintenance for simple machinery systems monitored by the Internet of Things (IoT). As these systems can be challenging to model due to their complexity, diverse typologies, and limited operational lifespans, traditional predictive maintenance approaches face obstacles due to [...] Read more.
This paper focuses on predictive maintenance for simple machinery systems monitored by the Internet of Things (IoT). As these systems can be challenging to model due to their complexity, diverse typologies, and limited operational lifespans, traditional predictive maintenance approaches face obstacles due to the lack of extensive historical data. To address this issue, we propose a novel clustering-based process that identifies potential machinery faults. The proposed approach lies in empowering decision-makers to define predictive maintenance policies based on the reliability of the proposed fault classification. Through a case study involving real sensor data from the doors of a transportation vehicle, specifically a bus, we demonstrate the practical applicability and effectiveness of our method in preemptively preventing faults and enhancing maintenance practices. By leveraging IoT sensor data and employing clustering techniques, our approach offers a promising avenue for cost-effective predictive maintenance strategies in simple machinery systems. As part of the quality assurance, a comparison between the predictive maintenance model for a simple machinery system, pattern recognition neural network, and support vector machine approaches has been conducted. For the last two methods, the performance is lower than the first one proposed. Full article
(This article belongs to the Special Issue Fault Diagnosis in the Internet of Things Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Schema of the proposed maintenance predictive steps.</p> Full article ">Figure 2
<p>System architecture. Using IoT sensors, the data are collected and stored for the data analysis.</p> Full article ">Figure 3
<p>Spatial representation of the set of input feature for door system 1.</p> Full article ">Figure 4
<p>Spatial representation of the set of input feature for door system 2.</p> Full article ">Figure 5
<p>Distances <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mi>D</mi> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>C</mi> <mi>s</mi> </msub> <mo>)</mo> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>b</mi> <mo>=</mo> <mi>D</mi> <mo>(</mo> <msub> <mi>C</mi> <mi>f</mi> </msub> <mo>,</mo> <msub> <mi>C</mi> <mi>s</mi> </msub> <mo>)</mo> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>d</mi> <mo>=</mo> <mi>D</mi> <mo>(</mo> <msub> <mi>C</mi> <mi>f</mi> </msub> <mo>,</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> </semantics></math> shown for a generic pattern <math display="inline"><semantics> <msub> <mi>x</mi> <mi>i</mi> </msub> </semantics></math> in door system 2.</p> Full article ">Figure 6
<p>Recall performance metric in door system 1 for the three clusters.</p> Full article ">Figure 7
<p>Precision performance metric in door system 1 for the three clusters.</p> Full article ">Figure 8
<p><math display="inline"><semantics> <msub> <mi>F</mi> <mn>1</mn> </msub> </semantics></math> score in door system 1.</p> Full article ">Figure 9
<p>Recall performance metric in door system 2 for the three clusters.</p> Full article ">Figure 10
<p>Precision performance metric in door system 2 for the three clusters.</p> Full article ">Figure 11
<p><math display="inline"><semantics> <msub> <mi>F</mi> <mn>1</mn> </msub> </semantics></math> score in door system 2.</p> Full article ">
27 pages, 1948 KiB  
Article
A Cross-Layer Approach to Analyzing Energy Consumption and Lifetime of a Wireless Sensor Node
by Fernando Ojeda, Diego Mendez, Arturo Fajardo, Maximilian Gottfried Becker and Frank Ellinger
J. Sens. Actuator Netw. 2024, 13(5), 56; https://doi.org/10.3390/jsan13050056 - 19 Sep 2024
Viewed by 3154
Abstract
Several wireless communication technologies, including Wireless Sensor Networks (WSNs), are essential for Internet of Things (IoT) applications. WSNs employ a layered framework to govern data exchanges between sender and recipient, which facilitates the establishment of rules and standards. However, in this conventional framework, [...] Read more.
Several wireless communication technologies, including Wireless Sensor Networks (WSNs), are essential for Internet of Things (IoT) applications. WSNs employ a layered framework to govern data exchanges between sender and recipient, which facilitates the establishment of rules and standards. However, in this conventional framework, network data sharing is limited to directly stacked layers, allowing manufacturers to develop proprietary protocols while impeding WSN optimization, such as energy consumption minimization, due to non-directly stacked layer effects on network performance. A Cross-Layer (CL) framework addresses implementation, modeling, and design challenges in IoT systems by allowing unrestricted data and parameter sharing between non-stacked layers. This holistic approach captures system dynamics, enabling network design optimization to address IoT network challenges. This paper introduces a novel CL modeling methodology for wireless communication systems, which is applied in two case studies to develop models for estimating energy consumption metrics, including node and network lifetime. Each case study validates the resulting model through experimental tests, demonstrating high accuracy with less than 3% error. Full article
(This article belongs to the Section Communications and Networking)
►▼ Show Figures

Figure 1

Figure 1
<p>A general architecture for an IoT system.</p> Full article ">Figure 2
<p>Layered framework and CL framework.</p> Full article ">Figure 3
<p>Block diagram of a generic node model resulting from the proposed CL modeling methodology.</p> Full article ">Figure 4
<p>Block diagram of a generic sink model resulting from the proposed CL modeling methodology.</p> Full article ">Figure 5
<p>CL node-level model.</p> Full article ">Figure 6
<p>Detailed block diagrams for different node-level models.</p> Full article ">Figure 7
<p>Experimental setup using two Zolertia Re-Motes.</p> Full article ">Figure 8
<p>Testbed implementation using two Zolertia Re-Motes.</p> Full article ">Figure 9
<p>Battery current measured from the Zolertia Re-Mote target.</p> Full article ">Figure 10
<p>Node-level lifetime estimation model and error comparison between simulation results and direct and indirect (powertrace) measurements.</p> Full article ">Figure 11
<p>CL node and network model and blocks.</p> Full article ">Figure 12
<p>Evaluated node-network level models.</p> Full article ">Figure 13
<p>Battery current measured from Zolertia Re-Mote target.</p> Full article ">Figure 14
<p>Network-level lifetime estimation model and error comparison between simulation results and direct and indirect (powertrace) measurements.</p> Full article ">
17 pages, 2339 KiB  
Article
Efficient Channel Estimation in OFDM Systems Using a Fast Super-Resolution CNN Model
by Sunita Khichar, Wiroonsak Santipach, Lunchakorn Wuttisittikulkij, Amir Parnianifard and Sushank Chaudhary
J. Sens. Actuator Netw. 2024, 13(5), 55; https://doi.org/10.3390/jsan13050055 - 5 Sep 2024
Cited by 1 | Viewed by 1031
Abstract
Channel estimation is a critical component in orthogonal frequency division multiplexing (OFDM) systems for ensuring reliable wireless communication. In this study, we propose a fast super-resolution convolutional neural network (FSRCNN) model for channel estimation, designed to reduce computational complexity while maintaining high estimation [...] Read more.
Channel estimation is a critical component in orthogonal frequency division multiplexing (OFDM) systems for ensuring reliable wireless communication. In this study, we propose a fast super-resolution convolutional neural network (FSRCNN) model for channel estimation, designed to reduce computational complexity while maintaining high estimation accuracy. The proposed FSRCNN model incorporates modifications such as replacing linear interpolation with zero padding and leveraging a new fast CNN architecture to estimate channel coefficients. Our numerical experiments and simulations demonstrate that the FSRCNN model significantly outperforms traditional methods, such as least square (LS) and linear minimum mean square error (LMMSE), in terms of mean square error (MSE) across various signal-to-noise ratios (SNRs). Specifically, the FSRCNN model achieves MSE values comparable to MMSE estimation, particularly at higher SNRs, while maintaining lower computational complexity. At an SNR of 20 dB, the FSRCNN model shows a notable improvement in MSE performance compared to the ChannelNet and LS methods. The proposed model also demonstrates robust performance across different SNR levels, with optimal results observed when the training SNR is close to the operating SNR. These findings validate the effectiveness of the FSRCNN model in providing a low-complexity, high-accuracy alternative for channel estimation, making it suitable for real-time applications and devices with limited computational resources. This advancement holds significant promise for enhancing the reliability and efficiency of current and future wireless communication networks. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>FSRCNN layers for the proposed estimator.</p> Full article ">Figure 2
<p>Our proposed channel estimation (<b>a</b>) Detailed process of channel estimation (<b>b</b>) A simplified example for channel estimation.</p> Full article ">Figure 3
<p>Execution time comparison for different CNN structures.</p> Full article ">Figure 4
<p>MSE of various channel estimation methods for VehA channel model (<b>a</b>) MSE comparison of FSRCNN (zero padding) and SRCNN (zero padding) only (<b>b</b>) MSE comparison of Proposed model with state of the art and other tradition techniques.</p> Full article ">Figure 4 Cont.
<p>MSE of various channel estimation methods for VehA channel model (<b>a</b>) MSE comparison of FSRCNN (zero padding) and SRCNN (zero padding) only (<b>b</b>) MSE comparison of Proposed model with state of the art and other tradition techniques.</p> Full article ">Figure 5
<p>The real parts of the estimates of the channel response from various estimation methods are compared with the actual response shown in solid blue lines. The channel response and the estimates are plotted for all subcarriers and 4 time slots as indicated.</p> Full article ">Figure 6
<p>Effect of different pilot lengths on MSE of channel estimation.</p> Full article ">
29 pages, 7003 KiB  
Article
Optimal Signal Wavelengths for Underwater Optical Wireless Communication under Sunlight in Stratified Waters
by Tharuka Govinda Waduge, Boon-Chong Seet and Kay Vopel
J. Sens. Actuator Netw. 2024, 13(5), 54; https://doi.org/10.3390/jsan13050054 - 4 Sep 2024
Viewed by 1122
Abstract
Underwater optical wireless communication (UOWC) is a field of research that has gained popularity with the development of unmanned underwater vehicle (UUV) technologies. Its utilization is crucial in offshore industries engaging in sustainable alternatives for food production and energy security. Although UOWC can [...] Read more.
Underwater optical wireless communication (UOWC) is a field of research that has gained popularity with the development of unmanned underwater vehicle (UUV) technologies. Its utilization is crucial in offshore industries engaging in sustainable alternatives for food production and energy security. Although UOWC can meet the high data rate and low latency requirements of underwater video transmission for UUV operations, the links that enable such communication are affected by the inhomogeneous light attenuation and the presence of sunlight. Here, we present how the underwater spectral distribution of the light field can be modeled along the depths of eight stratified oceanic water types. We considered other established models, such as SPCTRL2, Haltrin’s single parameter model for inherent optical properties, and a model for the estimation of the depth distribution of chlorophyll-a, and present insights based on transmission wavelength for the maximum signal-to-noise ratio (SNR) under different optical link parameter combinations such as beam divergence and transmit power under “daytime” and “nighttime” conditions. The results seem to challenge the common notion that the blue-green spectrum is the most suitable for underwater optical communication. We highlight a unique relationship between the transmission wavelength for the optimal SNR and the link parameters and distance, which varies with depth depending on the type of oceanic water stratification. Our analyses further highlighted potential implications for solar discriminatory approaches and strategies for routing in cooperative optical wireless networks in the photic region. Full article
(This article belongs to the Section Communications and Networking)
►▼ Show Figures

Figure 1

Figure 1
<p>A conceptualized offshore wireless optical network incorporating underwater optical wireless communication and free-space optical (FSO) links.</p> Full article ">Figure 2
<p>The key components of an underwater wireless optical communication system.</p> Full article ">Figure 3
<p>Spectral distribution of attenuation coefficient in Jerlov water types [<a href="#B36-jsan-13-00054" class="html-bibr">36</a>].</p> Full article ">Figure 4
<p>Spectral distribution of global irradiance generated for a clear summer day in Auckland, New Zealand, using SPCTRL2 on 7 December 2023 at 11:30 am; coordinates, 36.8509 S, 174.7645 E; 2.2 cm precipitable water vapor, 1013 mb atmospheric pressure, and 300 DU ozone coverage [<a href="#B52-jsan-13-00054" class="html-bibr">52</a>].</p> Full article ">Figure 5
<p>Reconstructed depth distribution of Chl-<span class="html-italic">a</span> concentration profiles for S1–S9.</p> Full article ">Figure 6
<p>The system model.</p> Full article ">Figure 7
<p>Change in attenuation coefficient with depth and wavelength in (<b>a</b>) clear ocean profile S2 and (<b>b</b>) coastal profile S8. Amber is preferred at the surface, shifting to blue-green with depth in S8.</p> Full article ">Figure 8
<p>(<b>a</b>) A vertical UOWC link where the receiver receives both signal and sunlight; the O-SNR is estimated for increasing link distance. The link is established for a vertical distance <math display="inline"><semantics> <mrow> <mi>d</mi> </mrow> </semantics></math> from a transmitter depth of <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>z</mi> </mrow> <mrow> <mn>0</mn> </mrow> </msub> </mrow> </semantics></math> to the receiver at a depth of <math display="inline"><semantics> <mrow> <mi>z</mi> </mrow> </semantics></math>; (<b>b</b>) a horizontal UOWC link where the receiver receives both the signal and horizontally diffused sunlight; the O-SNR is estimated for increasing link distance. The link is established for a horizontal distance <math display="inline"><semantics> <mrow> <mi>d</mi> </mrow> </semantics></math> at a transmitter and receiver depth of <math display="inline"><semantics> <mrow> <mi>z</mi> </mrow> </semantics></math>.</p> Full article ">Figure 9
<p>Downwelling solar spectral irradiance <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>E</mi> </mrow> <mrow> <mi>t</mi> </mrow> </msub> <mo> </mo> <mo>(</mo> <mi>λ</mi> <mo>,</mo> <mi>z</mi> <mo>)</mo> </mrow> </semantics></math> for profiles S1 (<b>a</b>) and S8 (<b>d</b>), and the resultant background power <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>P</mi> </mrow> <mrow> <mi>B</mi> <mi>G</mi> </mrow> </msub> <mo>(</mo> <mi>λ</mi> <mo>,</mo> <mi>z</mi> <mo>)</mo> </mrow> </semantics></math> for corresponding water profiles in the horizontal (<b>b</b>,<b>e</b>) and in the vertical (<b>c</b>,<b>f</b>) receiver directions, respectively.</p> Full article ">Figure 10
<p>Wavelength of best O-SNR for D-LOS UOWC at nighttime for profiles S1 (<b>a</b>–<b>d</b>), S4 (<b>e</b>–<b>h</b>), S5 (<b>i</b>–<b>l</b>), and S8 (<b>m</b>–<b>p</b>) with beam divergence and power parameters: (θ = 60°, 1 W) column 1; (θ = 60°, 10 W) column 2; (θ = 120°, 1 W) column 3; and (θ = 120°, 10 W) column 4. The magenta line shows the 0 dB distance, and the color beneath it is the wavelength that achieves it.</p> Full article ">Figure 11
<p>Wavelength of best O-SNR for P2P-LOS UOWC at nighttime for profiles S1 (<b>a</b>,<b>e</b>), S4 (<b>b</b>,<b>f</b>), S5 (<b>c</b>,<b>g</b>), and S8 (<b>d</b>,<b>h</b>) up to first 350 m with emitter powers of 1 W in the first row up and 10 W in the second row. The magenta line shows the 0 dB distance, and the color beneath it is the wavelength that achieves it.</p> Full article ">Figure 12
<p>Wavelength of best O-SNR for D-LOS (top) UOWC in daytime for profiles S1 (<b>a</b>–<b>d</b>), S4 (<b>e</b>–<b>h</b>), S5 (<b>i</b>–<b>l</b>), and S8 (<b>m</b>–<b>p</b>) with beam divergence and power parameters: (θ = 60°, 1 W) column 1; (θ = 60°, 10 W) column 2; (θ = 120°, 1 W) column 3; and (θ = 120°, 10 W) column 4. P2P-LOS results are at the bottom for profiles S1, S4, S5, and S8 (left-to-right) with a beam divergence of 5°, and power parameters: 1 W (top row, <b>q</b>–<b>t</b>) and 10 W (bottom row, <b>u</b>–<b>x</b>). The magenta line shows the 0 dB distance, and the color beneath it is the wavelength that achieves it for both sets.</p> Full article ">Figure 13
<p>The 0 dB link distances at each wavelength for D-LOS (top) UOWC in daytime for profiles S1 (<b>a</b>–<b>d</b>), S4 (<b>e</b>–<b>h</b>), S5 (<b>i</b>–<b>l</b>), and S8 (<b>m</b>–<b>p</b>) with beam divergence and power parameters: (θ = 60°, 1 W) column 1; (θ = 60°, 10 W) column 2; (θ = 120°, 1 W) column 3; and (θ = 120°, 10 W) column 4. P2P-LOS results are at the bottom for profiles S1, S4, S5, and S8 (left-to-right) with a beam divergence of 5° and power parameters: 1 W (top row, <b>q</b>–<b>t</b>) and 10 W (bottom row, <b>u</b>–<b>x</b>).</p> Full article ">Figure 14
<p>Wavelength of best O-SNR for downlink D-LOS UOWC for profiles S1 (<b>a</b>–<b>d</b>), S4 (<b>e</b>–<b>h</b>), S5 (<b>i</b>–<b>l</b>), and S8 (<b>m</b>–<b>p</b>) for full depth and link distance with emitter beam divergence and power parameters: (θ = 60°, 1 W) column 1; (θ = 60°, 10 W) column 2; (θ = 120°, 1 W) column 3; and (θ = 120°, 10 W) column 4.</p> Full article ">Figure 15
<p>Wavelength of best O-SNR for downlink P2P-LOS UOWC for profiles S1, S4, S5, and S8 (left-to-right) for full depth and link distance with emitter beam divergence of 5° and power parameters: 1 W (top row, <b>a</b>–<b>d</b>) and 10 W (bottom row, <b>e</b>–<b>h</b>).</p> Full article ">
33 pages, 6102 KiB  
Review
Machine-Learning- and Internet-of-Things-Driven Techniques for Monitoring Tool Wear in Machining Process: A Comprehensive Review
by Sudhan Kasiviswanathan, Sakthivel Gnanasekaran, Mohanraj Thangamuthu and Jegadeeshwaran Rakkiyannan
J. Sens. Actuator Netw. 2024, 13(5), 53; https://doi.org/10.3390/jsan13050053 - 4 Sep 2024
Viewed by 1292
Abstract
Tool condition monitoring (TCM) systems have evolved into an essential requirement for contemporary manufacturing sectors of Industry 4.0. These systems employ sensors and diverse monitoring techniques to swiftly identify and diagnose tool wear, defects, and malfunctions of computer numerical control (CNC) machines. Their [...] Read more.
Tool condition monitoring (TCM) systems have evolved into an essential requirement for contemporary manufacturing sectors of Industry 4.0. These systems employ sensors and diverse monitoring techniques to swiftly identify and diagnose tool wear, defects, and malfunctions of computer numerical control (CNC) machines. Their pivotal role lies in augmenting tool lifespan, minimizing machine downtime, and elevating productivity, thereby contributing to industry growth. However, the efficacy of CNC machine TCM hinges upon multiple factors, encompassing system type, data precision, reliability, and adeptness in data analysis. Globally, extensive research is underway to enhance real-time TCM system efficiency. This review focuses on the significance and attributes of proficient real-time TCM systems of CNC turning centers. It underscores TCM’s paramount role in manufacturing and outlines the challenges linked to TCM data processing and analysis. Moreover, the review elucidates various TCM system variants, including cutting force, acoustic emission, vibration, and temperature monitoring systems. Furthermore, the integration of industrial Internet of things (IIoT) and machine learning (ML) into CNC machine TCM systems are also explored. This article concludes by underscoring the ongoing necessity for research and development in TCM technology to empower modern intelligent industries to operate at peak efficiency. Full article
(This article belongs to the Special Issue AI and IoT Convergence for Sustainable Smart Manufacturing)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic diagram of TCM system stages.</p> Full article ">Figure 2
<p>Generalized TCM model.</p> Full article ">Figure 3
<p>Common TCM sensing and data collection techniques.</p> Full article ">Figure 4
<p>Process flow of the TCM topology with ML classifiers.</p> Full article ">Figure 5
<p>ML and DL structure comparison for TCM.</p> Full article ">Figure 6
<p>Layers of IIoT.</p> Full article ">Figure 7
<p>Schematic view of the intelligent condition monitoring system.</p> Full article ">Figure 8
<p>Industrial IoT structure for CNC real-time tool condition monitoring.</p> Full article ">Figure 9
<p>IoT system for online TCM.</p> Full article ">
27 pages, 8451 KiB  
Article
A Proof-of-Concept Study of Stability Monitoring of Implant Structure by Deep Learning of Local Vibrational Characteristics
by Manh-Hung Tran, Nhat-Duc Hoang, Jeong-Tae Kim, Hoang-Khanh Le, Ngoc-Loi Dang, Ngoc-Tuong-Vy Phan, Duc-Duy Ho and Thanh-Canh Huynh
J. Sens. Actuator Netw. 2024, 13(5), 52; https://doi.org/10.3390/jsan13050052 - 3 Sep 2024
Viewed by 824
Abstract
This study develops a structural stability monitoring method for an implant structure (i.e., a single-tooth dental implant) through deep learning of local vibrational modes. Firstly, the local vibrations of the implant structure are identified from the conductance spectrum, achieved by driving the structure [...] Read more.
This study develops a structural stability monitoring method for an implant structure (i.e., a single-tooth dental implant) through deep learning of local vibrational modes. Firstly, the local vibrations of the implant structure are identified from the conductance spectrum, achieved by driving the structure using a piezoelectric transducer within a pre-defined high-frequency band. Secondly, deep learning models based on a convolutional neural network (CNN) are designed to process the obtained conductance data of local vibrational modes. Thirdly, the CNN models are trained to autonomously extract optimal vibration features for structural stability assessment of the implant structure. We employ a validated predictive 3D numerical modeling approach to demonstrate the feasibility of the proposed approach. The proposed method achieved promising results for predicting material loss surrounding the implant, with the best CNN model demonstrating training and testing errors of 3.7% and 4.0%, respectively. The implementation of deep learning allows optimal feature extraction in a lower frequency band, facilitating the use of low-cost active sensing devices. This research introduces a novel approach for assessing the implant’s stability, offering promise for developing future radiation-free stability assessment tools. Full article
(This article belongs to the Section Actuators, Sensors and Devices)
►▼ Show Figures

Figure 1

Figure 1
<p>The proposed MBL method integrating the PZT-enabled conductance sensing technique with the 1D CNN algorithm.</p> Full article ">Figure 2
<p>The PZT–implant interaction and 1-dof simplified EMI model.</p> Full article ">Figure 3
<p>The architecture of 1D CNN-based bone-loss prediction models for implants. <span class="html-italic">x</span><sub>i</sub>: input: <span class="html-italic">y</span>: output, <span class="html-italic">Conv</span>: convolutional layer, <span class="html-italic">Batchnorm</span>: batch normalization layer, <span class="html-italic">ReLU</span>: rectified linear unit layer, <span class="html-italic">Maxpool</span>: maxpooling layer, <span class="html-italic">FC</span>: fully connected (FC) layer.</p> Full article ">Figure 4
<p>(<b>a</b>) The experimental test of a PZT-free beam structure; (<b>b</b>) FE modeling.</p> Full article ">Figure 5
<p>Validation of the FE modeling approach (<b>a</b>) Simulated conductance vs. experimental result; (<b>b</b>) simulated susceptance vs. experimental result; (<b>c</b>) simulated peak frequency vs. experimental result; (<b>d</b>) simulation error of the peak frequencies; (<b>e</b>) electro-mechanical interactions between the PZT transducer and the free beam at different resonances.</p> Full article ">Figure 6
<p>FE modeling of the PZT–implant–bone system: (<b>a</b>) the 3D numerical model; (<b>b</b>) the boundary conditions; (<b>c</b>) the meshed FE model.</p> Full article ">Figure 7
<p>The conductance responses: (<b>a</b>) the free PZT–implant–bone system; (<b>b</b>) the PZT–implant system.</p> Full article ">Figure 8
<p>The resonant vibration modes of (<b>a</b>) the PZT–implant–bone system (FEM1) and (<b>b</b>) the PZT–implant system (FEM2).</p> Full article ">Figure 9
<p>Simulation of marginal bone loss (MBL) cases.</p> Full article ">Figure 10
<p>The MBL-induced changes in the conductance spectrum: (<b>a</b>) the frequency band of 50 kHz–200 kHz; (<b>b</b>) the frequency band of 56 kHz–60 kHz; (<b>c</b>) the frequency band of 140 kHz–160 kHz.</p> Full article ">Figure 11
<p>Statistical metric-based MBL monitoring results: (<b>a</b>) RMSD metric and (<b>b</b>) CCD metric using the frequency range of 50 kHz–200 kHz; (<b>c</b>) RMSD metric and (<b>d</b>) CCD metric using the six sub-ranges.</p> Full article ">Figure 12
<p>The conductance signals of typical bone-loss cases before and after adding white noise: (<b>a</b>) original signals; (<b>b</b>) noisy signals.</p> Full article ">Figure 13
<p>Training of the four 1D CNN models: (<b>a</b>) training loss; (<b>b</b>) training RMSE.</p> Full article ">Figure 14
<p>The performance evaluation of (<b>a</b>) Model 1, (<b>b</b>) Model 2, (<b>c</b>) Model 3, and (<b>d</b>) Model 4.</p> Full article ">Figure 15
<p>Comparison of the four 1D CNN models.</p> Full article ">Figure 16
<p>The MBL assessment results of Model 3 (the best 1D CNN model) when evaluated on the testing dataset.</p> Full article ">
20 pages, 1563 KiB  
Article
Energy-Efficient Handover Algorithm for Sustainable Mobile Networks: Balancing Connectivity and Power Consumption
by Radhwan M. Abdullah, Ibrahim Al-Surmi, Gamil R. S. Qaid and Ali A. Alwan
J. Sens. Actuator Netw. 2024, 13(5), 51; https://doi.org/10.3390/jsan13050051 - 2 Sep 2024
Viewed by 944
Abstract
In the era of pervasive mobile and heterogeneous networks, maintaining seamless connectivity during handover events while minimizing energy consumption is paramount. Traditional handover mechanisms prioritize metrics such as signal strength, user mobility, and network load, often neglecting the critical aspect of energy consumption. [...] Read more.
In the era of pervasive mobile and heterogeneous networks, maintaining seamless connectivity during handover events while minimizing energy consumption is paramount. Traditional handover mechanisms prioritize metrics such as signal strength, user mobility, and network load, often neglecting the critical aspect of energy consumption. This study presents a novel approach to handover decision-making in mobile networks by incorporating energy-related metrics, such as battery level, energy consumption rate, and environmental context, to make informed handover decisions that balance connectivity quality and energy efficiency. Unlike traditional methods that primarily focus on signal strength and network load, our approach addresses the critical need for energy efficiency, particularly in high-mobility scenarios. This innovative framework not only enhances connectivity but also significantly improves power consumption management, offering a more sustainable solution for modern mobile networks. Through extensive simulations, we demonstrate the effectiveness of our proposed solution in reducing energy usage without compromising network performance. The results reveal significant improvements in energy savings for mobile devices, especially under high-mobility scenarios and varying network conditions. By prioritizing energy-efficient handovers, our approach not only extends the battery life of mobile devices but also contributes to the overall sustainability of mobile networks. This paper underscores the importance of incorporating energy metrics into handover decisions and sets the stage for future research in energy-aware network management. Full article
(This article belongs to the Section Network Services and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Aggregate energy consumed.</p> Full article ">Figure 2
<p>Comparison of successful handovers.</p> Full article ">Figure 3
<p>QoS latency compliance.</p> Full article ">Figure 4
<p>Comparison of QoS bandwidth maintenance.</p> Full article ">
26 pages, 3455 KiB  
Article
Energy-Efficient Online Path Planning for Internet of Drones Using Reinforcement Learning
by Zainab AlMania, Tarek Sheltami, Gamil Ahmed, Ashraf Mahmoud and Abdulaziz Barnawi
J. Sens. Actuator Netw. 2024, 13(5), 50; https://doi.org/10.3390/jsan13050050 - 29 Aug 2024
Viewed by 845
Abstract
Unmanned aerial vehicles (UAVs) have recently been applied in several contexts due to their flexibility, mobility, and fast deployment. One of the essential aspects of multi-UAV systems is path planning, which autonomously determines paths for drones from starting points to destination points. However, [...] Read more.
Unmanned aerial vehicles (UAVs) have recently been applied in several contexts due to their flexibility, mobility, and fast deployment. One of the essential aspects of multi-UAV systems is path planning, which autonomously determines paths for drones from starting points to destination points. However, UAVs face many obstacles in their routes, potentially causing loss or damage. Several heuristic approaches have been investigated to address collision avoidance. These approaches are generally applied in static environments where the environment is known in advance and paths are generated offline, making them unsuitable for unknown or dynamic environments. Additionally, limited flight times due to battery constraints pose another challenge in multi-UAV path planning. Reinforcement learning (RL) emerges as a promising candidate to generate collision-free paths for drones in dynamic environments due to its adaptability and generalization capabilities. In this study, we propose a framework to provide a novel solution for multi-UAV path planning in a 3D dynamic environment. The improved particle swarm optimization with reinforcement learning (IPSO-RL) framework is designed to tackle the multi-UAV path planning problem in a fully distributed and reactive manner. The framework integrates IPSO with deep RL to provide the drone with additional feedback and guidance to operate more sustainably. This integration incorporates a unique reward system that can adapt to various environments. Simulations demonstrate the effectiveness of the IPSO-RL approach, showing superior results in terms of collision avoidance, path length, and energy efficiency compared to other benchmarks. The results also illustrate that the proposed IPSO-RL framework can acquire a feasible and effective route successfully with minimum energy consumption in complicated environments. Full article
(This article belongs to the Topic Machine Learning in Communication Systems and Networks, 2nd Edition)
►▼ Show Figures

Figure 1

Figure 1
<p>Random function compared with logistic map [<a href="#B42-jsan-13-00050" class="html-bibr">42</a>].</p> Full article ">Figure 2
<p>Proposed IPSO-DRL system model.</p> Full article ">Figure 3
<p>Flowchart of IPSO-RL framework.</p> Full article ">Figure 4
<p>Cumulative reward with respect to the number of episodes for different values of <math display="inline"><semantics> <mi>α</mi> </semantics></math>, <math display="inline"><semantics> <mi>γ</mi> </semantics></math>, and <math display="inline"><semantics> <mi>ϵ</mi> </semantics></math>.</p> Full article ">Figure 5
<p>Scale where the improvement in the learning process occurs for IPSO-DRL with the energy model.</p> Full article ">Figure 6
<p>Reward curve for IPSO-RL with the energy model.</p> Full article ">Figure 7
<p>Energy consumption during the learning process.</p> Full article ">Figure 8
<p>Scale where the improvement in the learning process occurs for IPSO-RL without the energy model.</p> Full article ">Figure 9
<p>Reward curve for IPSO-RL without the energy model.</p> Full article ">Figure 10
<p>Moving average curve for IPSO with the energy model.</p> Full article ">Figure 11
<p>Moving average curve for IPSO with the energy model.</p> Full article ">Figure 12
<p>Reward curves for IPSO-DRL.</p> Full article ">Figure 13
<p>Reward curve for the recent study.</p> Full article ">
23 pages, 16203 KiB  
Article
Predictive Models for Aggregate Available Capacity Prediction in Vehicle-to-Grid Applications
by Luca Patanè, Francesca Sapuppo, Giuseppe Napoli and Maria Gabriella Xibilia
J. Sens. Actuator Netw. 2024, 13(5), 49; https://doi.org/10.3390/jsan13050049 - 27 Aug 2024
Viewed by 851
Abstract
The integration of vehicle-to-grid (V2G) technology into smart energy management systems represents a significant advancement in the field of energy suppliers for Industry 4.0. V2G systems enable a bidirectional flow of energy between electric vehicles and the power grid and can provide ancillary [...] Read more.
The integration of vehicle-to-grid (V2G) technology into smart energy management systems represents a significant advancement in the field of energy suppliers for Industry 4.0. V2G systems enable a bidirectional flow of energy between electric vehicles and the power grid and can provide ancillary services to the grid, such as peak shaving, load balancing, and emergency power supply during power outages, grid faults, or periods of high demand. In this context, reliable prediction of the availability of V2G as an energy source in the grid is fundamental in order to optimize both grid stability and economic returns. This requires both an accurate modeling framework that includes the integration and pre-processing of readily accessible data and a prediction phase over different time horizons for the provision of different time-scale ancillary services. In this research, we propose and compare two data-driven predictive modeling approaches to demonstrate their suitability for dealing with quasi-periodic time series, including those dealing with mobility data, meteorological and calendrical information, and renewable energy generation. These approaches utilize publicly available vehicle tracking data within the floating car data paradigm, information about meteorological conditions, and fuzzy weekend and holiday information to predict the available aggregate capacity with high precision over different time horizons. Two data-driven predictive modeling approaches are then applied to the selected data, and the performance is compared. The first approach is Hankel dynamic mode decomposition with control (HDMDc), a linear state-space representation technique, and the second is long short-term memory (LSTM), a deep learning method based on recurrent nonlinear neural networks. In particular, HDMDc performs well on predictions up to a time horizon of 4 h, demonstrating its effectiveness in capturing global dynamics over an entire year of data, including weekends, holidays, and different meteorological conditions. This capability, along with its state-space representation, enables the extraction of relationships among exogenous inputs and target variables. Consequently, HDMDc is applicable to V2G integration in complex environments such as smart grids, which include various energy suppliers, renewable energy sources, buildings, and mobility data. Full article
(This article belongs to the Special Issue Fault Diagnosis in the Internet of Things Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>HDMDc block diagram.</p> Full article ">Figure 2
<p>Sequence-to-sequence LSTM block diagram.</p> Full article ">Figure 3
<p>AAC prediction model framework.</p> Full article ">Figure 4
<p>Stop maps in different time intervals of the day integrated over the entire data time interval. The color bar shows the duration of the stops. The stop events started in the following time windows: (<b>a</b>) from 0 a.m. to 6 a.m., (<b>b</b>) from 6 a.m. to 12 a.m., (<b>c</b>) from 12 a.m. to 6 p.m., and (<b>d</b>) from 6 p.m. to 12 p.m.</p> Full article ">Figure 5
<p>Selection of the aggregation hub in the Ann Arbor Area: satellite view of <math display="inline"><semantics> <mrow> <mi>H</mi> <mi>u</mi> <msub> <mi>b</mi> <mn>1</mn> </msub> </mrow> </semantics></math> area in the city center and university zone.</p> Full article ">Figure 6
<p>Membership function for the fuzzification of the holiday rate: (<b>a</b>) weekend membership; (<b>b</b>) national holiday membership functions.</p> Full article ">Figure 7
<p>Exogenous inputs: left y-axis—holiday rate (blue); right y-axis—precipitation in mm (red), temperature in °C (green), and wind speed in km/h (magenta). (<b>a</b>) Selection of a training set week (Wednesday 3 to Tuesday 10 October 2018). (<b>b</b>) Selection of a test set week (Wednesday 17 to Tuesday 24 January 2018).</p> Full article ">Figure 8
<p>HDMDc time series prediction with different time horizons: 1 h, 2 h, 3 h, and 4 h. (<b>a</b>) Selection of a training set week (Wednesday 3 to Tuesday 10 October 2018). (<b>b</b>) Selection of a test set week (Wednesday 17 to Tuesday 24 January 2018).</p> Full article ">Figure 9
<p>HDMDc regression plots for prediction with different time horizons of the test dataset: (<b>a</b>) 1 h, (<b>b</b>) 2 h, (<b>c</b>) 3 h, and (<b>d</b>) 4 h.</p> Full article ">Figure 10
<p>LSTM time series prediction with different time horizons: 1 h, 2 h, 3 h, and 4 h. (<b>a</b>) Selection of a training set week (Wednesday 3 to Tuesday 10 October 2018). (<b>b</b>) Selection of a test set week (Wednesday 17 to Tuesday 24 January 2018).</p> Full article ">
26 pages, 3648 KiB  
Article
Classifying the Cognitive Performance of Drivers While Talking on Hands-Free Mobile Phone Based on Innovative Sensors and Intelligent Approach
by Boniface Ndubuisi Ossai, Mhd Saeed Sharif, Cynthia Fu, Jijomon Chettuthara Moncy, Arya Murali and Fahad Alblehai
J. Sens. Actuator Netw. 2024, 13(5), 48; https://doi.org/10.3390/jsan13050048 - 25 Aug 2024
Viewed by 1228
Abstract
The use of mobile phones while driving is restricted to hands-free mode. But even in the hands-free mode, the use of mobile phones while driving causes cognitive distraction due to the diverted attention of the driver. By employing innovative machine-learning approaches to drivers’ [...] Read more.
The use of mobile phones while driving is restricted to hands-free mode. But even in the hands-free mode, the use of mobile phones while driving causes cognitive distraction due to the diverted attention of the driver. By employing innovative machine-learning approaches to drivers’ physiological signals, namely electroencephalogram (EEG), heart rate (HR), and blood pressure (BP), the impact of talking on hands-free mobile phones in real time has been investigated in this study. The cognitive impact was measured using EEG, HR, and BP data. The authors developed an intelligent model that classified the cognitive performance of drivers using physiological signals that were measured while drivers were driving and reverse bay parking in real time and talking on hands-free mobile phones, considering all driver ages as a complete cohort. Participants completed two numerical tasks varying in difficulty while driving and reverse bay parking. The results show that when participants did the hard tasks, their theta and lower alpha EEG frequency bands increased and exceeded those when they did the easy tasks. The results also show that the BP and HR under phone condition were higher than the BP and HR under no-phone condition. Participants’ cognitive performance was classified using a feedforward neural network, and 97% accuracy was achieved. According to qualitative results, participants experienced significant cognitive impacts during the task completion. Full article
(This article belongs to the Special Issue Interpretable Strategies for Secure Vehicle Road Collaboration and Threat Tracing)
►▼ Show Figures

Figure 1

Figure 1
<p>Experimental testing site entrance.</p> Full article ">Figure 2
<p>Experimental testing site car park.</p> Full article ">Figure 3
<p>Flowchart of driver’s cognitive performance.</p> Full article ">Figure 4
<p>Block diagram of the project.</p> Full article ">Figure 5
<p>Data points representing age distribution of the participants.</p> Full article ">Figure 6
<p>Relationship between data points and theta easy.</p> Full article ">Figure 7
<p>Relationship between data points and theta hard.</p> Full article ">Figure 8
<p>Relationship between data points and lower alpha easy.</p> Full article ">Figure 9
<p>Relationship between data points and lower alpha hard.</p> Full article ">Figure 10
<p>Distribution of self-reported cognitive load.</p> Full article ">Figure 11
<p>ROC curve of ANN.</p> Full article ">Figure 12
<p>ANN model accuracy against epoch.</p> Full article ">
29 pages, 7057 KiB  
Review
A Review of Multi-UAV Task Allocation Algorithms for a Search and Rescue Scenario
by Sajjad A. Ghauri, Mubashar Sarfraz, Rahim Ali Qamar, Muhammad Farhan Sohail and Sheraz Alam Khan
J. Sens. Actuator Netw. 2024, 13(5), 47; https://doi.org/10.3390/jsan13050047 - 23 Aug 2024
Viewed by 1130
Abstract
Unmanned aerial vehicles (UAVs) play a crucial role in enhancing search and rescue (SAR) operations by accessing inaccessible areas, accomplishing challenging tasks, and providing real-time monitoring and modeling in situations where human presence is unsafe. Multi-UAVs can collaborate more efficiently and cost-effectively than [...] Read more.
Unmanned aerial vehicles (UAVs) play a crucial role in enhancing search and rescue (SAR) operations by accessing inaccessible areas, accomplishing challenging tasks, and providing real-time monitoring and modeling in situations where human presence is unsafe. Multi-UAVs can collaborate more efficiently and cost-effectively than a single large UAV for performing SAR operations. In multi-UAV systems, task allocation (TA) is a critical and complex process involving cooperative decision making and control to minimize the time and energy consumption of UAVs for task completion. This paper offers an exhaustive review of both static and dynamic TA algorithms, confidently assessing their strengths, weaknesses, and limitations. It provides valuable insights into addressing research questions related to specific UAV operations in SAR. The paper rigorously discusses outstanding issues and challenges and confidently presents potential directions for the future development of task assignment algorithms. Finally, it confidently highlights the challenges of multi-UAV dynamic TA methods for SAR. This work is crucial for gaining a comprehensive understanding of multi-UAV dynamic TA algorithms and confidently emphasizes critical open issues and research gaps for future SAR research and development, ensuring that readers feel informed and knowledgeable. Full article
(This article belongs to the Section Communications and Networking)
►▼ Show Figures

Figure 1

Figure 1
<p>A Comparison of single UAV and multi-UAV systems performing SAR operations.</p> Full article ">Figure 2
<p>Organization structure of the research article.</p> Full article ">Figure 3
<p>Multi-UAV task allocation: static vs. dynamic environments: (<b>a</b>) 12 tasks are assigned to 6 UAVs, where 2 tasks (<math display="inline"><semantics> <msub> <mi>t</mi> <mn>13</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>t</mi> <mn>14</mn> </msub> </semantics></math>) arrived after task allocation and during the task execution phase; (<b>b</b>) UAVs are reallocating tasks to accommodate new tasks <math display="inline"><semantics> <msub> <mi>t</mi> <mn>13</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>t</mi> <mn>14</mn> </msub> </semantics></math>, assigning them to UAV1 and UAV3.</p> Full article ">Figure 3 Cont.
<p>Multi-UAV task allocation: static vs. dynamic environments: (<b>a</b>) 12 tasks are assigned to 6 UAVs, where 2 tasks (<math display="inline"><semantics> <msub> <mi>t</mi> <mn>13</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>t</mi> <mn>14</mn> </msub> </semantics></math>) arrived after task allocation and during the task execution phase; (<b>b</b>) UAVs are reallocating tasks to accommodate new tasks <math display="inline"><semantics> <msub> <mi>t</mi> <mn>13</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>t</mi> <mn>14</mn> </msub> </semantics></math>, assigning them to UAV1 and UAV3.</p> Full article ">
24 pages, 2920 KiB  
Article
Opportunistic Interference Alignment in Cognitive Radio Networks with Space–Time Coding
by Yusuf Abdulkadir, Oluyomi Simpson and Yichuang Sun
J. Sens. Actuator Netw. 2024, 13(5), 46; https://doi.org/10.3390/jsan13050046 - 23 Aug 2024
Viewed by 569
Abstract
For a multiuser multiple-input–multiple-output (MIMO) overlay cognitive radio (CR) network, an opportunistic interference alignment (IA) technique has been proposed that allows spectrum sharing between primary users (PUs) and secondary users (SUs) while ensuring zero interference to the PU. The CR system consists of [...] Read more.
For a multiuser multiple-input–multiple-output (MIMO) overlay cognitive radio (CR) network, an opportunistic interference alignment (IA) technique has been proposed that allows spectrum sharing between primary users (PUs) and secondary users (SUs) while ensuring zero interference to the PU. The CR system consists of one PU and K SUs where the PU uses space-time water-filling (ST-WF) algorithm to optimize its transmission and in the process, frees up unused eigenmodes that can be exploited by the SU. The SUs make use of an optimal power allocation algorithm to align their transmitted signals in such a way their interference impairs only the PUs unused eigenmodes. Since the SUs optimal power allocation algorithm turns out to be an optimal beamformer with multiple eigen-beams, this work initially proposes combining the diversity gain property of space-time block codes, the zero-forcing function of IA and beamforming to optimize the SUs transmission rates. This proposed solution requires availability of channel state information (CSI), and to eliminate the need for CSI, this work then combines Differential Space-Time Block Coding (DSTBC) scheme with optimal IA precoders (consisting of beamforming and zero-forcing) to maximize the SUs data rates. Simulation results confirm the accuracy of the proposed solution. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Multiuser CR network model consisting of one PU link and multiple SUs.</p> Full article ">Figure 2
<p>Average sum rate vs. the SNR at the PUs link for water-filling (SWF and ST-WF) and MEB.</p> Full article ">Figure 3
<p>Outage probability curves for SWF and ST-WF.</p> Full article ">Figure 4
<p>(<b>a</b>) Single detection (<b>b</b>) Double detection.</p> Full article ">Figure 5
<p>Performance comparison of a conventional ED and a double-threshold ED scheme.</p> Full article ">Figure 6
<p>P<sub>d</sub> vs. SNR with P<sub>f</sub> = <math display="inline"><semantics> <mrow> <mn>0.1</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mn>0.01</mn> </mrow> </semantics></math> using a conventional ED and a double-threshold ED with an <math display="inline"><semantics> <mrow> <mi>R</mi> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math> scheme.</p> Full article ">Figure 7
<p>STBC process.</p> Full article ">Figure 8
<p>SER Curves for coherent STBC and DSTBC–beamforming schemes.</p> Full article ">Figure 9
<p>Average sum rate (b/s) against SNR (dB) for two SUs.</p> Full article ">Figure 10
<p>Average sum rate (b/s) against SNR (dB) for two SUs with DSTBC.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-22
Previous Issue
Next Issue
Back to TopTop