[go: up one dir, main page]
Next Issue
Volume 21, December-1
Previous Issue
Volume 21, November-1
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
7.3
3.4
Journals
Sensors
Volume 21
Issue 22
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 21, Issue 22 (November-2 2021) – 345 articles

Cover Story (view full-size image): The Ninapro dataset is a publicly available dataset designed to foster research on hand prosthesis, rehabilitation applications, and motor control studies. It has been exploited in several research projects in related fields. In this paper, an application on transfer learning is presented to test two domain adaptation techniques on a random forest classifier on EMG signals. The experiments were conducted on healthy subjects and amputees. Differently from several previous papers, no appreciable improvements were found in terms of accuracy, regardless of the transfer learning techniques tested. The lack of adaptive learning was also demonstrated for the first time in an intra-subject experimental setting when using as a source data acquisitions recorded from the same subject but on different days.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:
13 pages, 754 KiB  
Article
CoMeT: Configurable Tagged Memory Extension
by Jinjae Lee, Derry Pratama, Minjae Kim, Howon Kim and Donghyun Kwon
Sensors 2021, 21(22), 7771; https://doi.org/10.3390/s21227771 - 22 Nov 2021
Cited by 1 | Viewed by 2569
Abstract
Commodity processor architectures are releasing various instruction set extensions to support security solutions for the efficient mitigation of memory vulnerabilities. Among them, tagged memory extension (TME), such as ARM MTE and SPARC ADI, can prevent unauthorized memory access by utilizing tagged memory. However, [...] Read more.
Commodity processor architectures are releasing various instruction set extensions to support security solutions for the efficient mitigation of memory vulnerabilities. Among them, tagged memory extension (TME), such as ARM MTE and SPARC ADI, can prevent unauthorized memory access by utilizing tagged memory. However, our analysis found that TME has performance and security issues in practical use. To alleviate these, in this paper, we propose CoMeT, a new instruction set extension for tagged memory. The key idea behind CoMeT is not only to check whether the tag values in the address tag and memory tag are matched, but also to check the access permissions for each tag value. We implemented the prototype of CoMeT on the RISC-V platform. Our evaluation results confirm that CoMeT can be utilized to efficiently implement well-known security solutions, i.e., shadow stack and in-process isolation, without compromising security. Full article
(This article belongs to the Special Issue Access Control in the Internet of Things)
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of RISC-V MTE.</p> Full article ">Figure 2
<p>Overview of <tt>CoMeT</tt>.</p> Full article ">Figure 3
<p>TPCR (Tag Permission Configuration Register).</p> Full article ">Figure 4
<p>TPCR updating instructions.</p> Full article ">Figure 5
<p>Access Control in <tt>CoMeT</tt>.</p> Full article ">Figure 6
<p>Shadow stack and In-process isolation in <tt>CoMeT</tt>: (<b>a</b>) pseudo code of shadow stack in <tt>CoMeT</tt> and (<b>b</b>) pseudo code of in-process isolation in <tt>CoMeT</tt>.</p> Full article ">Figure 7
<p>Shadow stack and In-process isolation in TME: (<b>a</b>) pseudo code of shadow stack in TME and (<b>b</b>) pseudo code of in-process isolation in TME.</p> Full article ">Figure 8
<p>Execution time overhead for the shadow stack with <tt>CoMeT</tt> and TME.</p> Full article ">Figure 9
<p>Execution time overhead for the in-process isolation with <tt>CoMeT</tt> and TME.</p> Full article ">
19 pages, 5677 KiB  
Article
Metrological Characterization and Comparison of D415, D455, L515 RealSense Devices in the Close Range
by Michaela Servi, Elisa Mussi, Andrea Profili, Rocco Furferi, Yary Volpe, Lapo Governi and Francesco Buonamici
Sensors 2021, 21(22), 7770; https://doi.org/10.3390/s21227770 - 22 Nov 2021
Cited by 31 | Viewed by 4483
Abstract
RGB-D cameras are employed in several research fields and application scenarios. Choosing the most appropriate sensor has been made more difficult by the increasing offer of available products. Due to the novelty of RGB-D technologies, there was a lack of tools to measure [...] Read more.
RGB-D cameras are employed in several research fields and application scenarios. Choosing the most appropriate sensor has been made more difficult by the increasing offer of available products. Due to the novelty of RGB-D technologies, there was a lack of tools to measure and compare performances of this type of sensor from a metrological perspective. The recent ISO 10360-13:2021 represents the most advanced international standard regulating metrological characterization of coordinate measuring systems. Part 13, specifically, considers 3D optical sensors. This paper applies the methodology of ISO 10360-13 for the characterization and comparison of three RGB-D cameras produced by Intel® RealSense™ (D415, D455, L515) in the close range (100–1500 mm). ISO 10360-13 procedures, which focus on metrological performances, are integrated with additional tests to evaluate systematic errors (acquisition of flat objects, 3D reconstruction of objects). The present paper proposes an off-the-shelf comparison which considers the performance of the sensors throughout their acquisition volume. Results have exposed the strengths and weaknesses of each device. The D415 device showed better reconstruction quality on tests strictly related to the short range. The L515 device performed better on systematic depth errors; finally, the D455 device achieved better results on tests related to the standard. Full article
(This article belongs to the Special Issue Recent Advances in Depth Sensors and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Example of spherical artifact acquired in a voxel of the measurement volume.</p> Full article ">Figure 2
<p>Ball bar positions provided by the standard for the distortion error measurement.</p> Full article ">Figure 3
<p>Planar artifact positions provided by the standard to measure the flat distortion error.</p> Full article ">Figure 4
<p>Example of the artifact composed of 5 center-to-center length acquired in the concatenated measurement volume.</p> Full article ">Figure 5
<p>Three depth cameras under review in this study. From left to right: Intel<sup>®</sup> RealSense™ Depth Camera D415, Intel<sup>®</sup> RealSense™ Depth Camera D455, Intel<sup>®</sup> RealSense™ LiDAR Camera L515.</p> Full article ">Figure 6
<p>Adopted metrological characterization framework. The Figure shows which tests were performed depending on the acquisition distance.</p> Full article ">Figure 7
<p>Field of view of the three devices in the range 500–1500 mm. From left to right: (<b>a</b>) D415, (<b>b</b>) D455, (<b>c</b>) L515.</p> Full article ">Figure 8
<p>Acquisition setup for the very-close range test as in [<a href="#B24-sensors-21-07770" class="html-bibr">24</a>].</p> Full article ">Figure 9
<p>Artifacts selected for device characterization according to the standard ISO 10360-13:2021.</p> Full article ">Figure 10
<p>Acquisition setup to ensure camera-plane perpendicularity during the systematic depth error evaluation test.</p> Full article ">Figure 11
<p>Error of calibrated sphere diameter estimation in the range 100–500 mm for (<b>a</b>) the D415, (<b>b</b>) the D455 and (<b>c</b>) L515 devices.</p> Full article ">Figure 12
<p>Acquisitions of the sphere artifact performed by the three sensors according to the ISO 10360-13 standard. (<b>a</b>) D415 data; (<b>b</b>) D455 data; (<b>c</b>) L515 data.</p> Full article ">Figure 13
<p>Ball bar acquisitions performed by the three sensors according to the ISO 10360-13 standard. (<b>a</b>) D415 data; (<b>b</b>) D455 data; (<b>c</b>) L515 data.</p> Full article ">Figure 14
<p>Plane acquisitions performed by the three sensors according to the ISO 10360-13 standard. (<b>a</b>) D415 data; (<b>b</b>) D455 data; (<b>c</b>) L515 data.</p> Full article ">Figure 15
<p>(<b>a</b>) superior view (top) and lateral view (bottom) of the acquisition performed for the evaluation of the systematic depth errors; (<b>b</b>) detail of a few acquired planes simplified through lines that interpolate points in correspondence of two orthogonal planes (XZ plane, YZ plane) that intersect the optical axis of the camera.</p> Full article ">Figure 16
<p>Error for the planarity estimation test in the range 100–1500 mm for (<b>a</b>) the D415, (<b>b</b>) the D455 and (<b>c</b>) L515 devices.</p> Full article ">Figure 17
<p>Deviation results of the two 3D reconstructions (free form statue object and tangram object). From left to right, the figure shows the data obtained with D415, D455 and L515 sensors. The colour scale describes the distance between the reference model and the acquired model in mm; positive values are marked with colours from green to red, negative values are marked with colours from blue to green.</p> Full article ">
11 pages, 4686 KiB  
Article
Development of Road Surface Detection Algorithm Using CycleGAN-Augmented Dataset
by Wansik Choi, Jun Heo and Changsun Ahn
Sensors 2021, 21(22), 7769; https://doi.org/10.3390/s21227769 - 22 Nov 2021
Cited by 13 | Viewed by 4073
Abstract
Road surface detection is important for safely driving autonomous vehicles. This is because the knowledge of road surface conditions, in particular, dry, wet, and snowy surfaces, should be considered for driving control of autonomous vehicles. With the rise of deep learning technology, road [...] Read more.
Road surface detection is important for safely driving autonomous vehicles. This is because the knowledge of road surface conditions, in particular, dry, wet, and snowy surfaces, should be considered for driving control of autonomous vehicles. With the rise of deep learning technology, road surface detection methods using deep neural networks (DNN) have been widely used for developing road surface detection algorithms. To apply DNN in road surface detection, the dataset should be large and well-balanced for accurate and robust performance. However, most of the images of road surfaces obtained through usual data collection processes are not well-balanced. Most of the collected surface images tend to be of dry surfaces because road surface conditions are highly correlated with weather conditions. This could be a challenge in developing road surface detection algorithms. This paper proposes a method to balance the imbalanced dataset using CycleGAN to improve the performance of a road surface detection algorithm. CycleGAN was used to artificially generate images of wet and snow-covered roads. The road surface detection algorithm trained using the CycleGAN-augmented dataset had a better IoU than the method using imbalanced basic datasets. This result shows that CycleGAN-generated images can be used as datasets for road surface detection to improve the performance of DNN, and this method can help make the data acquisition process easy. Full article
(This article belongs to the Special Issue Artificial Intelligence and Their Applications in Smart Cities)
►▼ Show Figures

Figure 1

Figure 1
<p>Steps of the proposed road surface detection method.</p> Full article ">Figure 2
<p>Sample images of Mapillary Vistas dataset.</p> Full article ">Figure 3
<p>The number of images containing each road surface type in Mapillary Vistas dataset.</p> Full article ">Figure 4
<p>Concept of snowy road surface image data augmentation using CycleGAN.</p> Full article ">Figure 5
<p>Sample images translated from dry conditions (<b>top images</b>) to wet conditions (<b>bottom images</b>).</p> Full article ">Figure 6
<p>Sample images translated from dry conditions (<b>top images</b>) to snowy conditions (<b>bottom images</b>).</p> Full article ">Figure 7
<p>Sample images of the labeled images.</p> Full article ">Figure 8
<p>The structure of DeepLabv3+.</p> Full article ">Figure 9
<p>Road surface detection results.</p> Full article ">Figure 10
<p>Road surface detection results using the new real images.</p> Full article ">
20 pages, 9054 KiB  
Article
Point Cloud Resampling by Simulating Electric Charges on Metallic Surfaces
by Kyoungmin Han, Kyujin Jung, Jaeho Yoon and Minsik Lee
Sensors 2021, 21(22), 7768; https://doi.org/10.3390/s21227768 - 22 Nov 2021
Viewed by 2056
Abstract
3D point cloud resampling based on computational geometry is still a challenging problem. In this paper, we propose a point cloud resampling algorithm inspired by the physical characteristics of the repulsion forces between point electrons. The points in the point cloud are considered [...] Read more.
3D point cloud resampling based on computational geometry is still a challenging problem. In this paper, we propose a point cloud resampling algorithm inspired by the physical characteristics of the repulsion forces between point electrons. The points in the point cloud are considered as electrons that reside on a virtual metallic surface. We iteratively update the positions of the points by simulating the electromagnetic forces between them. Intuitively, the input point cloud becomes evenly distributed by the repulsive forces. We further adopt an acceleration and damping terms in our simulation. This system can be viewed as a momentum method in mathematical optimization and thus increases the convergence stability and uniformity performance. The net force of the repulsion forces may contain a normal directional force with respect to the local surface, which can make the point diverge from the surface. To prevent this, we introduce a simple restriction method that limits the repulsion forces between the points to an approximated local plane. This approach mimics the natural phenomenon in which positive electrons cannot escape from the metallic surface. However, this is still an approximation because the surfaces are often curved rather than being strict planes. Therefore, we project the points to the nearest local surface after the movement. In addition, we approximate the net repulsion force using the K-nearest neighbor to accelerate our algorithm. Furthermore, we propose a new measurement criterion that evaluates the uniformity of the resampled point cloud to compare the proposed algorithm with baselines. In experiments, our algorithm demonstrates superior performance in terms of uniformization, convergence, and run-time. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of point cloud resampling algorithm. The input point cloud <span class="html-italic">P</span> is assumed to be zero-centered and rescaled. First, the resampled point cloud <math display="inline"><semantics> <msup> <mi>Q</mi> <mn>0</mn> </msup> </semantics></math>, velocity <math display="inline"><semantics> <msup> <mi mathvariant="bold-italic">V</mi> <mn>0</mn> </msup> </semantics></math>, and the normal vectors <math display="inline"><semantics> <msubsup> <mi>N</mi> <mrow> <msup> <mi>Q</mi> <mn>0</mn> </msup> </mrow> <mi>P</mi> </msubsup> </semantics></math> of the local tangent plane are initialized. In each iteration, we perform the following procedures: We compute the <span class="html-italic">K</span>-nearest neighbors from <math display="inline"><semantics> <msup> <mi>Q</mi> <mrow> <mi>t</mi> <mo>−</mo> <mn>1</mn> </mrow> </msup> </semantics></math> to calculate the net electric force. Then, the normal vectors of the local tangent planes, calculated in the previous iteration, are used to project the forces to the local surfaces. The next velocities and the new query point cloud <math display="inline"><semantics> <msup> <mi>Q</mi> <mi>t</mi> </msup> </semantics></math> are computed based on the forces additionally modified with damping terms. Then, we obtain the <span class="html-italic">K</span>-nearest neighbor for the updated point cloud <math display="inline"><semantics> <msup> <mi>Q</mi> <mi>t</mi> </msup> </semantics></math> and calculate the local tangent planes. To prevent <math display="inline"><semantics> <msup> <mi>Q</mi> <mi>t</mi> </msup> </semantics></math> from diverging, we project it using these new tangent planes. These planes can be reused in the next iteration to project electric forces for efficiency. After the iteration converges, the final output point cloud is rescaled to the original scale and is relocated to have the original center point.</p> Full article ">Figure 2
<p>PCA projection restrains the surface approximation error when moved points shift away from the input point cloud’s surface. By using the PCA projection, we project the moved points to the nearest local plane.</p> Full article ">Figure 3
<p>Conceptual image of PCA-based local surface extraction. In a 3D space, the normal vector of the plane is the 3rd eigenvector of the PCA result.</p> Full article ">Figure 4
<p>Example results for the tangential noise cases. The first row is the input point cloud, the second row is the resampling result of the LOP algorithm, the third row is that of the WLOP, and the final row is that of the proposed algorithm. The odd columns are the resampled point cloud (from left to right, Horse, Bunny, Kitten, Buddha, and Armadillo), and the even columns are the corresponding enlarged views.</p> Full article ">Figure 5
<p>Quantitative results for the tangential noise cases. Each column shows the results of algorithms applied to Horse, Bunny, Kitten, Buddha, and Armadillo. The <span class="html-italic">x</span>-axes in the plots indicate the radius of evaluating <span class="html-italic">u</span>. The ranges of the radius were determined proportional to the square roots of the ratios between the surface areas of point clouds and the numbers of points.</p> Full article ">Figure 6
<p>Qualitative results for a tangential noise case (Horse). The second row shows the enlarged views of the red boxes in the first row. The first column shows the input point cloud. The second column shows the result of the LOP. The third column shows that of the WLOP. The last column shows that of the proposed algorithm.</p> Full article ">Figure 7
<p>Quantitative results for the omnidirectional noise cases. Each column represents different input data (first column: Horse; second column: Bunny; third column: Kitten; fourth column: Buddha; and fifth column: Armadillo).</p> Full article ">Figure 8
<p>Qualitative results for an omnidirectional noise case (Horse). First column: input point cloud; second column: LOP; third column: WLOP; and fourth column: proposed method. The second row shows enlarged views of the first row.</p> Full article ">Figure 9
<p>Hole-filling results for the tangential directional noise case (Horse). First column: input point cloud with holes and tangential noise; second column: LOP; third column: WLOP; and fourth column: proposed method. The second row shows enlarged views of the first row.</p> Full article ">Figure 10
<p>Quantitative results for the tangential noise cases with resampling ratio 0.5. Each column represents different input data (first column: Horse; second column: Bunny; third column: Kitten; fourth column: Buddha; and fifth column: Armadillo).</p> Full article ">Figure 11
<p>Qualitative results for a tangential noise case with resampling ratio 0.5 (Horse). First column: input point cloud; second column: LOP; third column: WLOP; and fourth column: the proposed method. The second row shows enlarged views of the first row.</p> Full article ">Figure 12
<p>Quantitative results for the omnidirectional noise cases with resampling ratio 0.5. Each column represents different input data (first column: Horse; second column: Bunny; third column: Kitten; fourth column: Buddha; and fifth column: Armadillo).</p> Full article ">Figure 13
<p>Qualitative results for an omnidirectional noise case with resampling ratio 0.5 (Horse). First column: input point cloud; second column: LOP; third column: WLOP; and fourth column: the proposed method. The second row shows enlarged views of the first row.</p> Full article ">Figure 14
<p>Quantitative results for the tangential noise cases with resampling ratio 2.0. Each column represents different input data (first column: Horse; second column: Bunny; third column: Kitten; fourth column: Buddha; and fifth column: Armadillo).</p> Full article ">Figure 15
<p>Qualitative results for an tangential noise case with resampling ratio 2.0 (Horse). First column: input point cloud; second column: LOP; third column: WLOP; and fourth column: the proposed method. The second row shows enlarged views of the first row.</p> Full article ">Figure 16
<p>Quantitative results for the omnidirectional noise cases with resampling ratio 2.0. Each column represents different input data (first column: Horse; second column: Bunny; third column: Kitten; fourth column: Buddha; and fifth column: Armadillo).</p> Full article ">Figure 17
<p>Qualitative results for an omnidirectional noise case with resampling ratio 2.0 (Horse). First column: input point cloud; second column: LOP; third column: WLOP; and fourth column: the proposed method. The second row shows enlarged views of the first row.</p> Full article ">Figure 18
<p>Quantitative result for real data sets. The first and second columns show the uniformity results of each algorithm for Lemon and Flashlight.</p> Full article ">Figure 19
<p>Qualitative results for real data sets. The first row shows the resampled results of Lemon. The second row shows enlarged views of the first row. The third row shows the resampled results of Flashlight. The fourth row shows enlarged views of the third row. First column: input point cloud; second column: LOP; third column: WLOP; and fourth column: proposed method.</p> Full article ">Figure 20
<p>Quantitative performance of the proposed method for various <math display="inline"><semantics> <mi>α</mi> </semantics></math>. The horizontal axis indicates the iteration, and the vertical axis indicates the uniformity value. Each column represents a different input point cloud (first column: Horse, second column: Bunny, third column: Kitten, fourth column: Buddha, and fifth column: Armadillo).</p> Full article ">Figure 21
<p>Quantitative performance of the proposed method for various <math display="inline"><semantics> <mi>β</mi> </semantics></math>. The horizontal axis indicates the iteration, and the vertical axis indicates the uniformity value. Each column represents a different input point cloud (first column: Horse, second column: Bunny, third column: Kitten, fourth column: Buddha, and fifth column: Armadillo).</p> Full article ">Figure 22
<p>Convergence results of compared methods for the resampling experiment with tangential case. (first column: Horse, second column: Bunny, third column: Kitten, fourth column: Buddha, and fifth column: Armadillo).</p> Full article ">Figure 23
<p>Resampling results of low-density inputs. The input point clouds were generated by randomly subsampling the input data of <a href="#sensors-21-07768-f005" class="html-fig">Figure 5</a>. The percentages in the parentheses represent the amount of subsampling. First row: LOP, second row: WLOP, and third row: proposed method.</p> Full article ">Figure 24
<p>Resampling result of a genus-one shape. Left: LOP, middle: WLOP, and right: proposed method.</p> Full article ">Figure 25
<p>Resampling results of Dragon. (<b>Left</b>): LOP, (<b>Middle</b>): WLOP, (<b>Right</b>): proposed method.</p> Full article ">
21 pages, 12102 KiB  
Article
Dual-Input and Multi-Channel Convolutional Neural Network Model for Vehicle Speed Prediction
by Jiaming Xing, Liang Chu, Chong Guo, Shilin Pu and Zhuoran Hou
Sensors 2021, 21(22), 7767; https://doi.org/10.3390/s21227767 - 22 Nov 2021
Cited by 10 | Viewed by 2682
Abstract
With the development of technology, speed prediction has become an important part of intelligent vehicle control strategies. However, the time-varying and nonlinear nature of vehicle speed increases the complexity and difficulty of prediction. Therefore, a CNN-based neural network architecture with two channel input [...] Read more.
With the development of technology, speed prediction has become an important part of intelligent vehicle control strategies. However, the time-varying and nonlinear nature of vehicle speed increases the complexity and difficulty of prediction. Therefore, a CNN-based neural network architecture with two channel input (DICNN) is proposed in this paper. With two inputs and four channels, DICNN can predict the speed changes in the next 5 s by extracting the temporal information of 10 vehicle signals and the driver’s intention. The prediction performances of DICNN are firstly examined. The best RMSE, MAE, ME and R2 are obtained compared with a Markov chain combined with Monte Carlo (MCMC) simulation, a support vector machine (SVM) and a single input CNN (SICNN). Secondly, equivalent fuel consumption minimization strategies (ECMS) combining different vehicle speed prediction methods are constructed. After verification by simulation, the equivalent fuel consumption of the simulation increases by only 4.89% compared with dynamic-programming-based energy management strategy and decreased by 5.40% compared with the speed prediction method with low accuracy. Full article
(This article belongs to the Special Issue Advanced Vehicle to Everything (V2X) Communication and Application in Vehicle-Environment Cooperative Control)
►▼ Show Figures

Figure 1

Figure 1
<p>Flow chart of vehicle speed prediction.</p> Full article ">Figure 2
<p>The schematic of the PHEV configuration.</p> Full article ">Figure 3
<p>The schematic of data collection.</p> Full article ">Figure 4
<p>The architecture of DICNN.</p> Full article ">Figure 5
<p>The speed prediction results of four methods for WLTC. The dots (lines) with color are predicted values and the black dots (lines) are actual values.</p> Full article ">Figure 6
<p>The prediction error (blue) and prediction speed (orange) of MCMC.</p> Full article ">Figure 7
<p>The prediction error (blue) and prediction speed (orange) of SVM.</p> Full article ">Figure 8
<p>The prediction error (blue) and prediction speed (orange) of SICNN.</p> Full article ">Figure 9
<p>The prediction error (blue) and prediction speed (orange) of DICNN.</p> Full article ">Figure 10
<p>The prediction error (blue) and prediction speed (orange) for the next 1 s.</p> Full article ">Figure 11
<p>The prediction error (blue) and prediction speed (orange) for the next 5 s.</p> Full article ">Figure 12
<p>The box plots of the prediction errors.</p> Full article ">Figure 13
<p>The performances of the four methods.</p> Full article ">Figure 14
<p>Equivalent fuel consumption per 100 km.</p> Full article ">Figure 15
<p>The changes of fuel consumption under various strategies.</p> Full article ">Figure 16
<p>The changes of SOC under various strategies.</p> Full article ">Figure 17
<p>The distribution area of engine operating points under various strategies.</p> Full article ">Figure 18
<p>The distribution area of motor operating points under various strategies.</p> Full article ">Figure 19
<p>The number of engine operating points in each efficiency range.</p> Full article ">Figure 20
<p>The number of motor operating points in each efficiency range.</p> Full article ">
14 pages, 271126 KiB  
Article
A DNN-Based UVI Calculation Method Using Representative Color Information of Sun Object Images
by Deog-Hyeon Ga, Seung-Taek Oh and Jae-Hyun Lim
Sensors 2021, 21(22), 7766; https://doi.org/10.3390/s21227766 - 22 Nov 2021
Viewed by 1873
Abstract
As outdoor activities are necessary for maintaining our health, research interest in environmental conditions such as the weather, atmosphere, and ultraviolet (UV) radiation is increasing. In particular, UV radiation, which can benefit or harm the human body depending on the degree of exposure, [...] Read more.
As outdoor activities are necessary for maintaining our health, research interest in environmental conditions such as the weather, atmosphere, and ultraviolet (UV) radiation is increasing. In particular, UV radiation, which can benefit or harm the human body depending on the degree of exposure, is recognized as an essential environmental factor that needs to be identified. However, unlike the weather and atmospheric conditions, which can be identified to some extent by the naked eye, UV radiation corresponds to wavelength bands that humans cannot recognize; hence, the intensity of UV radiation cannot be measured. Recently, although devices and sensors that can measure UV radiation have been launched, it is very difficult for ordinary users to acquire ambient UV radiation information directly because of the cost and inconvenience caused by operating separate devices. Herein, a deep neural network (DNN)-based ultraviolet index (UVI) calculation method is proposed using representative color information of sun object images. First, Mask-region-based convolutional neural networks (R-CNN) are applied to sky images to extract sun object regions and then detect the representative color of the sun object regions. Then, a deep learning model is constructed to calculate the UVI by inputting RGB color values, which are representative colors detected later along with the altitude angle and azimuth of the sun at that time. After selecting each day of spring and autumn, the performance of the proposed method was tested, and it was confirmed that accurate UVI could be calculated within a range of mean absolute error of 0.3. Full article
(This article belongs to the Special Issue Sensors and Advanced Sensing Techniques for Signal, Image and Computer Vision Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Process of the proposed method.</p> Full article ">Figure 2
<p>Process Mask R-CNN-based sun object detection process.</p> Full article ">Figure 3
<p>Grid search for model optimization.</p> Full article ">Figure 4
<p>Structure of DNN models for UVI calculation.</p> Full article ">Figure 5
<p>Detection and image scaling of sun object images for sky images over time (12 January 2020).</p> Full article ">Figure 6
<p>Detection results of major color components for sun object images over time (8 October 2020).</p> Full article ">Figure 7
<p>Performance evaluation of DNN model.</p> Full article ">Figure 7 Cont.
<p>Performance evaluation of DNN model.</p> Full article ">
26 pages, 4229 KiB  
Article
Authorized Shared Electronic Medical Record System with Proxy Re-Encryption and Blockchain Technology
by Weizhe Chen, Shunzhi Zhu, Jianmin Li, Jiaxin Wu, Chin-Ling Chen and Yong-Yuan Deng
Sensors 2021, 21(22), 7765; https://doi.org/10.3390/s21227765 - 22 Nov 2021
Cited by 16 | Viewed by 3847
Abstract
With the popularity of the internet 5G network, the network constructions of hospitals have also rapidly developed. Operations management in the healthcare system is becoming paperless, for example, via a shared electronic medical record (EMR) system. A shared electronic medical record system plays [...] Read more.
With the popularity of the internet 5G network, the network constructions of hospitals have also rapidly developed. Operations management in the healthcare system is becoming paperless, for example, via a shared electronic medical record (EMR) system. A shared electronic medical record system plays an important role in reducing diagnosis costs and improving diagnostic accuracy. In the traditional electronic medical record system, centralized database storage is typically used. Once there is a problem with the data storage, it could cause data privacy disclosure and security risks. Blockchain is tamper-proof and data traceable. It can ensure the security and correctness of data. Proxy re-encryption technology can ensure the safe sharing and transmission of relatively sensitive data. Based on the above situation, we propose an electronic medical record system based on consortium blockchain and proxy re-encryption to solve the problem of EMR security sharing. Electronic equipment in this process is connected to the blockchain network, and the security of data access is ensured through the automatic execution of blockchain chaincodes; the attribute-based access control method ensures fine-grained access to the data and improves the system security. Compared with the existing electronic medical records based on cloud storage, the system not only realizes the sharing of electronic medical records, but it also has advantages in privacy protection, access control, data security, etc. Full article
(This article belongs to the Special Issue Blockchain Security and Its Application in Internet of Things)
►▼ Show Figures

Figure 1

Figure 1
<p>The System Framework.</p> Full article ">Figure 2
<p>The consortium blockchain service center architecture.</p> Full article ">Figure 3
<p>The example of the scanner device digital certificate.</p> Full article ">Figure 4
<p>The example of a user’s digital certificate.</p> Full article ">Figure 5
<p>The chaincode data structure.</p> Full article ">Figure 6
<p>The flow chart of user registration.</p> Full article ">Figure 7
<p>The flow chart of scanner device registration.</p> Full article ">Figure 8
<p>The flow chart of appointment and EMR generation.</p> Full article ">Figure 9
<p>The flow chart of re-encryption key and access voucher generation.</p> Full article ">Figure 10
<p>The flow chart of scanning of access voucher and acquisition of re-encrypted EMR.</p> Full article ">Figure 11
<p>Block structure and Merkle tree in the proposed scheme.</p> Full article ">Figure 12
<p>CBSC Configuration.</p> Full article ">Figure 13
<p>Send rate (TPS).</p> Full article ">
21 pages, 51076 KiB  
Article
Verification of a Stiffness-Variable Control System with Feed-Forward Predictive Earthquake Energy Analysis
by Tzu-Kang Lin, Tappiti Chandrasekhara, Zheng-Jia Liu and Ko-Yi Chen
Sensors 2021, 21(22), 7764; https://doi.org/10.3390/s21227764 - 22 Nov 2021
Cited by 2 | Viewed by 1524
Abstract
Semi-active isolation systems with controllable stiffness have been widely developed in the field of seismic mitigation. Most systems with controllable stiffness perform more robustly and effectively for far-field earthquakes than for near-fault earthquakes. Consequently, a comprehensive system that provides comparable reductions in seismic [...] Read more.
Semi-active isolation systems with controllable stiffness have been widely developed in the field of seismic mitigation. Most systems with controllable stiffness perform more robustly and effectively for far-field earthquakes than for near-fault earthquakes. Consequently, a comprehensive system that provides comparable reductions in seismic responses to both near-fault and far-field excitations is required. In this regard, a new algorithm called Feed-Forward Predictive Earthquake Energy Analysis (FPEEA) is proposed to identify the ground motion characteristics of and reduce the structural responses to earthquakes. The energy distribution of the seismic velocity spectrum is considered, and the balance between the kinetic energy and potential energy is optimized to reduce the seismic energy. To demonstrate the performance of the FPEEA algorithm, a two-degree-of-freedom structure was used as the benchmark in the numerical simulation. The peak structural responses under two near-fault and far-field earthquakes of different earthquake intensities were simulated. The isolation layer displacement was suppressed most by the FPEEA, which outperformed the other three control methods. Moreover, superior control on superstructure acceleration was also supported by the FPEEA. Experimental verification was then conducted with shaking table test, and the satisfactory performance of the FPEEA on both isolation layer displacement and superstructure acceleration was demonstrated again. In summary, the proposed FPEEA has potential for practical application to unexpected near-fault and far-field earthquakes. Full article
(This article belongs to the Special Issue Sensors in Structural Health Monitoring and Smart Structural Control)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic diagram of LSCIS isolation system. (<b>a</b>) Physical model of the LSCIS isolation system. (<b>b</b>) Mathematical model of the LSCIS isolation system. (<b>c</b>) Drawing of the LSCIS in 3D. (<b>d</b>) Side view of the LSCIS mechanism.</p> Full article ">Figure 1 Cont.
<p>Schematic diagram of LSCIS isolation system. (<b>a</b>) Physical model of the LSCIS isolation system. (<b>b</b>) Mathematical model of the LSCIS isolation system. (<b>c</b>) Drawing of the LSCIS in 3D. (<b>d</b>) Side view of the LSCIS mechanism.</p> Full article ">Figure 2
<p>Input time–history data for theoretical simulation. (<b>a</b>) Whittier Narrows-01. (<b>b</b>) Chi-Chi TCU068-EW.</p> Full article ">Figure 3
<p>Comparison of different control parameters. (<b>a</b>) Superstructure acceleration (Whittier Narrows-01). (<b>b</b>) Superstructure acceleration (TCU068-EW). (<b>c</b>) Isolation layer displacement (Whittier Narrows-01). (<b>d</b>) Isolation layer displacement (TCU068-EW).</p> Full article ">Figure 4
<p>Comparison of the maximum responses of various control laws at different PGA values. (<b>a</b>) Superstructure acceleration (WhittierNarrows-01). (<b>b</b>) Superstructure acceleration (TCU068- EW). (<b>c</b>) Isolation layer displacement (Whittier Narrows-01). (<b>d</b>) Isolation layer displacement (TCU068-EW).</p> Full article ">Figure 4 Cont.
<p>Comparison of the maximum responses of various control laws at different PGA values. (<b>a</b>) Superstructure acceleration (WhittierNarrows-01). (<b>b</b>) Superstructure acceleration (TCU068- EW). (<b>c</b>) Isolation layer displacement (Whittier Narrows-01). (<b>d</b>) Isolation layer displacement (TCU068-EW).</p> Full article ">Figure 5
<p>Time history inputs for experimental testing.</p> Full article ">Figure 6
<p>Experimental setup and the instrumentation configuration. (<b>a</b>) Assembling of the isolation layer and superstructure. (<b>b</b>) Side view of the instrumentation. (<b>c</b>) Front view of the instrumentation.</p> Full article ">Figure 6 Cont.
<p>Experimental setup and the instrumentation configuration. (<b>a</b>) Assembling of the isolation layer and superstructure. (<b>b</b>) Side view of the instrumentation. (<b>c</b>) Front view of the instrumentation.</p> Full article ">Figure 7
<p>Comparison of the responses between theoretical and experimental values (TCU102-EW PGA = 0.1 g). (<b>a</b>) Superstructure displacement. (<b>b</b>) Isolation layer displacement. (<b>c</b>) Superstructure acceleration. (<b>d</b>) Isolation layer acceleration. (<b>e</b>) Pivot displacement.</p> Full article ">Figure 8
<p>Comparison for the FPEEA and passive controls (Northridge, PGA = 0.3 g). (<b>a</b>) Isolation layer displacement. (<b>b</b>) Superstructure acceleration.</p> Full article ">Figure 9
<p>Comparison for the FPEEA and LIEM controls (Northridge, PGA = 0.3 g). (<b>a</b>) Isolation layer displacement. (<b>b</b>) Superstructure acceleration. (<b>c</b>) Pivot displacement. (<b>d</b>) Hysteresis loop.</p> Full article ">Figure 10
<p>Comparison for the FPEEA and MEW controls (Northridge, PGA = 0.3 g). (<b>a</b>) Isolation layer displacement. (<b>b</b>) Superstructure acceleration. (<b>c</b>) Pivot displacement. (<b>d</b>) Hysteresis loop.</p> Full article ">Figure 11
<p>Comparison for the FPEEA and passive controls (TCU102, PGA = 0.1 g). (<b>a</b>) Isolation layer displacement. (<b>b</b>) Superstructure acceleration.</p> Full article ">Figure 12
<p>Comparison for the FPEEA and LIEM controls (TCU102, PGA = 0.1 g). (<b>a</b>) Isolation layer displacement. (<b>b</b>) Superstructure acceleration. (<b>c</b>) Pivot displacement. (<b>d</b>) Hysteresis loop.</p> Full article ">Figure 13
<p>Comparison for the FPEEA and MEW controls (TCU102, PGA = 0.1 g). (<b>a</b>) Isolation layer displacement. (<b>b</b>) Superstructure acceleration. (<b>c</b>) Pivot displacement. (<b>d</b>) Hysteresis loop.</p> Full article ">
15 pages, 2549 KiB  
Article
Electrochemical DNA Sensor Based on Acridine Yellow Adsorbed on Glassy Carbon Electrode
by Tatjana Kulikova, Anna Porfireva, Alexey Rogov and Gennady Evtugyn
Sensors 2021, 21(22), 7763; https://doi.org/10.3390/s21227763 - 22 Nov 2021
Cited by 9 | Viewed by 2894
Abstract
Electrochemical DNA sensors offer unique opportunities for the sensitive detection of specific DNA interactions. In this work, a voltametric DNA sensor is proposed on the base of glassy carbon electrode modified with carbon black, adsorbed acridine yellow and DNA for highly sensitive determination [...] Read more.
Electrochemical DNA sensors offer unique opportunities for the sensitive detection of specific DNA interactions. In this work, a voltametric DNA sensor is proposed on the base of glassy carbon electrode modified with carbon black, adsorbed acridine yellow and DNA for highly sensitive determination of doxorubicin antitumor drug. The signal recorded by cyclic voltammetry was attributed to irreversible oxidation of the dye. Its value was altered by aggregation of the hydrophobic dye molecules on the carbon black particles. DNA molecules promote disaggregation of the dye and increased the signal. This effect was partially suppressed by doxorubicin compensate for the charge of DNA in the intercalation. Sensitivity of the signal toward DNA and doxorubicin was additionally increased by treatment of the layer with dimethylformamide. In optimal conditions, the linear range of doxorubicin concentrations determined was 0.1 pM–1.0 nM, and the detection limit was 0.07 pM. No influence of sulfonamide medicines and plasma electrolytes on the doxorubicin determination was shown. The DNA sensor was tested on two medications (doxorubicin-TEVA and doxorubicin-LANS) and showed recoveries of 102–105%. The DNA sensor developed can find applications in the determination of drug residues in blood and for the pharmacokinetics studies. Full article
(This article belongs to the Section Biosensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Cyclic voltammogram of 1.0 μM AY in 0.025 M phosphate buffer containing 0.1 M NaNO<sub>3</sub>, pH = 7.0, on the bare GCE, 100 mV/s. Arrows indicate direction of the potential scamming. (<b>b</b>) Cathodic AY peak currents in the series of consecutive measurements on the same bare glassy carbon (1) and that covered with the CB suspension in chitosan (2). Average ± S.D. for six replications on individual electrodes.</p> Full article ">Figure 2
<p>Cyclic voltammograms recorded on the GCE covered with CB (2 μL of 1.35 mg/mL in chitosan) and DNA (20 min incubation) in 1.0 μM AY solution in phosphate buffer, pH 4.0. The numbers (1–5) correspond to the number of potential scan performed after the immersion of the electrode in the dye solution. Arrows indicate the direction of the potential scan, 100 mV/s.</p> Full article ">Figure 3
<p>The dependence of the anodic peak current on the GCE covered with the mixture of the CB and AY on the surface layer content and the pH. The surface layer content is expressed as v:v ratio of the components mixed prior to deposition on bare GCE. 1—5:1, 2—10:1, 3—1:1, 4—1:5, 5—1:10 (the numbers also correspond to appropriate lines of <a href="#app1-sensors-21-07763" class="html-app">Table S1</a>).</p> Full article ">Figure 4
<p>(<b>a</b>) Cyclic voltammograms recorded on the GCE covered with CB/AY (1:1 mixture, chitosan, 2 μL per electrode) after 10 min incubation of the electrode in organic solvent. (<b>b</b>) The AY anodic peak currents recorded after the contact of the GCE covered with 1:1 mixture of CB/AY in chitosan with DMF (1), ethanol (2), chloroform (3) and phosphate buffer, pH = 4.0 (4). Average from three repetitions with individual electrodes.</p> Full article ">Figure 5
<p>The dependence of the anodic AY peak current on the GCE covered with the CB/AY (1:1) and adsorbed DNA from 1 mg/mL solution (incubation 10 min) on the modification protocol and pH. 1,2—CB suspended in chitosan, 3,4—CB suspended in DMF; 1,3—DMF treatment after the DNA adsorption, 2.4–40 min incubation with DMF.</p> Full article ">Figure 6
<p>The Nyquist diagrams of the impedance spectra recorded on the GCE covered with CB (1), CB/AY (2) and that after adsorption of DNA (3). Frequency range from 0.04 Hz to 100 kHz, amplitude of the applied sine potential 5 mV, 0.025 M phosphate buffer, pH = 4.0. Inset: equivalent circuit applied for the data fitting. <span class="html-italic">R</span> is the charge transfer resistance and <span class="html-italic">C</span> constant phase element Rs is the electrolyte resistance. Index 1 corresponds to the solution–layer interface and index 2 to the electrode–layer interface.</p> Full article ">Figure 7
<p>SEM images of the electrode surface covered with CB/AY and treated with DMF (<b>a</b>,<b>b</b>) and that preliminary treated with DNA solution and then additionally treated with DMF (<b>c</b>,<b>d</b>).</p> Full article ">Figure 8
<p>(<b>a</b>) Cyclic voltammograms recorded on the GCE covered with CB/AY (1:1 mixture, DMF, DNA, 2 μL/mL) after 20 min incubation in doxorubicin solution (0, 0.1, 1.0 pM, 0.01, 0.1, 1.0 and 5.0 nM). (<b>b</b>) Calibration plot of doxorubicin (DOX), measurements in 0.025 M phosphate buffer, pH = 3.0. Average for three individual electrodes.</p> Full article ">
15 pages, 5535 KiB  
Article
A New Bearing Fault Diagnosis Method Based on Capsule Network and Markov Transition Field/Gramian Angular Field
by Bin Han, Hui Zhang, Ming Sun and Fengtong Wu
Sensors 2021, 21(22), 7762; https://doi.org/10.3390/s21227762 - 22 Nov 2021
Cited by 36 | Viewed by 3419
Abstract
Compared to time-consuming and unreliable manual analysis, intelligent fault diagnosis techniques using deep learning models can improve the accuracy of intelligent fault diagnosis with their multi-layer nonlinear mapping capabilities. This paper proposes a model to perform fault diagnosis and classification by using a [...] Read more.
Compared to time-consuming and unreliable manual analysis, intelligent fault diagnosis techniques using deep learning models can improve the accuracy of intelligent fault diagnosis with their multi-layer nonlinear mapping capabilities. This paper proposes a model to perform fault diagnosis and classification by using a time series of vibration sensor data as the input. The model encodes the raw vibration signal into a two-dimensional image and performs feature extraction and classification by a deep convolutional neural network or improved capsule network. A fault diagnosis technique based on the Gramian Angular Field (GAF), the Markov Transition Field (MTF), and the Capsule Network is proposed. Experiments conducted on a bearing failure dataset from Case Western Reserve University investigated the impact of two coding methods and different network structures on the diagnosis accuracy. The results show that the GAF technique retains more complete fault characteristics, while the MTF technique contains a small number of fault characteristics but more dynamic characteristics. Therefore, the proposed method incorporates GAF images and MTF images as a dual-channel image input to the capsule network, enabling the network to obtain a more complete fault signature. Multiple sets of experiments were conducted on the bearing fault dataset at Case Western Reserve University, and the Capsule Network in the proposed model has an advantage over other convolutional neural networks and performs well in the comparison of fault diagnosis methods proposed by other researchers. Full article
(This article belongs to the Special Issue Intelligent Systems for Fault Diagnosis and Prognosis)
►▼ Show Figures

Figure 1

Figure 1
<p>The whole process of fault diagnosis.</p> Full article ">Figure 2
<p>CWRU dataset experimental platform.</p> Full article ">Figure 3
<p>Different sizes of GAF images. (<b>a</b>) 64 × 64, (<b>b</b>) 128 × 128, (<b>c</b>) 256 × 256, (<b>d</b>) 512 × 512.</p> Full article ">Figure 4
<p>CapsNet model structure diagram.</p> Full article ">Figure 5
<p>GAF-Deep learning experimental results.</p> Full article ">Figure 6
<p>Accuracy of different MTF sizes.</p> Full article ">Figure 7
<p>Different signal lengths are converted to 128 × 128 MTF images. (<b>a</b>) MTF image when the original signal length is 128. (<b>b</b>) MTF image when the original signal length is 256. (<b>c</b>) MTF image when the original signal length is 512.</p> Full article ">Figure 8
<p>Accuracy of GAFMTF dataset in CapsNet and convolutional neural networks.</p> Full article ">Figure 9
<p>Confusion matrix of A-H eight datasets in CapsNet.</p> Full article ">Figure 9 Cont.
<p>Confusion matrix of A-H eight datasets in CapsNet.</p> Full article ">
15 pages, 4645 KiB  
Article
A Scheme with Acoustic Emission Hit Removal for the Remaining Useful Life Prediction of Concrete Structures
by Tuan-Khai Nguyen, Zahoor Ahmad and Jong-Myon Kim
Sensors 2021, 21(22), 7761; https://doi.org/10.3390/s21227761 - 22 Nov 2021
Cited by 11 | Viewed by 2106
Abstract
In this study, a scheme of remaining useful lifetime (RUL) prognosis from raw acoustic emission (AE) data is presented to predict the concrete structure’s failure before its occurrence, thus possibly prolong its service life and minimizing the risk of accidental damage. The deterioration [...] Read more.
In this study, a scheme of remaining useful lifetime (RUL) prognosis from raw acoustic emission (AE) data is presented to predict the concrete structure’s failure before its occurrence, thus possibly prolong its service life and minimizing the risk of accidental damage. The deterioration process is portrayed by the health indicator (HI), which is automatically constructed from raw AE data with a deep neural network pretrained and fine-tuned by a stacked autoencoder deep neural network (SAE-DNN). For the deep neural network structure to perform a more accurate construction of health indicator lines, a hit removal process with a one-class support vector machine (OC-SVM), which has not been investigated in previous studies, is proposed to extract only the hits which matter the most to the portrait of deterioration. The new set of hits is then harnessed as the training labels for the deep neural network. After the completion of the health indicator line construction, health indicators are forwarded to a long short-term memory recurrent neural network (LSTM-RNN) for the training and validation of the remaining useful life prediction, as this structure is capable of capturing the long-term dependencies, even with a limited set of data. Our prediction result shows a significant improvement in comparison with a similar scheme but without the hit removal process and other methods, such as the gated recurrent unit recurrent neural network (GRU-RNN) and the simple recurrent neural network. Full article
(This article belongs to the Special Issue Fault Detection and Data Analysis for Structure and Infrastructure Engineering)
►▼ Show Figures

Figure 1

Figure 1
<p>Illustration of the RC beam under a four-point bending test: (<b>a</b>) Loading and LDVT placement, (<b>b</b>) sensor placement, (<b>c</b>) pictorial of the test setup, (<b>d</b>) pictorial of the gridded specimen.</p> Full article ">Figure 2
<p>Loading versus the bottom displacement during the two testing phases.</p> Full article ">Figure 3
<p>Flowchart of the proposed scheme.</p> Full article ">Figure 4
<p>CFAR diagram.</p> Full article ">Figure 5
<p>AE hit parameters.</p> Full article ">Figure 6
<p>Application of the CFAR on sensor 8, specimen A during time steps 400–401. (<b>a</b>) Raw data, (<b>b</b>) Cell power versus threshold, (<b>c</b>) Detected hits.</p> Full article ">Figure 7
<p>Visualization of normal versus anomalous hits in regard to the hit arrival time, rise time, and energy.</p> Full article ">Figure 8
<p>HI lines constructed by SAE-DNN (<b>a</b>) Specimen A, (<b>b</b>) Specimen B, (<b>c</b>) Specimen C.</p> Full article ">Figure 9
<p>HI prediction results from sensor 8 at two different time steps during the three tests. (<b>a</b>) Specimen A at the 350th s. (<b>b</b>) Specimen A at the 450th s. (<b>c</b>) Specimen B at the 350th s. (<b>d</b>) Specimen B at the 450th s. (<b>e</b>) Specimen C at the 350th s. (<b>f</b>) Specimen C at the 450th s.</p> Full article ">
21 pages, 6320 KiB  
Article
DDS and OPC UA Protocol Coexistence Solution in Real-Time and Industry 4.0 Context Using Non-Ideal Infrastructure
by Alexandru Ioana and Adrian Korodi
Sensors 2021, 21(22), 7760; https://doi.org/10.3390/s21227760 - 22 Nov 2021
Cited by 13 | Viewed by 3783
Abstract
Continuing the evolution towards Industry 4.0, the industrial communication protocols represent a significant topic of interest, as real-time data exchange between multiple devices constitute the pillar of Industrial Internet of Things (IIoT) scenarios. Although the legacy protocols are still persistent in the industry, [...] Read more.
Continuing the evolution towards Industry 4.0, the industrial communication protocols represent a significant topic of interest, as real-time data exchange between multiple devices constitute the pillar of Industrial Internet of Things (IIoT) scenarios. Although the legacy protocols are still persistent in the industry, the transition was initiated by the key Industry 4.0 facilitating protocol, the Open Platform Communication Unified Architecture (OPC UA). OPC UA has to reach the envisioned applicability, and it therefore has to consider coexistence with other emerging real-time oriented protocols in the production lines. The Data Distribution Service (DDS) will certainly be present in future architectures in some areas as robots, co-bots, and compact units. The current paper proposes a solution to evaluate the real-time coexistence of OPC UA and DDS protocols, functioning in parallel and in a gateway context. The purpose is to confirm the compatibility and feasibility between the two protocols alongside a general definition of criteria and expectations from an architectural point of view, pointing out advantages and disadvantages in a neutral manner, shaping a comprehensive view of the possibilities. The researched architecture is meant to comply with both performance comparison scenarios and interaction scenarios over a gateway application. Considering the industrial tendencies, the developed solution is applied using non-ideal infrastructures to provide a more feasible and faster applicability in the production lines. Full article
(This article belongs to the Section Internet of Things)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic view of OPC UA—DDS protocol coexistence in the Industry 4.0 context.</p> Full article ">Figure 2
<p>System architecture.</p> Full article ">Figure 3
<p>Multithreading nodes from an architectural perspective.</p> Full article ">Figure 4
<p>Data-buffering success rate percent-based results.</p> Full article ">Figure 5
<p>Generated Digital Signal based on payload delivered by the Gateway Application at 100 ms recurrence.</p> Full article ">Figure 6
<p>Generated Digital Signal based on payload delivered by the Gateway Application at 10 ms recurrence.</p> Full article ">Figure 7
<p>Generated Digital Signal based on payload delivered by the Gateway Application at 5 ms recurrence.</p> Full article ">Figure 8
<p>Generated Digital Signal based on payload delivered by the Gateway Application at 2 ms recurrence.</p> Full article ">Figure 9
<p>Generated Digital Signal based on payload delivered by the Gateway Application at 1 ms recurrence.</p> Full article ">
10 pages, 2011 KiB  
Communication
Evaluation of HPC Acceleration and Interconnect Technologies for High-Throughput Data Acquisition
by Alessandro Cilardo
Sensors 2021, 21(22), 7759; https://doi.org/10.3390/s21227759 - 22 Nov 2021
Cited by 1 | Viewed by 2182
Abstract
Efficient data movement in multi-node systems is a crucial issue at the crossroads of scientific computing, big data, and high-performance computing, impacting demanding data acquisition applications from high-energy physics to astronomy, where dedicated accelerators such as FPGA devices play a key role coupled [...] Read more.
Efficient data movement in multi-node systems is a crucial issue at the crossroads of scientific computing, big data, and high-performance computing, impacting demanding data acquisition applications from high-energy physics to astronomy, where dedicated accelerators such as FPGA devices play a key role coupled with high-performance interconnect technologies. Building on the outcome of the RECIPE Horizon 2020 research project, this work evaluates the use of high-bandwidth interconnect standards, namely InfiniBand EDR and HDR, along with remote direct memory access functions for direct exposure of FPGA accelerator memory across a multi-node system. The prototype we present aims at avoiding dedicated network interfaces built in the FPGA accelerator itself, leaving most of the resources for user acceleration and supporting state-of-the-art interconnect technologies. We present the detail of the proposed system and a quantitative evaluation in terms of end-to-end bandwidth as concretely measured with a real-world FPGA-based multi-node HPC workload. Full article
(This article belongs to the Special Issue Intelligent IoT Circuits and Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic of the baseline system. The dashed arrows indicate the communication paths exercised in our experimental evaluation.</p> Full article ">Figure 2
<p>(<b>a</b>) Schematic of the backplane-based system. The dashed arrow indicates the communication path exercised in our experimental evaluation. (<b>b</b>) Photograph of the system.</p> Full article ">Figure 3
<p>Host-to-host RDMA performance, as measured in the baseline system along Path 1.</p> Full article ">Figure 4
<p>Performance of remote direct accelerator memory access based on PCIe peer-to-peer communication, as measured in the baseline system along Path 2.</p> Full article ">Figure 5
<p>Performance of the stencil case study with remote direct accelerator memory access based on PCIe peer-to-peer communication, as measured in the baseline system along Path 2.</p> Full article ">Figure 6
<p>Comparisons of the performance of remote direct accelerator memory access based on PCIe peer-to-peer communication in the case of the baseline and the backplane-based system. The backplane cases (<b>a</b>) and (<b>b</b>) differ in the internal HBM interface clock frequency, 250 MHz and 450 MHz, respectively.</p> Full article ">
15 pages, 3904 KiB  
Article
Dual Properties of Polyvinyl Alcohol-Based Magnetorheological Plastomer with Different Ratio of DMSO/Water
by Norhiwani Mohd Hapipi, Saiful Amri Mazlan, Ubaidillah Ubaidillah, Siti Aishah Abdul Aziz, Seung-Bok Choi, Nur Azmah Nordin, Nurhazimah Nazmi, Zhengbin Pang and Shahir Mohd Yusuf
Sensors 2021, 21(22), 7758; https://doi.org/10.3390/s21227758 - 22 Nov 2021
Cited by 1 | Viewed by 2347
Abstract
Polyvinyl alcohol (PVA)-based magnetorheological plastomer (MRP) possesses excellent magnetically dependent mechanical properties such as the magnetorheological effect (MR effect) when exposed to an external magnetic field. PVA-based MRP also shows a shear stiffening (ST) effect, which is very beneficial in fabricating pressure sensor. [...] Read more.
Polyvinyl alcohol (PVA)-based magnetorheological plastomer (MRP) possesses excellent magnetically dependent mechanical properties such as the magnetorheological effect (MR effect) when exposed to an external magnetic field. PVA-based MRP also shows a shear stiffening (ST) effect, which is very beneficial in fabricating pressure sensor. Thus, it can automatically respond to external stimuli such as shear force without the magnetic field. The dual properties of PVA-based MRP mainly on the ST and MR effect are rarely reported. Therefore, this work empirically investigates the dual properties of this smart material under the influence of different solvent compositions (20:80, 40:60, 60:40, and 80:20) by varying the ratios of binary solvent mixture (dimethyl sulfoxide (DMSO) to water). Upon applying a shear stress with excitation frequencies from 0.01 to 10 Hz, the storage modulus (G′) for PVA-based MRP with DMSO to water ratio of 20:40 increases from 6.62 × 10−5 to 0.035 MPa. This result demonstrates an excellent ST effect with the relative shear stiffening effect (RSTE) up to 52,827%. In addition, both the ST and MR effect show a downward trend with increasing DMSO content to water. Notably, the physical state of hydrogel MRP could be changed with different solvent ratios either in the liquid-like or solid-like state. On the other hand, a transient stepwise experiment showed that the solvent’s composition had a positive effect on the arrangement of CIPs within the matrix as a function of the external magnetic field. Therefore, the solvent ratio (DMSO/water) can influence both ST and MR effects of hydrogel MRP, which need to be emphasized in the fabrication of hydrogel MRP for appropriate applications primarily with soft sensors and actuators for dynamic motion control. Full article
(This article belongs to the Special Issue Recent Advances in Smart Materials for Sensor and Actuator Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Basic structure of the proposed PVA-based MRP as a pressure/strain sensor to detect human motion (bending and stretching).</p> Full article ">Figure 2
<p>ESEM micrographs of the distribution of CIPs in the PVA-based MRP matrix, with magnifications of (<b>a</b>) ×5k and (<b>b</b>) ×10k, and (<b>c</b>) EDS mapping of the sample. Dot scanning analysis was conducted on the sample image as shown in the red dot circled in (<b>a</b>).</p> Full article ">Figure 3
<p>Shear storage and loss modulus of PVA-based MRP samples with different ratios of DMSO:water content with continuously changes in shear frequencies for (<b>a</b>) HMRP-20, (<b>b</b>) HMRP-40, (<b>c</b>) HMRP-60 and (<b>d</b>) HMRP-80.</p> Full article ">Figure 4
<p>Shear storage modulus of PVA-based MRP samples with different ratios of DMSO:water under continuously changing frequency.</p> Full article ">Figure 5
<p>Time dependence of G′ for PVA-based MRP with different solvent ratios in response to a stepwise magnetic field (<b>a</b>) and the dimensionless transient response of G′ (<b>b</b>) under a magnetic field of 500 mT.</p> Full article ">Figure 6
<p>Schematic illustration of the mechanism of CIP movement upon the application of external magnetic field in water-rich and DMSO-rich PVA-based MRP.</p> Full article ">Figure 7
<p>Shear storage modulus of PVA-based MRP samples with the different ratios of DMSO:water (<b>a</b>), and the calculated ASTE and RSTE (<b>b</b>) of the samples under a continuously changing magnetic flux density.</p> Full article ">Figure 8
<p>Relationship between loss factor and magnetic flux density for PVA-based MRP samples with different DMSO contents.</p> Full article ">
12 pages, 4947 KiB  
Article
Lensless Multispectral Camera Based on a Coded Aperture Array
by Jianwei Wang and Yan Zhao
Sensors 2021, 21(22), 7757; https://doi.org/10.3390/s21227757 - 22 Nov 2021
Cited by 4 | Viewed by 2402
Abstract
Multispectral imaging can be applied to water quality monitoring, medical diagnosis, and other applications, but the principle of multispectral imaging is different from the principle of hyper-spectral imaging. Multispectral imaging is generally achieved through filters, so multiple photos are required to obtain spectral [...] Read more.
Multispectral imaging can be applied to water quality monitoring, medical diagnosis, and other applications, but the principle of multispectral imaging is different from the principle of hyper-spectral imaging. Multispectral imaging is generally achieved through filters, so multiple photos are required to obtain spectral information. Using multiple detectors to take pictures at the same time increases the complexity and cost of the system. This paper proposes a simple multispectral camera based on lensless imaging, which does not require multiple lenses. The core of the system is the multispectral coding aperture. The coding aperture is divided into different regions and each region transmits the light of one wavelength, such that the spectral information of the target can be coded. By solving the inverse problem of sparse constraints, the multispectral information of the target is inverted. Herein, we analyzed the characteristics of this multispectral camera and developed a principle prototype to obtain experimental results. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Multispectral coding aperture, including filter array and coded aperture array.</p> Full article ">Figure 2
<p>Schematic diagram of the first type of coded aperture multispectral imaging principle. The distance between the centers of the sub-encoding apertures is D, and the width of the area where the target is projected onto the detector through each sub-aperture is d. When d is less than D, the light rays passing through the sub-apertures do not interfere with each other.</p> Full article ">Figure 3
<p>Schematic diagram of the second type of coded aperture multispectral imaging principle. When d is greater than D, the parts of light passing through different sub-apertures overlap on the detector, and the information between different wavebands is aliased.</p> Full article ">Figure 4
<p>Handwritten digital simulation image. The RGB images of the three color maps can be arranged in order to simulate different band information of the target.</p> Full article ">Figure 5
<p>Multispectral image data of band simulation target. The RGB image of the color image of the first number 5 corresponds to the first to third bands, the RGB image of color 3 and 4 of the second image corresponds to the 4th to 6th bands, and so on.</p> Full article ">Figure 6
<p>A set of 3 × 3 multispectral coding apertures. It is composed of nine different sub-encoding apertures, and each sub-aperture only encodes the image of one waveband in <a href="#sensors-21-07757-f005" class="html-fig">Figure 5</a>, simulating imaging of different wavebands.</p> Full article ">Figure 7
<p>The detector receives the image which is blurred after being encoded by sub-aperture.</p> Full article ">Figure 8
<p>Reconstruction of the multispectral image (PSNR = 19.704); comparison of combined pseudo-color images of different bands with original data. (<b>a</b>) Color map of reconstructed image; (<b>b</b>) reconstructed spectrum of pixel <span class="html-italic">x</span> = 52, <span class="html-italic">y</span> = 45 (D = 120 px, d = 20 px, 200 iterations, constraint term coefficient is 0.01).</p> Full article ">Figure 9
<p>The changes in the reconstructed image quality when changing the duty cycle (proportion of small holes in coded aperture) and D of the sub-encoding aperture (sub-aperture size 20 × 20). The reconstructed image quality only shows a correlation with D (Abscissa), and there is no significant difference between different curves (each curve is calculated using different duty cycle parameters—0.1, 0.2, 0.3, 0.4, 0.5).</p> Full article ">Figure 10
<p>Different sub-encoding aperture size, the root mean square error of reconstructed image under different spacing conditions. (Each curve is calculated using different sizes of sub-coded aperture—20, 25, 30, 35, 40).</p> Full article ">Figure 11
<p>(<b>a</b>) is a designed 31 × 31 SDTA as sub-coded aperture, (<b>b</b>) is multispectral coded aperture, which consists of nine sub-encoding apertures and nine filters; each filter covers a sub-encoding aperture, and the interval between adjacent sub-encoding apertures is 7.4 mm.</p> Full article ">Figure 12
<p>Multispectral coding aperture lensless imaging prototype.</p> Full article ">Figure 13
<p>Multispectral coded aperture imaging system prototype acquisition data.</p> Full article ">Figure 14
<p>(<b>a</b>) Image of the color led screen. The camera is 28.5 mm away from the screen. (<b>b</b>) Raw data collected. Since the LED has no near-infrared light, there is no image in the near-infrared band. (<b>c</b>) Image of a single lit LED, and the obtained image is preprocessed into a nine-band PSF. (<b>d</b>) Selection of the three bands (RGB) of the reconstructed multispectral image to form a color image, which is more consistent with the color displayed on the LED screen.</p> Full article ">
22 pages, 7126 KiB  
Article
Assessment and Feedback Control of Paving Quality of Earth-Rock Dam Based on OODA Loop
by Cheng Wang, Jiajun Wang, Wenlong Chen, Jia Yu, Zheng Jiao and Hongling Yu
Sensors 2021, 21(22), 7756; https://doi.org/10.3390/s21227756 - 22 Nov 2021
Cited by 4 | Viewed by 2413
Abstract
Paving thickness and evenness are two key factors that affect the paving operation quality of earth-rock dams. However, in the recent study, both of the key factors characterising the paving quality were measured using finite point random sampling, which resulted in subjectivity in [...] Read more.
Paving thickness and evenness are two key factors that affect the paving operation quality of earth-rock dams. However, in the recent study, both of the key factors characterising the paving quality were measured using finite point random sampling, which resulted in subjectivity in the detection and a lag in the feedback control. At the same time, the on-site control of the paving operation quality based on experience results in a poor and unreliable paving quality. To address the above issues, in this study, a novel assessment and feedback control framework for the paving operation quality based on the observe–orient–decide–act (OODA) loop is presented. First, in the observation module, a cellular automaton is used to convert the location of the bulldozer obtained by monitoring devices into the paving thickness of the levelling layer. Second, in the orient module, the learning automaton is used to update the state of the corresponding and surrounding cells. Third, in the decision module, an overall path planning method is developed to realise feedback control of the paving thickness and evenness. Finally, in the act module, the paving thickness and evenness of the entire work unit are calculated and compared to their control thresholds to determine whether to proceed with the next OODA loop. The experiments show that the proposed method can maintain the paving thickness less than the designed standard value and effectively prevent the occurrence of ultra-thick or ultra-thin phenomena. Furthermore, the paving evenness is improved by 21.5% as compared to that obtained with the conventional paving quality control method. The framework of the paving quality assessment and feedback control proposed in this paper has extensive popularisation and application value for the same paving construction scene. Full article
(This article belongs to the Topic GNSS Measurement Technique in Aerial Navigation)
►▼ Show Figures

Figure 1

Figure 1
<p>Paving construction process scene in core wall area of earth-rock dam.</p> Full article ">Figure 2
<p>Composition and application of the proposed assessment and feedback control framework.</p> Full article ">Figure 3
<p>Theoretical framework of OODA loop.</p> Full article ">Figure 4
<p>Relationship between the LA and the environment.</p> Full article ">Figure 5
<p>Unloading process diagram of dump truck.</p> Full article ">Figure 6
<p>Calculation process of paving thickness of storehouse surface: (<b>a</b>) elevation of the previous storehouse; (<b>b</b>) elevation of storehouse in construction; (<b>c</b>) superimposition of (<b>a</b>,<b>b</b>); (<b>d</b>) calculation of paving thickness of storehouse surface.</p> Full article ">Figure 7
<p>Feedback control of bulldozer by OODA decision module.</p> Full article ">Figure 8
<p>Real-time acquisition process of paving operation parameters.</p> Full article ">Figure 9
<p>Equipment arrangement of monitoring terminal.</p> Full article ">Figure 10
<p>Three-dimensional colour map of the elevation of the entire working area.</p> Full article ">Figure 11
<p>Three-dimensional surface colour map of the elevation at each position at the end of the paving of the two sets.</p> Full article ">Figure 12
<p>Real-time path optimisation based on feedback control.</p> Full article ">Figure 13
<p>Comparative analysis results of paving efficiency and flatness.</p> Full article ">Figure 14
<p>Trajectory of bulldozers and corresponding graphic reports for each test and contrast set.</p> Full article ">
8 pages, 2892 KiB  
Communication
In Situ Femtosecond-Laser-Induced Fluorophores on Surface of Polyvinyl Alcohol for H2O/Co2+ Sensing and Data Security
by Weiliang Chen, Jichao Gao, Jie Tian and Jingyu Zhang
Sensors 2021, 21(22), 7755; https://doi.org/10.3390/s21227755 - 22 Nov 2021
Viewed by 2394
Abstract
In situ fluorophores were induced on polyvinyl alcohol (PVA) bulk materials by direct femtosecond laser writing. The generation of fluorophores was ascribed to localized laser-assisted carbonization. The carbonization of PVA polymers was confirmed through X-ray photoelectron spectroscopy analysis. The distinct fluorescence responses of [...] Read more.
In situ fluorophores were induced on polyvinyl alcohol (PVA) bulk materials by direct femtosecond laser writing. The generation of fluorophores was ascribed to localized laser-assisted carbonization. The carbonization of PVA polymers was confirmed through X-ray photoelectron spectroscopy analysis. The distinct fluorescence responses of fluorophores in various solutions were harnessed for implementing in situ reagent sensors, metal ion sensors, data encryption, and data security applications. The demonstrated water detection sensor in acetone exhibited a sensitivity of 3%. Meanwhile, a data encryption scheme and a “burn after reading” technique were demonstrated by taking advantage of the respective reversible and irreversible switching properties of the in situ laser-induced fluorophores. Taking a step further, a quantitative cobalt ion measurement was demonstrated based on the concentration-dependent fluorescence recovery. Combined with a laser-induced hydrophilic modification, our scheme could enable “lab-on-a-chip” microfluidics sensors with arbitrary shape, varied flow delay, designed reaction zones, and targeted functionalities in the future. Full article
(This article belongs to the Special Issue State-of-the-Art Optical Sensors Technology in China)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Schematic of a pattern written in PVA using a femtosecond laser (<b>left</b>). Schematic of the carbonization and dehydration processes after femtosecond-laser-based irradiation (<b>right</b>).</p> Full article ">Figure 2
<p>(<b>a</b>) Photoluminescence spectra of modified PVA samples for different excitation wavelengths and pristine PVA sample for the excitation wavelength of 488 nm. (<b>b</b>) Confocal fluorescence microscopy image of voxel array excited at 488 nm. Fluorescence intensity as a function of pulse energy. The inset is a 4-level picture of a bus. The scale bars are 50 μm. XPS spectra of (<b>c</b>) the pristine PVA sample and (<b>d</b>) the laser-modified PVA sample.</p> Full article ">Figure 3
<p>Reagent sensing characteristics of the fluorophores. (<b>a</b>) Fluorescence retention of different solvents on the laser-modified PVA samples. (<b>b</b>) Fluorescence images of two patterned PVA samples before and after water (<b>i</b>,<b>ii</b>) and acetone (<b>iii</b>,<b>iv</b>) dropping, respectively. The scale bars are 50 µm. (<b>c</b>) The fluorescence retention as a function of water content in acetone. (<b>d</b>) Fluorescence intensities of the laser-modified PVA after continuous exposure to 488-nm-wavelength excitation. The intensities are normalized to the first measurement.</p> Full article ">Figure 4
<p>(<b>a</b>) Normalized fluorescence intensities of the laser-induced patterns following solvent deposition and drying. (<b>b</b>) A flower pattern fabricated using a femtosecond laser on the PVA surface, followed by 10 μL water dropping (<b>c</b>), and consequent heating at 35 °C for 2 h (<b>d</b>). (<b>e</b>) A fish pattern fabricated using a femtosecond laser on the PVA surface, followed by 10 μL Co<sup>2+</sup> aqueous solution (0.1 mol/L) dropping (<b>f</b>), and consequent heating at 35 °C for 2 h (<b>g</b>). (<b>h</b>) Recovered fluorescence intensity as a function of cobalt ion concentration. The insets are corresponding patterns. The intensities are normalized to the measurement of the untreated laser-patterned sample. The scale bars are 50 µm.</p> Full article ">Figure 5
<p>Hydrophilic surface laser patterning. Fluorescence images of the hydrophilic patterns (<b>a</b>) before and (<b>b</b>) after dropping the Co<sup>2+</sup> aqueous solution (0.1 mol/L) at the far right of the patterns. The gaps between the left squares and the right rectangles are 100 μm. The scale bars are 500 μm. (<b>c</b>) Corresponding SEM and (<b>d</b>) birefringent slow axis orientation images of the laser-patterned PVA surface. Pseudo colors (inset (<b>d</b>)) indicate the direction of the slow axis. The scale bars are 10 μm. Laser parameters: 300 mW, 10 mm/s scanning speed, 1 MHz repetition rate, 0.25 NA objective lens.</p> Full article ">
20 pages, 6020 KiB  
Article
A Methodology for the Multi-Point Characterization of Short-Term Temperature Fluctuations in Complex Microclimates Based on the European Standard EN 15757:2010: Application to the Archaeological Museum of L’Almoina (Valencia, Spain)
by Ignacio Díaz-Arellano, Manuel Zarzo, Fernando-Juan García-Diego and Angel Perles
Sensors 2021, 21(22), 7754; https://doi.org/10.3390/s21227754 - 22 Nov 2021
Cited by 3 | Viewed by 2075
Abstract
The monitoring and control of thermo-hygrometric indoor conditions is necessary for an adequate preservation of cultural heritage. The European standard EN 15757:2010 specifies a procedure for determining if seasonal patterns of relative humidity (RH) and temperature are adequate for the long-term preservation of [...] Read more.
The monitoring and control of thermo-hygrometric indoor conditions is necessary for an adequate preservation of cultural heritage. The European standard EN 15757:2010 specifies a procedure for determining if seasonal patterns of relative humidity (RH) and temperature are adequate for the long-term preservation of hygroscopic materials on display at museums, archives, libraries or heritage buildings. This procedure is based on the characterization of the seasonal patterns and the calculation of certain control limits, so that it is possible to assess whether certain changes in the microclimate can be harmful for the preventive conservation of artworks, which would lead to the implementation of corrective actions. In order to discuss the application of this standard, 27 autonomous data-loggers were located in different points at the Archaeological Museum of l’Almoina (Valencia). The HVAC system (heating, ventilation and air conditioning) at the museum tries to reach certain homogeneous environment, which becomes a challenge because parts of the ruins are covered by a skylight that produces a greenhouse effect in summer, resulting in severe thermo-hygrometric gradients. Based on the analysis of temperatures recorded during 16 months, the air conditions in this museum are discussed according to the standard EN 15757:2010, and some corrective measures are proposed to improve the conservation conditions. Although this standard is basically intended for data recorded from a single sensor, an alternative approach proposed in this work is to find zones inside the museum with a homogeneous microclimate and to discuss next the average values collected in each area. A methodology is presented to optimize the application of this standard in places with a complex microclimate like this case, when multiple sensors are located at different positions. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Position of 27 data-loggers installed at the Archaeological Center of l’Almoina [<a href="#B46-sensors-21-07754" class="html-bibr">46</a>]. Color codes indicate similarity in the pattern of recorded temperatures. The position of the skylight is marked as a dashed rectangle in black. The walkway used to visit the museum is indicated as a light-brown graticule (grid area).</p> Full article ">Figure 2
<p>Raw trajectories of the time series of temperature recorded every 5 min by the 27 data-loggers installed in l’Almoina Museum from 10th October 2019 until 8th February 2021. Some time series contains missing values.</p> Full article ">Figure 3
<p>Evolution over time of temperatures recorded by data-logger B2. The seasonal trend is indicated as a black line. Security limits calculated according to EN 15757:2010 are depicted in blue. The box in dashed lines indicates the period with a higher frequency of short-term fluctuations, between 20 April and 1 September.</p> Full article ">Figure 4
<p>Short-term fluctuations of temperature for data-logger B2 (i.e., deviations with respect to the seasonal trend). Security limits according to EN 15757:2010 are depicted as horizontal lines in red. The box in dashed lines (same as <a href="#sensors-21-07754-f003" class="html-fig">Figure 3</a>) indicates the period with a higher frequency of short-term fluctuations.</p> Full article ">Figure 5
<p>Short-term fluctuations of temperature for data-logger B2 between 13 June and 19 June (i.e., zoom of <a href="#sensors-21-07754-f004" class="html-fig">Figure 4</a> for those six days). It can be noticed that values outside the limits appear at specific hours every day.</p> Full article ">Figure 6
<p>Short-term fluctuations of temperature for data-logger C5. Security limits according to EN 15757:2010 are depicted as dashed lines in red. Vertical equidistant dashed lines highlight instants of time when the time series reaches a minimum.</p> Full article ">Figure 7
<p>Periodogram computed with the short-term fluctuations of the mean time series of temperature (i.e., daily average of records from all data-loggers inside the museum). The two highest peaks indicate a seasonal pattern in the time series every 23 and 26 days.</p> Full article ">Figure 8
<p>Evolution over time of daily-mean temperatures: (i) average from all sensors located inside the museum (in green), (ii) from a nearby weather station at 2 km away (in blue), and (iii) from the data-logger installed outside (in orange; part of this trajectory is missing, mainly from March to April 2020). The two vertical dashed lines indicate a period when the indoor temperature suddenly drops lower than the outdoor temperature.</p> Full article ">Figure 9
<p>Number of observations outside the security limits for each data-logger (in columns) and each month (in rows). Such limits were determined by applying the standard EN 15757:2010 to each data-logger individually. Columns were sorted conveniently according to the three microclimates identified by PCA-clustering (i.e., the three blocks of columns correspond to the three color codes used for data-loggers in <a href="#sensors-21-07754-f001" class="html-fig">Figure 1</a>). For each column, the highest three values are highlighted in red, and the three lowest values, in green.</p> Full article ">Figure 10
<p>Band width (i.e., separation of security limits) for each data-logger according to EN 15757:2010. Each microclimate is represented by a different color (same as in <a href="#sensors-21-07754-f001" class="html-fig">Figure 1</a>): the central area under the skylight in red; the ruins of the Andalusian period in blue, and the rest in green.</p> Full article ">Figure 11
<p>Number of out-of-band observations for each month (in row) and each data-logger (in column) by considering a common band width. Such band was computed as the median of the band resulting from applying the standard EN 15757:2010 individually to each data-logger. The lowest tercile (i.e., percentile 33) of values are highlighted in green, the highest tercile in red, and the rest in light orange.</p> Full article ">Figure 12
<p>Number of out-of-band observations (i.e., values outside the security limits) for each data-logger computed by establishing the median band width (2.5 °C, see <a href="#sensors-21-07754-f010" class="html-fig">Figure 10</a>) as a common criterion for all data-loggers.</p> Full article ">Figure 13
<p>Percentage of out-of-band observations, for each month and each data-logger, with respect to the total. The number of out-of-band observations was computed by considering as a reference the median band width among data-loggers derived from EN 15757:2010. Values ≥ 2 are highlighted in red, &lt;1 in green and the rest in orange.</p> Full article ">
12 pages, 3391 KiB  
Article
Design and Performance Verification of a Space Radiation Detection Sensor Based on Graphene
by Heng An, Detian Li, Shengsheng Yang, Xuan Wen, Chenguang Zhang, Zhou Cao and Jun Wang
Sensors 2021, 21(22), 7753; https://doi.org/10.3390/s21227753 - 22 Nov 2021
Cited by 4 | Viewed by 2207
Abstract
In order to verify the performance of a graphene-based space radiation detection sensor, the radiation detection principle based on two-dimensional graphene material was analyzed according to the band structure and electric field effect of graphene. The method of space radiation detection based on [...] Read more.
In order to verify the performance of a graphene-based space radiation detection sensor, the radiation detection principle based on two-dimensional graphene material was analyzed according to the band structure and electric field effect of graphene. The method of space radiation detection based on graphene was studied and then a new type of space radiation sensor samples with small volume, high resolution, and radiation-resistance was formed. Using protons and electrons, the electrical performance of GFET radiation sensor was verified. The designed graphene space radiation detection sensor is expected to be applied in the radiation environment monitoring of the space station and the moon, and can also achieve technological breakthroughs in pulsar navigation and other fields. Full article
(This article belongs to the Section Sensor Materials)
►▼ Show Figures

Figure 1

Figure 1
<p>The electric field effect of graphene. (<b>A</b>) the basic structure of graphene field effect transistors [<a href="#B5-sensors-21-07753" class="html-bibr">5</a>]. (<b>B</b>) the change of graphene resistance with the intensity of the electric field without irradiation, with the green circle in the figure showing the output resistance position of field-effect transistors. (<b>C</b>) the graphene transistor has been exposed to radia-tion and that the substrate portion of the cell is ionized. (<b>D</b>) the output circuit of the graphene field effect transistor has shifted.</p> Full article ">Figure 2
<p>How graphene space radiation detectors work.</p> Full article ">Figure 3
<p>500 keV electron trajectory in Si.</p> Full article ">Figure 4
<p>Penetration depth and ionization energy loss of 40 KeV and 5 MeV proton.</p> Full article ">Figure 5
<p>CVD graphene and SEM image.</p> Full article ">Figure 6
<p>Graphene electrode pattern and graphene device channel diagram.</p> Full article ">Figure 7
<p>Schematic diagram of gold wire welding graphene device.</p> Full article ">Figure 8
<p>The output characteristic curve of graphene samples before and after electron irradiation. (<b>A</b>) Linear coordinates and (<b>B</b>) Logarithmic coordinates.</p> Full article ">Figure 9
<p>The transfer characteristic curve of graphene samples before and after electron irradiation. (<b>A</b>) Linear coordinates and (<b>B</b>) Logarithmic coordinates.</p> Full article ">Figure 10
<p>The output characteristic curve of graphene samples before and after proton irradiation. (<b>A</b>) Linear coordinates and (<b>B</b>) logarithmic coordinates.</p> Full article ">Figure 11
<p>The transfer characteristic curve of graphene samples before and after proton irradiation. (<b>A</b>) Linear coordinates and (<b>B</b>) logarithmic coordinates.</p> Full article ">Figure 12
<p>I–t curve of graphene samples before and after electron irradiation.</p> Full article ">Figure 13
<p>The normalized leakage current of graphene samples varies with the dose of electron irradiation.</p> Full article ">Figure 14
<p>The I–t curve of the graphene sample before and after proton irradiation.</p> Full article ">Figure 15
<p>The normalized leakage current of graphene samples varies with proton irradiation dose.</p> Full article ">
15 pages, 2198 KiB  
Article
In-Vehicle Alcohol Detection Using Low-Cost Sensors and Genetic Algorithms to Aid in the Drinking and Driving Detection
by Jose M. Celaya-Padilla, Jonathan S. Romero-González, Carlos E. Galvan-Tejada, Jorge I. Galvan-Tejada, Huizilopoztli Luna-García, Jose G. Arceo-Olague, Nadia K. Gamboa-Rosales, Claudia Sifuentes-Gallardo, Antonio Martinez-Torteya, José I. De la Rosa and Hamurabi Gamboa-Rosales
Sensors 2021, 21(22), 7752; https://doi.org/10.3390/s21227752 - 21 Nov 2021
Cited by 16 | Viewed by 10521
Abstract
Worldwide, motor vehicle accidents are one of the leading causes of death, with alcohol-related accidents playing a significant role, particularly in child death. Aiming to aid in the prevention of this type of accidents, a novel non-invasive method capable of detecting the presence [...] Read more.
Worldwide, motor vehicle accidents are one of the leading causes of death, with alcohol-related accidents playing a significant role, particularly in child death. Aiming to aid in the prevention of this type of accidents, a novel non-invasive method capable of detecting the presence of alcohol inside a motor vehicle is presented. The proposed methodology uses a series of low-cost alcohol MQ3 sensors located inside the vehicle, whose signals are stored, standardized, time-adjusted, and transformed into 5 s window samples. Statistical features are extracted from each sample and a feature selection strategy is carried out using a genetic algorithm, and a forward selection and backwards elimination methodology. The four features derived from this process were used to construct an SVM classification model that detects presence of alcohol. The experiments yielded 7200 samples, 80% of which were used to train the model. The rest were used to evaluate the performance of the model, which obtained an area under the ROC curve of 0.98 and a sensitivity of 0.979. These results suggest that the proposed methodology can be used to detect the presence of alcohol and enforce prevention actions. Full article
(This article belongs to the Special Issue Smartphone Sensors for Driver Behavior Monitoring Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of the proposed methodology.</p> Full article ">Figure 2
<p>Layout of the sensor placement.</p> Full article ">Figure 3
<p>Output from two MQ3 sensors exposed to the same alcohol sample.</p> Full article ">Figure 4
<p>Standardized signal from two MQ3 sensors exposed to the same alcohol sample.</p> Full article ">Figure 5
<p>Longitudinal behavior of a sensor without an alcohol sample.</p> Full article ">Figure 6
<p>Fitted line obtained after performing a linear regression.</p> Full article ">Figure 7
<p>Time-adjusted signal.</p> Full article ">Figure 8
<p>Model generation and validation methodology.</p> Full article ">Figure 9
<p>Average fitness of the models throughout the 200 generations.</p> Full article ">Figure 10
<p>Feature rank stability.</p> Full article ">Figure 11
<p>Accuracy of the models during the forward selection methodology.</p> Full article ">Figure 12
<p>ROC curve of the model on test data set.</p> Full article ">Figure 13
<p>Model performance on train/test samples: (<b>a</b>) confusion matrix for the train samples. (<b>b</b>) confusion matrix for the test samples.</p> Full article ">
15 pages, 20202 KiB  
Article
Highly Discriminative Physiological Parameters for Thermal Pattern Classification
by Laura Benita Alvarado-Cruz, Carina Toxqui-Quitl, Raúl Castro-Ortega, Alfonso Padilla-Vivanco and José Humberto Arroyo-Núñez
Sensors 2021, 21(22), 7751; https://doi.org/10.3390/s21227751 - 21 Nov 2021
Viewed by 1836
Abstract
Infrared Thermography (IRT) is a non-contact, non-intrusive, and non-ionizing radiation tool used for detecting breast lesions. This paper analyzes the surface temperature distribution (STD) on an optimal Region of Interest (RoI) for extraction of suitable internal heat source parameters. The physiological parameters are [...] Read more.
Infrared Thermography (IRT) is a non-contact, non-intrusive, and non-ionizing radiation tool used for detecting breast lesions. This paper analyzes the surface temperature distribution (STD) on an optimal Region of Interest (RoI) for extraction of suitable internal heat source parameters. The physiological parameters are estimated through the inverse solution of the bio-heat equation and the STD of suspicious areas related to the hottest spots of the RoI. To reach these values, the STD is analyzed by means: the Depth-Intensity-Radius (D-I-R) measurement model and the fitting method of Lorentz curve. A highly discriminative pattern vector composed of the extracted physiological parameters is proposed to classify normal and abnormal breast thermograms. A well-defined RoI is delimited at a radial distance, determined by the Support Vector Machines (SVM). Nevertheless, this distance is less than or equal to 1.8 cm due to the maximum temperature location close to the boundary image. The methodology is applied to 87 breast thermograms that belong to the Database for Mastology Research with Infrared Image (DMR-IR). This methodology does not apply any image enhancements or normalization of input data. At an optimal position, the three-dimensional scattergrams show a correct separation between normal and abnormal thermograms. In other cases, the feature vectors are highly correlated. According to our experimental results, the proposed pattern vector extracted at optimal position a=1.6 cm reaches the highest sensitivity, specificity, and accuracy. Even more, the proposed technique utilizes a reduced number of physiological parameters to obtain a Correct Rate Classification (CRC) of 100%. The precision assessment confirms the performance superiority of the proposed method compared with other techniques for the breast thermogram classification of the DMR-IR. Full article
(This article belongs to the Section Biomedical Sensors)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Scheme of the theoretical model of an internal heat source with depth <span class="html-italic">d</span>, intensity <span class="html-italic">q</span>, and radius <span class="html-italic">R</span>.</p> Full article ">Figure 2
<p>Flowchart of the proposed method. (<b>a</b>) Thermal data were obtained from the <math display="inline"><semantics> <mrow> <mi>D</mi> <mi>M</mi> <mi>R</mi> </mrow> </semantics></math>-<math display="inline"><semantics> <mrow> <mi>I</mi> <mi>R</mi> </mrow> </semantics></math>. (<b>b</b>) A well-defined RoI is delimited at an optimal radial distance <span class="html-italic">a</span>. As we can see, the RoI encircle the temperature area to be analyzed. (<b>c</b>) Surface temperature distribution related to the hottest spot of the RoI. (<b>d</b>) A proposed highly discriminative pattern vector is composed by the physiological parameters from the point heat source. (<b>e</b>) Classification step using SVM.</p> Full article ">Figure 3
<p>RoI delimitation process. (<b>a</b>) Segmentation procedure based on the detection of the inframammary line by a polynomial curve fitting. (<b>b</b>) Visualization of a clustered thermal pattern through thermal gradients. (<b>c</b>) Centered RoI around the hottest point with radial distance <span class="html-italic">a</span>.</p> Full article ">Figure 4
<p>(<b>a</b>) The extraction process of the STD using thermal input data. (<b>b</b>) Mean temperature distribution around the hottest point with radial distance <span class="html-italic">a</span>.</p> Full article ">Figure 5
<p>Comparison of the efficiency of two methods for extracting physiological parameters: fitting method of Lorentz curve (blue line) and the D-I-R model (red line). (<b>a</b>) STD fitted using the Lorentz curve method. We use the coefficient of determination (R-squared) to quantify the fitting between the surface temperature curve and the Lorentz curve. Estimation of physiological parameters (<b>b</b>) Depth <span class="html-italic">d</span> and (<b>c</b>) Intensity <span class="html-italic">q</span>.</p> Full article ">Figure 6
<p>Three-dimensional space using the proposed physiological parameters <math display="inline"><semantics> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>a</mi> </mrow> <mi>n</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mspace width="3.33333pt"/> </mrow> </semantics></math>{<math display="inline"><semantics> <mrow> <mi>θ</mi> <mo>,</mo> <mi>d</mi> <mo>,</mo> <mi>q</mi> </mrow> </semantics></math>} obtained through the D-I-R model, corresponding to <math display="inline"><semantics> <mrow> <mi>n</mi> <mo>=</mo> <mn>2</mn> </mrow> </semantics></math> at a given position <span class="html-italic">a</span>. The support vectors define the margin’s greatest separation between the normal and abnormal classes.</p> Full article ">Figure 7
<p>The temperature at the vicinity of affected tissue is about 2 °C higher than normal tissue.</p> Full article ">Figure 8
<p>Surface temperature distribution from (<b>a</b>) normal and (<b>b</b>) abnormal breast thermograms.</p> Full article ">Figure 9
<p>Three-dimensional scattergrams using the physiological parameters obtained by means of the fitting method of Lorentz curve at different positions <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0102</mn> </mrow> </semantics></math> m, <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0168</mn> </mrow> </semantics></math> m, and <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.018</mn> </mrow> </semantics></math> m. Column (<b>a</b>) corresponds to the pattern vector <math display="inline"><semantics> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>a</mi> </mrow> <mn>1</mn> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> </mrow> </semantics></math>{<math display="inline"><semantics> <msub> <mi>T</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> </semantics></math>, <span class="html-italic">d</span>, <span class="html-italic">q</span>} and column (<b>b</b>) corresponds to the pattern vector <math display="inline"><semantics> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>a</mi> </mrow> <mn>1</mn> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> </mrow> </semantics></math>{<math display="inline"><semantics> <mi>θ</mi> </semantics></math>, <span class="html-italic">d</span>, <span class="html-italic">q</span>}. As can be seen, at the optimal position <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0168</mn> </mrow> </semantics></math> m, the scattergrams show a correct separation between normal and abnormal thermograms. In other cases, the feature vectors are highly correlated.</p> Full article ">Figure 10
<p>Three-dimensional scattergrams using the physiological parameters obtained by means of the D-I-R model at different positions <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0102</mn> </mrow> </semantics></math> m, <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0168</mn> </mrow> </semantics></math> m, and <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.018</mn> </mrow> </semantics></math> m. Column (<b>a</b>) corresponds to the pattern vector <math display="inline"><semantics> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>a</mi> </mrow> <mn>2</mn> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> </mrow> </semantics></math>{<math display="inline"><semantics> <msub> <mi>T</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> </semantics></math>, <span class="html-italic">d</span>, <span class="html-italic">q</span>} and column (<b>b</b>) corresponds to the pattern vector <math display="inline"><semantics> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>a</mi> </mrow> <mn>2</mn> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> </mrow> </semantics></math>{<math display="inline"><semantics> <mi>θ</mi> </semantics></math>, <span class="html-italic">d</span>, <span class="html-italic">q</span>}. As can be observed, at the optimal position <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0168</mn> </mrow> </semantics></math> m, the scattergram shows a correct separation between normal and abnormal thermograms. In other cases, the feature vectors are highly correlated.</p> Full article ">Figure 11
<p>Three-dimensional scattergrams using the physiological parameters extracted from (<b>a</b>) the fitting method of Lorentz curve and (<b>b</b>) the D-I-R model. As can be seen, at the same optimal position <math display="inline"><semantics> <mrow> <mi>a</mi> <mo>=</mo> <mn>0.0168</mn> </mrow> </semantics></math> m, the scattergrams show a correct separation between normal and abnormal thermograms in both cases.</p> Full article ">Figure 12
<p>Classification results using the pattern vector <math display="inline"><semantics> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>a</mi> </mrow> <mi>n</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> </mrow> </semantics></math>{<math display="inline"><semantics> <mrow> <msub> <mi>T</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>,</mo> <mi>d</mi> <mo>,</mo> <mi>q</mi> <mo>,</mo> <mi>R</mi> <mo>,</mo> <mi>θ</mi> </mrow> </semantics></math>} obtained through the fitting method of Lorentz curve and D-I-R model at different <span class="html-italic">a</span> positions using SVM as a classifier.</p> Full article ">Figure 13
<p>ROC curves.</p> Full article ">
17 pages, 1552 KiB  
Article
An Enhanced Ensemble Approach for Non-Intrusive Energy Use Monitoring Based on Multidimensional Heterogeneity
by Yu Liu, Qianyun Shi, Yan Wang, Xin Zhao, Shan Gao and Xueliang Huang
Sensors 2021, 21(22), 7750; https://doi.org/10.3390/s21227750 - 21 Nov 2021
Cited by 1 | Viewed by 1680
Abstract
Acting as a virtual sensor network for household appliance energy use monitoring, non-intrusive load monitoring is emerging as the technical basis for refined electricity analysis as well as home energy management. Aiming for robust and reliable monitoring, the ensemble approach has been expected [...] Read more.
Acting as a virtual sensor network for household appliance energy use monitoring, non-intrusive load monitoring is emerging as the technical basis for refined electricity analysis as well as home energy management. Aiming for robust and reliable monitoring, the ensemble approach has been expected in load disaggregation, but the obstacles of design difficulty and computational inefficiency still exist. To address this, an ensemble design integrated with multi-heterogeneity is proposed for non-intrusive energy use disaggregation in this paper. Firstly, the idea of utilizing a heterogeneous design is presented, and the corresponding ensemble framework for load disaggregation is established. Then, a sparse coding model is allocated for individual classifiers, and the combined classifier is diversified by introducing different distance and similarity measures without consideration of sparsity, forming mutually heterogeneous classifiers. Lastly, a multiple-evaluations-based decision process is fine-tuned following the interactions of multi-heterogeneous committees, and finally deployed as the decision maker. Through verifications on both a low-voltage network simulator and a field measurement dataset, the proposed approach is demonstrated to be effective in enhancing load disaggregation performance robustly. By appropriately introducing the heterogeneous design into the ensemble approach, load monitoring improvements are observed with reduced computational burden, which stimulates research enthusiasm in investigating valid ensemble strategies for practical non-intrusive load monitoring implementations. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Multidimensional-heterogeneity-enhanced ensemble framework for NILM.</p> Full article ">Figure 2
<p>Heterogeneous design idea for the combined classifier and individual classifiers.</p> Full article ">Figure 3
<p>Design illustration of heterogeneous committees of the combined classifier.</p> Full article ">Figure 4
<p>Comparison visualization of <span class="html-italic">PHA</span> and <span class="html-italic">EPA</span> in metric performance.</p> Full article ">
16 pages, 5966 KiB  
Article
Ground-Based GNSS and Satellite Observations of Auroral Ionospheric Irregularities during Geomagnetic Disturbances in August 2018
by Irina Zakharenkova, Iurii Cherniak and Andrzej Krankowski
Sensors 2021, 21(22), 7749; https://doi.org/10.3390/s21227749 - 21 Nov 2021
Cited by 3 | Viewed by 2290
Abstract
The 25–26 August 2018 space weather event occurred during the solar minimum period and surprisingly became the third largest geomagnetic storm of the entire 24th solar cycle. We analyzed the ionospheric response at high latitudes of both hemispheres using multi-site ground-based GNSS observations [...] Read more.
The 25–26 August 2018 space weather event occurred during the solar minimum period and surprisingly became the third largest geomagnetic storm of the entire 24th solar cycle. We analyzed the ionospheric response at high latitudes of both hemispheres using multi-site ground-based GNSS observations and measurements onboard Swarm and DMSP satellites. With the storm development, the zones of intense ionospheric irregularities of auroral origin largely expanded in size and moved equatorward towards midlatitudes as far as ~55–60° magnetic latitude (MLAT) in the American, European, and Australian longitudinal sectors. The main ionospheric trough, associated with the equatorward side of the auroral oval, shifted as far equatorward as 45–50° MLAT at both hemispheres. The interhemispheric comparison revealed a high degree of similarity in a large expansion of the auroral irregularities oval towards midlatitudes, in addition to asymmetrical differences in terms of larger intensity of plasma density gradients and structures over the Southern auroral and polar cap regions. Evolution of the intense ionospheric irregularities and equatorward expansion of the auroral irregularities oval were well correlated with increases of geomagnetic activity and peaks of the auroral electrojet index. Full article
(This article belongs to the Special Issue Ground and Space-Borne Multi-Instrumental Radio-Wave-Based Research of Ionospheric Irregularities and Structures)
►▼ Show Figures

Figure 1

Figure 1
<p>Geomagnetic conditions during 25–27 August 2018: (<b>a</b>) IMF Bz component, (<b>b</b>) velocity and (<b>c</b>) dynamic pressure of the solar wind, (<b>d</b>) auroral electrojet index AE, (<b>e</b>) Kp index, and (<b>f</b>) SYM-H index.</p> Full article ">Figure 2
<p>Global GNSS ROTI maps for selected times on 25–26 August 2018. The thick black line marks the magnetic equator, the grey shaded area shows nighttime. High ROTI values (intense red color) depict severe ionospheric irregularities occurrence at equatorial and auroral zones.</p> Full article ">Figure 3
<p>Daily MLT-MLAT ROTI maps for the Northern Hemisphere for (<b>a</b>) 25 August, (<b>b</b>) 26 August, and (<b>c</b>) 27 August 2018. The maps cover 50–90° N MLAT with 10° latitude circles; magnetic local noon/midnight is at the top/bottom, and dusk/dawn is toward the left/right. Blue color corresponds to the absence or very weak ionospheric irregularities, red color—severe ionospheric irregularities occurrence.</p> Full article ">Figure 4
<p>North–south cross-sections (keograms) representing the spatio-temporal features of the storm-induced ionospheric irregularities as detected in GNSS ROTI along (<b>a</b>) 90° W, (<b>b</b>) 20° E, (<b>c</b>) 60° W, and (<b>d</b>) 150° E longitudes in the Northern and Southern Hemispheres during 25–26 August 2018. Vertical axes show (left) geographic and (right) geomagnetic latitudes, horizontal—UT and LT time. White color depicts empty cells due to lack of actual observations. The auroral oval boundaries predicted by the Feldstein–Starkov model are marked by the black solid lines for the poleward and equatorward boundaries, and by a dashed line for the diffuse boundary.</p> Full article ">Figure 5
<p>Two-dimensional GNSS ROTI maps in a polar view projection illustrating development of the auroral irregularities oval over the Northern (NH) and Southern (SH) Hemispheres for (<b>a</b>) 23:00 UT on 25 August, (<b>b</b>) 06:20 UT on 26 August, and (<b>c</b>) 16:20 UT on 26 August 2018. The maps cover 30°–90° N/S with 30° latitude/longitude grid. The grey shaded area shows nighttime, the maps are rotated with local midnight to be at the bottom. Black dot depicts location of the geomagnetic poles, the dashed magenta line shows the projection of the Swarm satellite overpass. Bottom panel of each plot shows variability of in situ electron density (Ne) along the Swarm overpass together with information about corresponding UT and geographic coordinates.</p> Full article ">Figure 6
<p>L1B Daily Summary Images from the Special Sensor Ultraviolet Scanning Imager (SSUSI) sensor onboard DMSP F17 satellite with a map projection of nightside disk data with nighttime solar zenith angles for 25–27 August 2018 (Day of Year 2018/237–2018/239) (<a href="https://ssusi.jhuapl.edu/images_daily_l1b" target="_blank">https://ssusi.jhuapl.edu/images_daily_l1b</a> (accessed on 16 November 2021)). For 25 August (top plot), the first DMSP F17 pass was near 80° W, next passes progressed to the left.</p> Full article ">Figure 7
<p>Satellite-based in situ density variations |∆N/N| plotted as a function of geomagnetic latitude (the y axis) and UT time (the x axis) during 25–27 August 2018 for the (<b>left</b>) Northern and (<b>right</b>) Southern Hemispheres for (<b>a,b</b>) Swarm-A, (<b>c,d</b>) Swarm-B, (<b>e,f</b>) DMSP F16, and (<b>g,h</b>) DMSP F17 satellites. Continuity leaps (white areas close to 90° N/S MLAT) appear due to satellite pass displacement from the magnetic poles.</p> Full article ">
25 pages, 8487 KiB  
Article
Effects of Plant Crown Shape on Microwave Backscattering Coefficients of Vegetation Canopy
by Xiangchen Liu, Yun Shao, Long Liu, Kun Li, Jingyuan Wang, Shuo Li, Jinning Wang and Xuexiao Wu
Sensors 2021, 21(22), 7748; https://doi.org/10.3390/s21227748 - 21 Nov 2021
Cited by 3 | Viewed by 3278
Abstract
A microwave scattering model is a powerful tool for determining relationships between vegetation parameters and backscattering characteristics. The crown shape of the vegetation canopy is an important parameter in forestry and affects the microwave scattering modeling results. However, there are few numerical models [...] Read more.
A microwave scattering model is a powerful tool for determining relationships between vegetation parameters and backscattering characteristics. The crown shape of the vegetation canopy is an important parameter in forestry and affects the microwave scattering modeling results. However, there are few numerical models or methods to describe the relationships between crown shapes and backscattering features. Using the Modified Tor Vergata Model (MTVM), a microwave scattering model based on the Matrix Doubling Algorithm (MDA), we quantitatively characterized the effects of crown shape on the microwave backscattering coefficients of the vegetation canopy. FEKO was also used as a computational electromagnetic method to make a complement and comparison with MTVM. In a preliminary experiment, the backscattering coefficients of two ideal vegetation canopies with four representative crown shapes (cylinder, cone, inverted cone and ellipsoid) were simulated: MTVM simulations were performed for the L (1.2 GHz), C (5.3 GHz) and X (9.6 GHz) bands in fully polarimetric mode, and FEKO simulations were carried out for the C (5.3 GHz) band at VV and VH polarization. The simulation results show that, for specific input parameters, the mean relative differences in backscattering coefficients due to variations in crown shape are as high as 127%, which demonstrates that the crown shape has a non-negligible influence on microwave backscattering coefficients of the vegetation canopy. In turn, this also suggests that investigation on effects of plant crown shape on microwave backscattering coefficients may have the potential to improve the accuracy of vegetation microwave scattering models, especially in canopies where volume scattering is the predominant mechanism. Full article
(This article belongs to the Special Issue Microwave Sensing and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Crown shapes in nature [<a href="#B22-sensors-21-07748" class="html-bibr">22</a>]. Vegetation crown shapes are written in bold; the corresponding typical vegetation types are written in normal font and are: (<b>a</b>–<b>e</b>).</p> Full article ">Figure 2
<p>Tridimensional geometry representation of the ideal crown shape in the cylindrical coordinate system <math display="inline"><semantics> <mrow> <mfenced> <mrow> <mrow> <mi>ρ</mi> <mo>,</mo> <mo> </mo> </mrow> <mi>ϕ</mi> <mrow> <mo>,</mo> <mo> </mo> <mi>h</mi> </mrow> </mrow> </mfenced> </mrow> </semantics></math>, where an ellipse represents a leaf, and a cylinder represents a branch; <math display="inline"><semantics> <mi>H</mi> </semantics></math> is crown height, and <math display="inline"><semantics> <mrow> <mrow> <mi>ρ</mi> <mo>(</mo> </mrow> <mi>ϕ</mi> <mrow> <mo>,</mo> <mo> </mo> <mi>h</mi> <mo>)</mo> <mo>=</mo> <mi>F</mi> <mo>(</mo> </mrow> <mi>ϕ</mi> <mrow> <mo>,</mo> <mo> </mo> <mi>h</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> is the crown envelope equation.</p> Full article ">Figure 3
<p>Geometric representations of the three specific crown shapes. The crown shapes, from left to right, are (<b>a</b>) cone, (<b>b</b>) inverted cone and (<b>c</b>) ellipsoid.</p> Full article ">Figure 4
<p>The process of using the Matrix Doubling Algorithm to calculate the multiple scattering of unit incident power between two close thin layers.</p> Full article ">Figure 5
<p>Two different structures of vegetation canopy components: (<b>a</b>) the component structure of canopy A; (<b>b</b>) the component structure of canopy B.</p> Full article ">Figure 6
<p>The four 3D geometrical vegetation canopies with different crown shapes when <span class="html-italic">H</span> = 1.5 m. In (<b>a</b>–<b>d</b>), the parameters from <a href="#sensors-21-07748-t001" class="html-table">Table 1</a> were used as model inputs; in (<b>e</b>–<b>h</b>), the parameters from <a href="#sensors-21-07748-t002" class="html-table">Table 2</a> were used as model inputs.</p> Full article ">Figure 7
<p>The MTVM simulation results of the backscattering coefficients for different crown shapes in L band (1.2 GHz) at (<b>a</b>) VV, (<b>b</b>) HH, (<b>c</b>) VH and (<b>d</b>) HV polarizations. The model input parameters are from <a href="#sensors-21-07748-t001" class="html-table">Table 1</a> (canopy A).</p> Full article ">Figure 8
<p>The MTVM simulation results of the backscattering coefficients for different crown shapes in C band (5.3 GHz) at (<b>a</b>) VV, (<b>b</b>) HH, (<b>c</b>) VH and (<b>d</b>) HV polarizations. The model input parameters are from <a href="#sensors-21-07748-t001" class="html-table">Table 1</a> (canopy A).</p> Full article ">Figure 9
<p>The MTVM simulation results of the backscattering coefficients for different crown shapes in X band (9.6 GHz) at (<b>a</b>) VV, (<b>b</b>) HH, (<b>c</b>) VH and (<b>d</b>) HV polarizations. The model input parameters are from <a href="#sensors-21-07748-t001" class="html-table">Table 1</a> (canopy A).</p> Full article ">Figure 10
<p>The MTVM simulation results of the backscattering coefficients for different crown shapes in L band (1.2 GHz) at (<b>a</b>) VV, (<b>b</b>) HH, (<b>c</b>) VH and (<b>d</b>) HV polarizations. The model input parameters are from <a href="#sensors-21-07748-t002" class="html-table">Table 2</a> (canopy B).</p> Full article ">Figure 11
<p>The MTVM simulation results of the backscattering coefficients for different crown shapes in C band (5.3 GHz) at (<b>a</b>) VV, (<b>b</b>) HH, (<b>c</b>) VH and (<b>d</b>) HV polarizations. The model input parameters are from <a href="#sensors-21-07748-t002" class="html-table">Table 2</a> (canopy B).</p> Full article ">Figure 12
<p>The MTVM simulation results of the backscattering coefficients for different crown shapes in X band (9.6 GHz) at (<b>a</b>) VV, (<b>b</b>) HH, (<b>c</b>) VH and (<b>d</b>) HV polarizations. The model input parameters are from <a href="#sensors-21-07748-t002" class="html-table">Table 2</a> (canopy B).</p> Full article ">Figure 13
<p>The feasible region of parabola factors a and b. The blue line represents the feasible region, the cross symbols represent the integer feasible solutions, and the bold dots with coordinates denote the solutions of a and b, which correspond to the four crown shapes.</p> Full article ">Figure 14
<p>The squares of the parabola equations with different integer feasible solutions, which correspond to the cross symbols in <a href="#sensors-21-07748-f013" class="html-fig">Figure 13</a>. The yellow, green, red and blue represent the square of the cylinder, cone, inverted cone and ellipsoid equations, respectively, and correspond to the bold dots in <a href="#sensors-21-07748-f013" class="html-fig">Figure 13</a>.</p> Full article ">Figure 15
<p>Backscattering coefficient simulation results of different crown shapes with discrete a and b in the feasible region; the bold dots correspond to the four crown shapes.</p> Full article ">Figure 16
<p>Rendering graphs of five transition crown shapes from the inverted cone to the cone. The parabola factors are (<b>a</b>) a = 1, b = 0; (<b>b</b>) a = 1, b = −0.5; (<b>c</b>) a = 1, b = −1; (<b>d</b>) a = 1, b = −1.5; and (<b>e</b>) a = 1, b = −2.</p> Full article ">Figure 17
<p>The simulated and fitted backscattering coefficients of vegetation canopies when using canopy A’s parameters as inputs at different canopy heights.</p> Full article ">Figure 18
<p>The averaged FEKO simulation results of the backscattering coefficients for different crown shapes at VV and VH polarizations of C band (5.3 GHz) when using canopy A’s parameters as inputs. (<b>a</b>) VV polarization; (<b>b</b>) VH polarization.</p> Full article ">Figure 19
<p>The averaged FEKO simulation results of the backscattering coefficients for different crown shapes at VV and VH polarizations of C band (5.3 GHz) when using canopy B’s parameters as inputs. (<b>a</b>) VV polarization; (<b>b</b>) VH polarization.</p> Full article ">
18 pages, 5988 KiB  
Article
Results of Large-Scale Propagation Models in Campus Corridor at 3.7 and 28 GHz
by Md Abdus Samad, Feyisa Debo Diba, Young-Jin Kim and Dong-You Choi
Sensors 2021, 21(22), 7747; https://doi.org/10.3390/s21227747 - 21 Nov 2021
Cited by 15 | Viewed by 2752
Abstract
The indoor application of wave propagation in the 5G network is essential to fulfill the increasing demands of network access in an indoor environment. This study investigated the wave propagation properties of line-of-sight (LOS) links at two long corridors of Chosun University (CU). [...] Read more.
The indoor application of wave propagation in the 5G network is essential to fulfill the increasing demands of network access in an indoor environment. This study investigated the wave propagation properties of line-of-sight (LOS) links at two long corridors of Chosun University (CU). We chose wave propagation measurements at 3.7 and 28 GHz, since 3.7 GHz is the closest to the roll-out frequency band of 3.5 GHz in South Korea and 28 GHz is next allocated frequency band for Korean telcos. In addition, 28 GHz is the promising millimeter band adopted by the Federal Communications Commission (FCC) for the 5G network. Thus, the 5G network can use 3.7 and 28 GHz frequencies to achieve the spectrum required for its roll-out frequency band. The results observed were applied to simulate the path loss of the LOS links at extended indoor corridor environments. The minimum mean square error (MMSE) approach was used to evaluate the distance and frequency-dependent optimized coefficients of the close-in (CI) model with a frequency-weighted path loss exponent (CIF), floating-intercept (FI), and alpha–beta–gamma (ABG) models. The outcome shows that the large-scale FI and CI models fitted the measured results at 3.7 and 28 GHz. Full article
(This article belongs to the Special Issue Modeling and Measurements of Propagation Environments for 5G and beyond Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Channel sounder architecture.</p> Full article ">Figure 2
<p>Measurement locations of the transmitter and receiver during the campaign. There are some structural changes in the corridor. These 2-dimensional changes of such spaces are marked with circled numbers and the length of the irregularities are: ➀ 8.894 m; ➁ 2.950 m; ➂ 1.540 m; ➃ 6.55 m; ➄ 2.322 m; ➅ 3.162 m; ➆ 1.371 m; and ➇ 3.267 m.</p> Full article ">Figure 3
<p>The experimental outlet is on the 3rd floor of the main building corridor. Structures that create irregularities in the corridor are marked with circled numbers: ➀; ➁; ➂; and ➃.</p> Full article ">Figure 4
<p>(<b>a</b>) The figure shows the location of the transmitter of the measurement campaign in the corridor of the IT convergence building on the 10th floor. The 1.75 m height transmitter was installed at 4.3 m along the wall; (<b>b</b>) the figure shows a measurement location while moving the receiver in a particular position along the corridor of the IT convergence building on the 10th floor. The picture includes the Rx horn antenna and the TAS antenna.</p> Full article ">Figure 5
<p>(<b>a</b>) The figure shows the 375 m-long corridor without human movement on the 3rd floor of the main building just before the measurement campaign; and (<b>b</b>) it shows a measurement scenario of the 3rd floor corridor of the main building.</p> Full article ">Figure 6
<p>The figure depicts CI, CIF, FI, ABG, and the measured path loss in LOS link at the frequency of 3.7 GHz (<b>a</b>,<b>c</b>) and at 28 GHz (<b>b</b>,<b>d</b>).</p> Full article ">Figure 7
<p>The figure depicts the point-to-point standard deviation of recorded data at 3.7 and 28 GHz under various antenna settings.</p> Full article ">
15 pages, 6013 KiB  
Article
New Results on Small and Dim Infrared Target Detection
by Hao Wang, Zehao Zhao, Chiman Kwan, Geqiang Zhou and Yaohong Chen
Sensors 2021, 21(22), 7746; https://doi.org/10.3390/s21227746 - 21 Nov 2021
Cited by 3 | Viewed by 2209
Abstract
Real-time small infrared (IR) target detection is critical to the performance of the situational awareness system in high-altitude aircraft. However, current IR target detection systems are generally hardware-unfriendly and have difficulty in achieving a robust performance in datasets with clouds occupying a large [...] Read more.
Real-time small infrared (IR) target detection is critical to the performance of the situational awareness system in high-altitude aircraft. However, current IR target detection systems are generally hardware-unfriendly and have difficulty in achieving a robust performance in datasets with clouds occupying a large proportion of the image background. In this paper, we present new results by using an efficient method that extracts the candidate targets in the pre-processing stage and fuses the local scale, blob-based contrast map and gradient map in the detection stage. We also developed mid-wave infrared (MWIR) and long-wave infrared (LWIR) cameras for data collection experiments and algorithm evaluations. Experimental results using both publicly available datasets and image sequences acquired by our cameras clearly demonstrated that the proposed method achieves high detection accuracy with the mean AUC being at least 22.3% higher than comparable methods, and the computational cost beating the other methods by a large margin. Full article
(This article belongs to the Special Issue Mid-Infrared Sensors and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Framework of the proposed IR small target detection method.</p> Full article ">Figure 2
<p>Demonstration of the 3-D and 1-D analysis for small IR targets with different sizes: (<b>a</b>) local patches contain targets; (<b>b</b>) 3-D mesh; (<b>c</b>) 1-D cross-section profile analysis.</p> Full article ">Figure 3
<p>The filtering results of the targets in the <a href="#sensors-21-07746-f002" class="html-fig">Figure 2</a>.</p> Full article ">Figure 4
<p>Illustrations of (<b>a</b>) input image, (<b>b</b>) binarized mask, (<b>c</b>) local operation region, (<b>d</b>) binarized filter kernel when <span class="html-italic">r</span> = 2, and (<b>e</b>) the surrounding region represented by the mask.</p> Full article ">Figure 5
<p>(<b>a</b>) Illustration for the operation region for the gradient analysis; (<b>b</b>–<b>e</b>) gradient filters for different directions.</p> Full article ">Figure 6
<p>(<b>a</b>) Illustration of index distance calculation for quadrant 1 in different situations; (<b>b</b>) example of the four quadrants having the same maximum gradient directions; (<b>c</b>) example of dominant direction of two quadrants having a negative gradient score.</p> Full article ">Figure 7
<p>(<b>a</b>) Illustration of the target board, (<b>b</b>) single-pixel target acquisition, and (<b>c</b>) sequence acquisition with cloudy weather.</p> Full article ">Figure 8
<p>The performance comparison of different methods on single-pixel target detection: (<b>a</b>) target image; (<b>b</b>–<b>h</b>) normalized 3-D mesh obtained by the LCM [<a href="#B18-sensors-21-07746" class="html-bibr">18</a>], IPI [<a href="#B10-sensors-21-07746" class="html-bibr">10</a>], MCPM [<a href="#B19-sensors-21-07746" class="html-bibr">19</a>], LIG [<a href="#B22-sensors-21-07746" class="html-bibr">22</a>], PSTNN [<a href="#B12-sensors-21-07746" class="html-bibr">12</a>], FKRW [<a href="#B15-sensors-21-07746" class="html-bibr">15</a>], and the proposed method; and (<b>i</b>) PRC comparison. The red arrows indicate the targets, and the green arrows indicate the noisy clutters.</p> Full article ">Figure 9
<p>The performance comparison of different methods on single-pixel image with defective pixels: (<b>a</b>) target image; (<b>b</b>–<b>h</b>) normalized 3-D mesh obtained by the LCM [<a href="#B18-sensors-21-07746" class="html-bibr">18</a>], IPI [<a href="#B10-sensors-21-07746" class="html-bibr">10</a>], MCPM [<a href="#B19-sensors-21-07746" class="html-bibr">19</a>], LIG [<a href="#B22-sensors-21-07746" class="html-bibr">22</a>], PSTNN [<a href="#B12-sensors-21-07746" class="html-bibr">12</a>], FKRW [<a href="#B15-sensors-21-07746" class="html-bibr">15</a>], and the proposed method; and (<b>i</b>) PRC comparison.</p> Full article ">Figure 10
<p>The detection results of different methods on S1: (<b>a</b>) the 49th image in S1 and (<b>b</b>–<b>h</b>) normalized 3-D mesh obtained by the LCM [<a href="#B18-sensors-21-07746" class="html-bibr">18</a>], IPI [<a href="#B10-sensors-21-07746" class="html-bibr">10</a>], MCPM [<a href="#B19-sensors-21-07746" class="html-bibr">19</a>], LIG [<a href="#B22-sensors-21-07746" class="html-bibr">22</a>], PSTNN [<a href="#B12-sensors-21-07746" class="html-bibr">12</a>], FKRW [<a href="#B15-sensors-21-07746" class="html-bibr">15</a>], and the proposed method, respectively.</p> Full article ">Figure 11
<p>The detection results of different methods on S2: (<b>a</b>) the 17th image in S2 and (<b>b</b>–<b>h</b>) normalized 3-D mesh obtained by the LCM [<a href="#B18-sensors-21-07746" class="html-bibr">18</a>], IPI [<a href="#B10-sensors-21-07746" class="html-bibr">10</a>], MCPM [<a href="#B19-sensors-21-07746" class="html-bibr">19</a>], LIG [<a href="#B22-sensors-21-07746" class="html-bibr">22</a>], PSTNN [<a href="#B12-sensors-21-07746" class="html-bibr">12</a>], FKRW [<a href="#B15-sensors-21-07746" class="html-bibr">15</a>], and the proposed method, respectively.</p> Full article ">Figure 12
<p>The detection results of different methods on S3: (<b>a</b>) the 1st image in S3 and (<b>b</b>–<b>h</b>) normalized 3-D mesh obtained by the LCM [<a href="#B18-sensors-21-07746" class="html-bibr">18</a>], IPI [<a href="#B10-sensors-21-07746" class="html-bibr">10</a>], MCPM [<a href="#B19-sensors-21-07746" class="html-bibr">19</a>], LIG [<a href="#B22-sensors-21-07746" class="html-bibr">22</a>], PSTNN [<a href="#B12-sensors-21-07746" class="html-bibr">12</a>], FKRW [<a href="#B15-sensors-21-07746" class="html-bibr">15</a>], and the proposed method, respectively.</p> Full article ">Figure 13
<p>The detection results of different methods on S4: (<b>a</b>) the 79th image in S4 and (<b>b</b>–<b>h</b>) normalized 3-D mesh obtained by the LCM [<a href="#B18-sensors-21-07746" class="html-bibr">18</a>], IPI [<a href="#B10-sensors-21-07746" class="html-bibr">10</a>], MCPM [<a href="#B19-sensors-21-07746" class="html-bibr">19</a>], LIG [<a href="#B22-sensors-21-07746" class="html-bibr">22</a>], PSTNN [<a href="#B12-sensors-21-07746" class="html-bibr">12</a>], FKRW [<a href="#B15-sensors-21-07746" class="html-bibr">15</a>], and the proposed method, respectively.</p> Full article ">Figure 14
<p>The PRC on different sequences: (<b>a</b>) PRC for S1, (<b>b</b>) PRC for S2, (<b>c</b>) PRC for S3, and (<b>d</b>) PRC for S4.</p> Full article ">
4 pages, 200 KiB  
Editorial
Editorial: Special Issue “Antenna Design for 5G and Beyond”
by Naser Ojaroudi Parchin, Chan Hwang See and Raed A. Abd-Alhameed
Sensors 2021, 21(22), 7745; https://doi.org/10.3390/s21227745 - 21 Nov 2021
Cited by 2 | Viewed by 1911
Abstract
The demand for high data rate transfer and large capacities of traffic is continuously growing as the world witnesses the development of the fifth generation (5G) of wireless communications with the fastest broadband speed yet and low latency [...] Full article
(This article belongs to the Special Issue Antenna Design for 5G and Beyond)
18 pages, 1189 KiB  
Article
Latency Reduction in Vehicular Sensing Applications by Dynamic 5G User Plane Function Allocation with Session Continuity
by Pablo Fondo-Ferreiro, David Candal-Ventureira, Francisco Javier González-Castaño and Felipe Gil-Castiñeira
Sensors 2021, 21(22), 7744; https://doi.org/10.3390/s21227744 - 21 Nov 2021
Cited by 3 | Viewed by 2377
Abstract
Vehicle automation is driving the integration of advanced sensors and new applications that demand high-quality information, such as collaborative sensing for enhanced situational awareness. In this work, we considered a vehicular sensing scenario supported by 5G communications, in which vehicle sensor data need [...] Read more.
Vehicle automation is driving the integration of advanced sensors and new applications that demand high-quality information, such as collaborative sensing for enhanced situational awareness. In this work, we considered a vehicular sensing scenario supported by 5G communications, in which vehicle sensor data need to be sent to edge computing resources with stringent latency constraints. To ensure low latency with the resources available, we propose an optimization framework that deploys User Plane Functions (UPFs) dynamically at the edge to minimize the number of network hops between the vehicles and them. The proposed framework relies on a practical Software-Defined-Networking (SDN)-based mechanism that allows seamless re-assignment of vehicles to UPFs while maintaining session and service continuity. We propose and evaluate different UPF allocation algorithms that reduce communications latency compared to static, random, and centralized deployment baselines. Our results demonstrated that the dynamic allocation of UPFs can support latency-critical applications that would be unfeasible otherwise. Full article
(This article belongs to the Special Issue Sensor Networks for Vehicular Communications)
►▼ Show Figures

Figure 1

Figure 1
<p>Architecture of the proposed solution.</p> Full article ">Figure 2
<p>Resulting interconnection network topology.</p> Full article ">Figure 3
<p>Histogram of the minimum distances (number of hops) between every pair of base stations in the resulting graph.</p> Full article ">Figure 4
<p>The 90th-percentile of the latency (number of hops) perceived by the UEs for a growing percentage of base stations containing a UPF.</p> Full article ">Figure 5
<p>Distribution of 4112 UEs in the 3000th time slot of the simulation. The red cross marks the location where 1 UPF is allocated using the K-means algorithm.</p> Full article ">Figure 6
<p>Distribution of 4112 UEs in the 3000th time slot of the simulation. The red crosses mark the location where 2 UPFs are allocated using the K-means algorithm.</p> Full article ">Figure 7
<p>Distribution of 4112 UEs in the 3000th time slot of the simulation. The red crosses mark the location where 3 UPFs are allocated using the K-means algorithm.</p> Full article ">Figure 8
<p>Average execution time for the different algorithms for a growing percentage of base stations containing a UPF.</p> Full article ">
19 pages, 513 KiB  
Article
Sensor-Based Human Activity Recognition Using Adaptive Class Hierarchy
by Kazuma Kondo and Tatsuhito Hasegawa
Sensors 2021, 21(22), 7743; https://doi.org/10.3390/s21227743 - 21 Nov 2021
Cited by 5 | Viewed by 3067
Abstract
In sensor-based human activity recognition, many methods based on convolutional neural networks (CNNs) have been proposed. In the typical CNN-based activity recognition model, each class is treated independently of others. However, actual activity classes often have hierarchical relationships. It is important to consider [...] Read more.
In sensor-based human activity recognition, many methods based on convolutional neural networks (CNNs) have been proposed. In the typical CNN-based activity recognition model, each class is treated independently of others. However, actual activity classes often have hierarchical relationships. It is important to consider an activity recognition model that uses the hierarchical relationship among classes to improve recognition performance. In image recognition, branch CNNs (B-CNNs) have been proposed for classification using class hierarchies. B-CNNs can easily perform classification using hand-crafted class hierarchies, but it is difficult to manually design an appropriate class hierarchy when the number of classes is large or there is little prior knowledge. Therefore, in our study, we propose a class hierarchy-adaptive B-CNN, which adds a method to the B-CNN for automatically constructing class hierarchies. Our method constructs the class hierarchy from training data automatically to effectively train the B-CNN without prior knowledge. We evaluated our method on several benchmark datasets for activity recognition. As a result, our method outperformed standard CNN models without considering the hierarchical relationship among classes. In addition, we confirmed that our method has performance comparable to a B-CNN model with a class hierarchy based on human prior knowledge. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Model overview of the related methods (<b>a</b>,<b>b</b>) and of our method (<b>c</b>). (<b>a</b>) Hierarchical Classification. (<b>b</b>) B-CNN [<a href="#B7-sensors-21-07743" class="html-bibr">7</a>]. (<b>c</b>) Class Hierarchy Adaptive B-CNN.</p> Full article ">Figure 2
<p>Overview of the class hierarchy construction method.</p> Full article ">Figure 3
<p>Model structure of the B-CNN with VGG16 as the base model.</p> Full article ">Figure 4
<p>Effect of varying the number of subjects used for training on accuracy.</p> Full article ">Figure 5
<p>Distribution of accuracy when attempting all possible patterns of class hierarchy in the HASC dataset.</p> Full article ">Figure 6
<p>Percentage of two classes merged at Level 2 of the top 1% class hierarchy.</p> Full article ">Figure 7
<p>Percentage of two classes merged in Level 2 of the lower 1% class hierarchy.</p> Full article ">Figure 8
<p>The number of combinations of class hierarchies of height 3 per original number of classes.</p> Full article ">
23 pages, 4806 KiB  
Review
Electrochemistry/Photoelectrochemistry-Based Immunosensing and Aptasensing of Carcinoembryonic Antigen
by Jingjing Jiang, Jili Xia, Yang Zang and Guowang Diao
Sensors 2021, 21(22), 7742; https://doi.org/10.3390/s21227742 - 21 Nov 2021
Cited by 11 | Viewed by 3721
Abstract
Recently, electrochemistry- and photoelectrochemistry-based biosensors have been regarded as powerful tools for trace monitoring of carcinoembryonic antigen (CEA) due to the fact of their intrinsic advantages (e.g., high sensitivity, excellent selectivity, small background, and low cost), which play an important role in early [...] Read more.
Recently, electrochemistry- and photoelectrochemistry-based biosensors have been regarded as powerful tools for trace monitoring of carcinoembryonic antigen (CEA) due to the fact of their intrinsic advantages (e.g., high sensitivity, excellent selectivity, small background, and low cost), which play an important role in early cancer screening and diagnosis and benefit people’s increasing demands for medical and health services. Thus, this mini-review will introduce the current trends in electrochemical and photoelectrochemical biosensors for CEA assay and classify them into two main categories according to the interactions between target and biorecognition elements: immunosensors and aptasensors. Some recent illustrative examples are summarized for interested readers, accompanied by simple descriptions of the related signaling strategies, advanced materials, and detection modes. Finally, the development prospects and challenges of future electrochemical and photoelectrochemical biosensors are considered. Full article
(This article belongs to the Section Biosensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of electrochemistry- and photoelectrochemistry-based immunosensing and aptasensing of CEA.</p> Full article ">Figure 2
<p>Schematic illustration of the preparation process for PtNPs@rGO@PS NSs and the fabrication of the electrochemical label-free immunosensor. Reprinted with permission from ref. [<a href="#B24-sensors-21-07742" class="html-bibr">24</a>]. Copyright 2020 American Chemical Society.</p> Full article ">Figure 3
<p>(<b>A</b>) The preparation procedures of Au@Pt DNs/NG/Cu<sup>2+</sup>-Ab<sub>2</sub>. (<b>B</b>) The fabrication process of the sandwich-type electrochemical immunosensor. Reprinted with permission from ref. [<a href="#B41-sensors-21-07742" class="html-bibr">41</a>]. Copyright 2018 Elsevier.</p> Full article ">Figure 4
<p>The preparation process of Au@Pd NDs/Fe<sup>2+</sup>-CS/PPy NTs and the schemata of the fabrication process of the working electrode for label-free immunosensors. Reprinted with permission from ref. [<a href="#B47-sensors-21-07742" class="html-bibr">47</a>]. Copyright 2018 Elsevier.</p> Full article ">Figure 5
<p>Scheme diagram of the DNA-gated MOF-based electrochemical biosensing platform of CEA. (<b>A</b>) Assembly procedure of MB@DNA/MOFs. (<b>B</b>) Target-triggered nicking endonuclease cleavage process. (<b>C</b>) Signal molecule release from MB@DNA/MOFs on the electrode. Reprinted with permission from ref. [<a href="#B74-sensors-21-07742" class="html-bibr">74</a>]. Copyright 2020 American Chemical Society.</p> Full article ">Figure 6
<p>Fabrication and detection mechanism of the PEC immunosensor. Reprinted with permission from ref. [<a href="#B94-sensors-21-07742" class="html-bibr">94</a>]. Copyright 2021 American Chemical Society.</p> Full article ">Figure 7
<p>Illustration of the PEC immunoassay for the detection of CEA. Reprinted with permission from ref. [<a href="#B103-sensors-21-07742" class="html-bibr">103</a>]. Copyright 2020 American Chemical Society.</p> Full article ">Figure 8
<p>Construction of the designed photoelectrochemical immunosensor for CEA monitoring based on the quenching of HCR-modulated Cu<sup>2+</sup> sources toward TiO<sub>2</sub>-sensitized DS-ZnCdS HNs. Reprinted with permission from ref. [<a href="#B105-sensors-21-07742" class="html-bibr">105</a>]. Copyright 2021 Elsevier.</p> Full article ">Figure 9
<p>Analytical principle of the salt bridge-connected electrochromic PEC immunosensor with DMM readout: (<b>A</b>) The sensing cell; (<b>B</b>) The electrochromic cell. Reprinted with permission from ref. [<a href="#B109-sensors-21-07742" class="html-bibr">109</a>]. Copyright 2020 Elsevier.</p> Full article ">Figure 10
<p>Schematic diagram of this proposed PEC biosensor for CEA determination. (<b>A</b>) Enzyme-free target cycling amplification strategy for generating S1; (<b>B</b>) Fabrication of the PEC “signal-off-on” biosensor based on ZnSe QDs/Au NPs and 3D DNA nanospheres. Reprinted with permission from ref. [<a href="#B124-sensors-21-07742" class="html-bibr">124</a>]. Copyright 2020 Elsevier.</p> Full article ">Figure 11
<p>Schematic illustration of the proposed PEC biosensor for CEA determination. Reprinted with permission from ref. [<a href="#B127-sensors-21-07742" class="html-bibr">127</a>]. Copyright 2021 Elsevier.</p> Full article ">Figure 12
<p>Construction of the up-conversion-mediated ratiometric PEC aptasensor for CEA detection. Reprinted with permission from ref. [<a href="#B129-sensors-21-07742" class="html-bibr">129</a>]. Copyright 2019 American Chemical Society.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-345
Previous Issue
Next Issue
Back to TopTop