[go: up one dir, main page]
Next Issue
Volume 19, May-2
Previous Issue
Volume 19, April-2
 
 
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 19
Issue 9
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 19, Issue 9 (May-1 2019) – 257 articles

Cover Story (view full-size image): The gas sensor market is growing fast, driven by many social, economic, and industrial factors. Optical gas sensors operating in the mid-infrared spectral region offer excellent performance for an increasing number of potential applications in healthcare, smart homes, and the automotive sector. With the emerging trend of miniaturization of optical devices, such sensors could be integrated into smartphones, watches, wearables, or medical devices, to widen their appeal to a much broader consumer application area. In this issue, we discuss major optical sensor technologies and architectures and present a path towards their miniaturization and monolithic integration with a focus on low-cost, low-power-consumption, and high volume manufacturability. 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:
15 pages, 4186 KiB  
Article
Novel Local Coding Algorithm for Finger Multimodal Feature Description and Recognition
by Shuyi Li, Haigang Zhang, Yihua Shi and Jinfeng Yang
Sensors 2019, 19(9), 2213; https://doi.org/10.3390/s19092213 - 13 May 2019
Cited by 24 | Viewed by 4185
Abstract
Recently, finger-based biometrics, including fingerprint (FP), finger-vein (FV) and finger-knuckle-print (FKP) with high convenience and user friendliness, have attracted much attention for personal identification. The features expression which is insensitive to illumination and pose variation are beneficial for finger trimodal recognition performance improvement. [...] Read more.
Recently, finger-based biometrics, including fingerprint (FP), finger-vein (FV) and finger-knuckle-print (FKP) with high convenience and user friendliness, have attracted much attention for personal identification. The features expression which is insensitive to illumination and pose variation are beneficial for finger trimodal recognition performance improvement. Therefore, exploring suitable method of reliable feature description is of great significance for developing finger-based biometric recognition system. In this paper, we first propose a correction approach for dealing with the pose inconsistency among the finger trimodal images, and then introduce a novel local coding-based feature expression method to further implement feature fusion of FP, FV, and FKP traits. First, for the coding scheme a bank of oriented Gabor filters is used for direction feature enhancement in finger images. Then, a generalized symmetric local graph structure (GSLGS) is developed to fully express the position and orientation relationships among neighborhood pixels. Experimental results on our own-built finger trimodal database show that the proposed coding-based approach achieves excellent performance in improving the matching accuracy and recognition efficiency. Full article
(This article belongs to the Special Issue Biometric Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Finger multimodal recognition process based on the Gabor generalized symmetric local graph structure (Gabor-GSLGS).</p> Full article ">Figure 2
<p>A finger trimodal image acquisition system. (<b>a</b>) the imaging schematic diagram; (<b>b</b>) a homemade image capture device; (<b>c</b>) a system interface of image acquisition.</p> Full article ">Figure 2 Cont.
<p>A finger trimodal image acquisition system. (<b>a</b>) the imaging schematic diagram; (<b>b</b>) a homemade image capture device; (<b>c</b>) a system interface of image acquisition.</p> Full article ">Figure 3
<p>Computing finger posture angle. (<b>a</b>) the edge line of the finger; (<b>b</b>) the coordinate extraction of the finger edge line; (<b>c</b>) finger rotation direction extraction.</p> Full article ">Figure 4
<p>Some corrected image samples after rotation.</p> Full article ">Figure 5
<p>The finger trimodal region of interest (ROI) images of four fingers. (<b>a</b>) fingerprint (FP) ROIs samples; (<b>b</b>) samples of finger-vein (FV) ROIs; (<b>c</b>) finger-knuckle-print (FKP) ROIs samples.</p> Full article ">Figure 6
<p>The enhanced images of the finger three modalities.</p> Full article ">Figure 7
<p>The design of the GSLGS operator (0° direction, <span class="html-italic">n</span> = 3).</p> Full article ">Figure 8
<p>The GSLGS operator (<span class="html-italic">k</span> = 4).</p> Full article ">Figure 9
<p>The coding process of GSLGS operator at 0° and 45°.</p> Full article ">Figure 10
<p>The fusion of finger trimodal features.</p> Full article ">Figure 11
<p>Receiver operating characteristic (ROC) of different <span class="html-italic">k</span>.</p> Full article ">Figure 12
<p>ROC of different neighborhoods in <span class="html-italic">M</span> = 6, 7, 8, 9.</p> Full article ">Figure 13
<p>ROC of different division blocks <span class="html-italic">M</span> in a 5 × 5 neighborhood.</p> Full article ">Figure 14
<p>Comparison results of different modal combinations. (<b>a</b>) ROC of unimodal recognition; (<b>b</b>) ROC of multimodal recognition.</p> Full article ">Figure 15
<p>Comparisons of different methods.</p> Full article ">
16 pages, 5852 KiB  
Article
Emotion Recognition from Multiband EEG Signals Using CapsNet
by Hao Chao, Liang Dong, Yongli Liu and Baoyun Lu
Sensors 2019, 19(9), 2212; https://doi.org/10.3390/s19092212 - 13 May 2019
Cited by 251 | Viewed by 12051
Abstract
Emotion recognition based on multi-channel electroencephalograph (EEG) signals is becoming increasingly attractive. However, the conventional methods ignore the spatial characteristics of EEG signals, which also contain salient information related to emotion states. In this paper, a deep learning framework based on a multiband [...] Read more.
Emotion recognition based on multi-channel electroencephalograph (EEG) signals is becoming increasingly attractive. However, the conventional methods ignore the spatial characteristics of EEG signals, which also contain salient information related to emotion states. In this paper, a deep learning framework based on a multiband feature matrix (MFM) and a capsule network (CapsNet) is proposed. In the framework, the frequency domain, spatial characteristics, and frequency band characteristics of the multi-channel EEG signals are combined to construct the MFM. Then, the CapsNet model is introduced to recognize emotion states according to the input MFM. Experiments conducted on the dataset for emotion analysis using EEG, physiological, and video signals (DEAP) indicate that the proposed method outperforms most of the common models. The experimental results demonstrate that the three characteristics contained in the MFM were complementary and the capsule network was more suitable for mining and utilizing the three correlation characteristics. Full article
(This article belongs to the Special Issue Novel Approaches to EEG Signal Processing)
►▼ Show Figures

Figure 1

Figure 1
<p>Three-dimensional space of the emotion model.</p> Full article ">Figure 2
<p>Images used for self-assessment: (<b>top</b>) Valence self-assessment manikin (SAM); (<b>middle</b>) Arousal SAM; (<b>bottom</b>) Dominance SAM.</p> Full article ">Figure 3
<p>International 10–20 system and 9 × 9 square matrix.</p> Full article ">Figure 4
<p>The mapping process of multiband feature matrix according to the raw electroencephalograph (EEG) signals of 32 channels.</p> Full article ">Figure 5
<p>(<b>a</b>) Information transfer between capsules; (<b>b</b>) Routing process.</p> Full article ">Figure 6
<p>Architecture of the capsule network (CapsNet)-based model. ReLU: rectified linear unit.</p> Full article ">Figure 7
<p>Accuracy of different models in three dimensions in the first step.</p> Full article ">Figure 8
<p>Accuracy of different models in three dimensions in the second step.</p> Full article ">Figure 9
<p>Comparison of emotion recognition results between CapsNet and comparison classifiers. CNN: convolutional neural network; k-NN: k-nearest neighbor; RDF: random decision forest; SVM: support vector machine.</p> Full article ">Figure 10
<p>Comparison of 2D-CNN and CapsNet using multiband feature matrices (MFMs).</p> Full article ">
16 pages, 1661 KiB  
Article
Decision-Making of Underwater Cooperative Confrontation Based on MODPSO
by Na Wei, Mingyong Liu and Weibin Cheng
Sensors 2019, 19(9), 2211; https://doi.org/10.3390/s19092211 - 13 May 2019
Cited by 11 | Viewed by 3712
Abstract
This paper proposes a multi-objective decision-making model for underwater countermeasures based on a multi-objective decision theory and solves it using the multi-objective discrete particle swarm optimization (MODPSO) algorithm. Existing decision-making models are based on fully allocated assignment without considering the weapon consumption and [...] Read more.
This paper proposes a multi-objective decision-making model for underwater countermeasures based on a multi-objective decision theory and solves it using the multi-objective discrete particle swarm optimization (MODPSO) algorithm. Existing decision-making models are based on fully allocated assignment without considering the weapon consumption and communication delay, which does not conform to the actual naval combat process. The minimum opponent residual threat probability and minimum own-weapon consumption are selected as two functions of the multi-objective decision-making model in this paper. Considering the impact of the communication delay, the multi-objective discrete particle swarm optimization (MODPSO) algorithm is proposed to obtain the optimal solution of the distribution scheme with different weapon consumptions. The algorithm adopts the natural number coding method, and the particle corresponds to the confrontation strategy. The simulation result shows that underwater communication delay impacts the decision-making selection. It verifies the effectiveness of the proposed model and the proposed multi-objective discrete particle swarm optimization algorithm. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Allocation in the process of underwater cooperative confrontation.</p> Full article ">Figure 2
<p>Task allocation instance.</p> Full article ">Figure 3
<p>Chart of the multi-objective discrete particle swarm optimization (MODPSO) algorithm.</p> Full article ">Figure 4
<p>Results of the two algorithms.</p> Full article ">Figure 5
<p>Distribution uniformity of Pareto optimal set of algorithms.</p> Full article ">Figure 6
<p>Target allocation scheme under different weapon consumption.</p> Full article ">Figure 7
<p>Weapon consumption under different communication delay influence factors.</p> Full article ">
22 pages, 2450 KiB  
Article
Differential Structure of Inductive Proximity Sensor
by Yi-Xin Guo, Cong Lai, Zhi-Biao Shao, Kai-Liang Xu and Ting Li
Sensors 2019, 19(9), 2210; https://doi.org/10.3390/s19092210 - 13 May 2019
Cited by 20 | Viewed by 8955
Abstract
The inductive proximity sensor (IPS) is applicable to displacement measurements in the aviation field due to its non-mechanical contact, safety, and durability. IPS can increase reliability of position detection and decrease maintenance cost of the system effectively in aircraft applications. Nevertheless, the specialty [...] Read more.
The inductive proximity sensor (IPS) is applicable to displacement measurements in the aviation field due to its non-mechanical contact, safety, and durability. IPS can increase reliability of position detection and decrease maintenance cost of the system effectively in aircraft applications. Nevertheless, the specialty in the aviation field proposes many restrictions and requirements on the application of IPS, including the temperature drift effect of the resistance component of the IPS sensing coil. Moreover, reliability requirements of aircrafts restrict the use of computational-intensive algorithms and avoid the use of process control components. Furthermore, the environment of airborne electronic equipment restricts measurements driven by large current and proposes strict requirements on emission tests of radio frequency (RF) energy. For these reasons, a differential structured IPS measurement method is proposed in this paper. This measurement method inherits the numerical separation of the resistance and inductance components of the IPS sensing coil to improve the temperature adaptation of the IPS. The computational complexity is decreased by combining the dimension-reduced look-up table method to prevent the use of process control components. The proposed differential structured IPS is equipped with a differential structure of distant and nearby sensing coils to increase the detection accuracy. The small electric current pulse excitation decreases the RF energy emission. Verification results demonstrate that the differential structured IPS realizes the numerical decoupling calculation of the vector impedance of the sensing coil by using 61 look-up table units. The measuring sensitivity increased from 135.5 least significant bits (LSB)/0.10 mm of a single-sensing-coil structured IPS to 1201.4 LSB/0.10 mm, and the linear approximation distance error decreased from 99.376 μm to −3.240 μm. The proposed differential structured IPS method has evident comparative advantages compared with similar measuring techniques. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Block diagram of single-sensor-coil structured inductive proximity sensor (IPS).</p> Full article ">Figure 2
<p>Relationships between vector impedance of the response circuit and environmental factors.</p> Full article ">Figure 3
<p>Block diagram of differential structured IPS.</p> Full article ">Figure 4
<p>Equivalent circuit of the nearby sensing coil.</p> Full article ">Figure 5
<p>Damped cases of the system determined by the value of <span class="html-italic">f</span>(<span class="html-italic">C</span><sub>pn</sub>).</p> Full article ">Figure 6
<p>Calibration of distance vs. <span class="html-italic">L</span>.</p> Full article ">Figure 7
<p>Response waveforms of the inertial system.</p> Full article ">Figure 8
<p>Response waveforms of the differential structured IPS.</p> Full article ">Figure 9
<p>Nonlinear relationship between differential response and ambient temperature.</p> Full article ">Figure 10
<p>Process of calibration and calculation.</p> Full article ">Figure 11
<p>Maximum variation range in the analog−digital converter (ADC) dynamic range.</p> Full article ">
15 pages, 2039 KiB  
Article
Instant Mercury Ion Detection in Industrial Waste Water with a Microchip Using Extended Gate Field-Effect Transistors and a Portable Device
by Revathi Sukesan, Yi-Ting Chen, Suman Shahim, Shin-Li Wang, Indu Sarangadharan and Yu-Lin Wang
Sensors 2019, 19(9), 2209; https://doi.org/10.3390/s19092209 - 13 May 2019
Cited by 14 | Viewed by 6367
Abstract
Mercury ion selective membrane (Hg-ISM) coated extended gate Field Effect transistors (ISM-FET) were used to manifest a novel methodology for ion-selective sensors based on FET’s, creating ultra-high sensitivity (−36 mV/log [Hg2+]) and outweighing ideal Nernst sensitivity limit (−29.58 mV/log [Hg2+ [...] Read more.
Mercury ion selective membrane (Hg-ISM) coated extended gate Field Effect transistors (ISM-FET) were used to manifest a novel methodology for ion-selective sensors based on FET’s, creating ultra-high sensitivity (−36 mV/log [Hg2+]) and outweighing ideal Nernst sensitivity limit (−29.58 mV/log [Hg2+]) for mercury ion. This highly enhanced sensitivity compared with the ion-selective electrode (ISE) (10−7 M) has reduced the limit of detection (10−13 M) of Hg2+ concentration’s magnitude to considerable orders irrespective of the pH of the test solution. Systematical investigation was carried out by modulating sensor design and bias voltage, revealing that higher sensitivity and a lower detection limit can be attained in an adequately stronger electric field. Our sensor has a limit of detection of 10−13 M which is two orders lower than Inductively Coupled Plasma Mass Spectrometry (ICP-MS), having a limit of detection of 10−11 M. The sensitivity and detection limit do not have axiomatic changes under the presence of high concentrations of interfering ions. The technology offers economic and consumer friendly water quality monitoring options intended for homes, offices and industries. Full article
(This article belongs to the Special Issue Potentiometric Bio/Chemical Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Schematic representation of extended gate Hg-ISMFET connected to the prototype. (<b>b</b>) Structural representation of extended gate Hg-ISMFET. (<b>c</b>) Real view image of sensor chip mounted on the portable measurement system connected to personal computer. (<b>d</b>) Current gain average verses time in air and in 0.02X PBS by extended gate Hg-ISMFET. (<b>e</b>) Gain leakage current of ISMFET in the presence of 10<sup>−8</sup> M mercury ions.</p> Full article ">Figure 2
<p>Characteristics of Mercury ion selective ISMFET (Hg-ISMFET). (<b>a</b>) Current gain signal of Hg-ISMFET for differing conductivity of test sample. (<b>b</b>) Current gain signal of Hg-ISMFET for differing pH of a test sample.</p> Full article ">Figure 3
<p>Sensor response; (<b>a</b>) Current gain response of extended gate Hg-ISMFET through time. (<b>b</b>) Current gain average verses different concentration of Hg<sup>2+</sup> prepared in 0.02X PBS by extended gate Hg-ISMFET (error bars obtained from multiple tests with <span class="html-italic">n</span> = 3).</p> Full article ">Figure 4
<p>Effect of gap distance between sensing and reference electrodes and applied Vg on current gain. (<b>a</b>)–(<b>f</b>) Current gain versus different gap distance for fixed Vg.</p> Full article ">Figure 5
<p>Mercury ion detection using extended gate Hg-ISMFET and sensitivity comparison. (<b>a</b>) Gain of Hg-ISMFET versus gate voltage (<b>b</b>) Gain of Hg-ISMFET versus log mercury ion concentration. (<b>c</b>) Effective gate voltage obtained with respect to log mercury ion concentration. (<b>d</b>) Schematic representation of the capacitive model of the sensor.</p> Full article ">Figure 6
<p>Selectivity characteristics of extended gate Hg-ISMFET sensor measured at 1 V Vg. (<b>a</b>) Gain versus heavy metal ion concentration graph of Hg-ISMFET, using fixed interference method. (<b>b</b>) Gain versus heavy metal ion concentration graph of Hg-ISMFET, using separate solution method (error bars obtained from multiple tests with <span class="html-italic">n</span> = 3).</p> Full article ">
20 pages, 1511 KiB  
Article
A Novel FEM Based T-S Fuzzy Particle Filtering for Bearings-Only Maneuvering Target Tracking
by Xiaoli Wang, Liangqun Li and Weixin Xie
Sensors 2019, 19(9), 2208; https://doi.org/10.3390/s19092208 - 13 May 2019
Cited by 7 | Viewed by 3208
Abstract
In this paper, we propose a novel fuzzy expectation maximization (FEM) based Takagi-Sugeno (T-S) fuzzy particle filtering (FEMTS-PF) algorithm for a passive sensor system. In order to incorporate target spatial-temporal information into particle filtering, we introduce a T-S fuzzy modeling algorithm, in which [...] Read more.
In this paper, we propose a novel fuzzy expectation maximization (FEM) based Takagi-Sugeno (T-S) fuzzy particle filtering (FEMTS-PF) algorithm for a passive sensor system. In order to incorporate target spatial-temporal information into particle filtering, we introduce a T-S fuzzy modeling algorithm, in which an improved FEM approach is proposed to adaptively identify the premise parameters, and the model probability is adjusted by the premise membership functions. In the proposed FEM, the fuzzy parameter is derived by the fuzzy C-regressive model clustering method based on entropy and spatial-temporal characteristics, which can avoid the subjective influence caused by the artificial setting of the initial value when compared to the traditional FEM. Furthermore, using the proposed T-S fuzzy model, the algorithm samples particles, which can effectively reduce the particle degradation phenomenon and the parallel filtering, can realize the real-time performance of the algorithm. Finally, the results of the proposed algorithm are evaluated and compared to several existing filtering algorithms through a series of Monte Carlo simulations. The simulation results demonstrate that the proposed algorithm is more precise, robust and that it even has a faster convergence rate than the interacting multiple model unscented Kalman filter (IMMUKF), interacting multiple model extended Kalman filter (IMMEKF) and interacting multiple model Rao-Blackwellized particle filter (IMMRBPF). Full article
(This article belongs to the Special Issue Multi-Sensor Systems for Positioning and Navigation)
►▼ Show Figures

Figure 1

Figure 1
<p>Block diagram of the T-S fuzzy modeling method.</p> Full article ">Figure 2
<p>Performance comparison of the IMMUKF (triangle sign (<math display="inline"><semantics> <mi mathvariant="sans-serif">Δ</mi> </semantics></math>)), IMMEKF (plus sign (<math display="inline"><semantics> <mo>+</mo> </semantics></math>)), IMMRBPF (dotted line (<math display="inline"><semantics> <mrow> <mo>−</mo> <mo>−</mo> </mrow> </semantics></math>)), FPF (solid line (<math display="inline"><semantics> <mo>−</mo> </semantics></math>)), traditional FEMTS-PF (circle (<math display="inline"><semantics> <mi>o</mi> </semantics></math>)) and proposed FEMTS-PF (star sign (<math display="inline"><semantics> <mo>∗</mo> </semantics></math>)). (<b>a</b>) The target trajectory of the proposed algorithm and the actual position; (<b>b</b>) position root-mean-square error (RMSE); (<b>c</b>) <span class="html-italic">X</span>-axis RMSE; and (<b>d</b>) <span class="html-italic">Y</span>-axis RMSE.</p> Full article ">Figure 3
<p>The position RMSE for different process noises. (<b>a</b>) The standard deviation of the process noise is 0.02; and (<b>b</b>) the standard deviation of the process noise is 0.04.</p> Full article ">Figure 4
<p>The position RMSE for different measurement noises. (<b>a</b>) The standard deviation of the measurement noise is 0.002; and (<b>b</b>) the standard deviation of the measurement noise is 0.005.</p> Full article ">Figure 5
<p>Performance comparison of the IMMUKF (triangle sign (<math display="inline"><semantics> <mi mathvariant="sans-serif">Δ</mi> </semantics></math>)), IMMEKF (plus sign (<math display="inline"><semantics> <mo>+</mo> </semantics></math>)), IMMRBPF (dotted line (<math display="inline"><semantics> <mrow> <mo>−</mo> <mo>−</mo> </mrow> </semantics></math>)), traditional FEMTS-PF (circle (<math display="inline"><semantics> <mi>o</mi> </semantics></math>)) and FEMTS-PF (star sign (<math display="inline"><semantics> <mo>*</mo> </semantics></math>)). (<b>a</b>) The target trajectory of the proposed algorithm and the actual position; (<b>b</b>) position root-mean-square error (RMSE); (<b>c</b>) <span class="html-italic">X</span>-axis RMSE; and (<b>d</b>) <span class="html-italic">Y</span>-axis RMSE.</p> Full article ">
15 pages, 5786 KiB  
Article
Optimal Design of Angular Displacement Sensor with Shared Magnetic Field Based on the Magnetic Equivalent Loop Method
by Pinggui Luo, Qifu Tang and Huan Jing
Sensors 2019, 19(9), 2207; https://doi.org/10.3390/s19092207 - 13 May 2019
Cited by 5 | Viewed by 3710
Abstract
Angular displacement sensor with shared magnetic field has strong environmental adaptability and high measurement accuracy. However, its 3-D structure is multi-pole double-layer structure, using time stepping finite element method (TSFEM) to optimize the structure is time-consuming and uneconomical. Therefore, a magnetic equivalent loop [...] Read more.
Angular displacement sensor with shared magnetic field has strong environmental adaptability and high measurement accuracy. However, its 3-D structure is multi-pole double-layer structure, using time stepping finite element method (TSFEM) to optimize the structure is time-consuming and uneconomical. Therefore, a magnetic equivalent loop method (MELM) is proposed to simplify the optimal design of sensors. By reasonably setting the node position, the mechanical structure parameters, winding coefficients and input voltage of the sensor are integrated into a mathematical model to calculate of the induced voltage. The calculation results are compared with the simulation results, and a sensor prototype is made to test the optimized effect of the MELM. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>The structure of sensor: (<b>a</b>) Stator; and (<b>b</b>) rotor.</p> Full article ">Figure 2
<p>The simplified structure and size symbols: (<b>a</b>) Stator; and (<b>b</b>) rotor.</p> Full article ">Figure 3
<p>The magnetic equivalent loop method of sensor: (<b>a</b>) The magnetic potential of each node; and (<b>b</b>) the permeability between the nodes.</p> Full article ">Figure 4
<p>Decomposition diagram of permeability at each node of rotor tooth: (<b>a</b>) Decomposition diagram of <math display="inline"><semantics> <mrow> <msubsup> <mi>P</mi> <mi>j</mi> <mrow> <mi>r</mi> <mi>r</mi> </mrow> </msubsup> </mrow> </semantics></math>; (<b>b</b>) decomposition diagram of <math display="inline"><semantics> <mrow> <msubsup> <mi>P</mi> <mi>j</mi> <mi>r</mi> </msubsup> </mrow> </semantics></math>; and (<b>c</b>) decomposition diagram of <math display="inline"><semantics> <mrow> <msubsup> <mi>P</mi> <mi>j</mi> <mrow> <mi>r</mi> <mi>t</mi> </mrow> </msubsup> </mrow> </semantics></math>.</p> Full article ">Figure 5
<p>The mesh diagram of the simulation sensor: (<b>a</b>) Rotor X and stator; and (<b>b</b>) rotor Y and stator.</p> Full article ">Figure 6
<p>Induced voltage waveform of the inner rotor by FEM and MELM under two-phase Excitation: (<b>a</b>) Rotor X; and (<b>b</b>) rotor Y.</p> Full article ">Figure 7
<p>The values of <math display="inline"><semantics> <mrow> <msub> <mo>Δ</mo> <mi>k</mi> </msub> </mrow> </semantics></math> of rotor X and rotor Y.</p> Full article ">Figure 8
<p>The relationship between VAA, VADC and trapezoidal groove clearance <math display="inline"><semantics> <mrow> <msub> <mi>θ</mi> <mn>6</mn> </msub> </mrow> </semantics></math>: (<b>a</b>) VAA; and (<b>b</b>) VADC.</p> Full article ">Figure 9
<p>The relationship between VAA, VADC and trapezoidal groove height <math display="inline"><semantics> <mrow> <msub> <mi>r</mi> <mn>6</mn> </msub> <mo>−</mo> <msub> <mi>r</mi> <mn>5</mn> </msub> </mrow> </semantics></math>: (<b>a</b>) VAA; and (<b>b</b>) VADC.</p> Full article ">Figure 10
<p>The relationship between VAA, VADC, and stator teeth width <math display="inline"><semantics> <mrow> <msub> <mi>d</mi> <mn>2</mn> </msub> </mrow> </semantics></math>: (<b>a</b>) VAA; and (<b>b</b>) VADC.</p> Full article ">Figure 11
<p>The relationship between VAA, VADC and air-gap length <math display="inline"><semantics> <mrow> <msub> <mi>d</mi> <mrow> <mi>r</mi> <mi>s</mi> </mrow> </msub> </mrow> </semantics></math>: (<b>a</b>) VAA; and (<b>b</b>) VADC.</p> Full article ">Figure 12
<p>Experimental system: (<b>a</b>) Experimental table; and (<b>b</b>) sensor prototype; (<b>c</b>) stator; (<b>d</b>) rotor.</p> Full article ">Figure 13
<p>One period measurement error of rotor X at 0.36 r/min.</p> Full article ">Figure 14
<p>One period measurement error of rotor Y at 0.36 r/min.</p> Full article ">Figure 15
<p>Full range measurement error of rotor X at 22 r/min.</p> Full article ">Figure 16
<p>Full range measurement error of rotor Y at 22 r/min.</p> Full article ">
34 pages, 1765 KiB  
Article
Smarter Traffic Prediction Using Big Data, In-Memory Computing, Deep Learning and GPUs
by Muhammad Aqib, Rashid Mehmood, Ahmed Alzahrani, Iyad Katib, Aiiad Albeshri and Saleh M. Altowaijri
Sensors 2019, 19(9), 2206; https://doi.org/10.3390/s19092206 - 13 May 2019
Cited by 96 | Viewed by 8356
Abstract
Road transportation is the backbone of modern economies, albeit it annually costs 1.25 million deaths and trillions of dollars to the global economy, and damages public health and the environment. Deep learning is among the leading-edge methods used for transportation-related predictions, however, the [...] Read more.
Road transportation is the backbone of modern economies, albeit it annually costs 1.25 million deaths and trillions of dollars to the global economy, and damages public health and the environment. Deep learning is among the leading-edge methods used for transportation-related predictions, however, the existing works are in their infancy, and fall short in multiple respects, including the use of datasets with limited sizes and scopes, and insufficient depth of the deep learning studies. This paper provides a novel and comprehensive approach toward large-scale, faster, and real-time traffic prediction by bringing four complementary cutting-edge technologies together: big data, deep learning, in-memory computing, and Graphics Processing Units (GPUs). We trained deep networks using over 11 years of data provided by the California Department of Transportation (Caltrans), the largest dataset that has been used in deep learning studies. Several combinations of the input attributes of the data along with various network configurations of the deep learning models were investigated for training and prediction purposes. The use of the pre-trained model for real-time prediction was explored. The paper contributes novel deep learning models, algorithms, implementation, analytics methodology, and software tool for smart cities, big data, high performance computing, and their convergence. Full article
(This article belongs to the Section State-of-the-Art Sensors Technologies)
►▼ Show Figures

Figure 1

Figure 1
<p>Prediction workflow.</p> Full article ">Figure 2
<p>A depiction of vehicles data collected from PeMS (24 h) (November 2017).</p> Full article ">Figure 3
<p>Architecture of a deep neural network model with one input, one output and <span class="html-italic">k</span> hidden layers.</p> Full article ">Figure 4
<p>Traffic flow: Actual and predicted values for a single VDS.</p> Full article ">Figure 5
<p>Traffic flow: Sctual and predicted values for all 26 VDS.</p> Full article ">Figure 6
<p>Traffic flow: Actual and predicted values (average of all VDS) (48 h).</p> Full article ">Figure 7
<p>Traffic flow: Actual and predicted values (total of all VDS) (48 h).</p> Full article ">Figure 8
<p>Traffic Flow: MAE, MAPE, AND RMSE Evaluation Metrics.</p> Full article ">Figure 9
<p>Traffic speed: Actual and predicted values for a single VDS.</p> Full article ">Figure 10
<p>Traffic speed: Actual and predicted values (all VDS) (peak hour: 26 June 2017, 16:00–17:00).</p> Full article ">Figure 11
<p>Traffic speed: Actual and predicted values (average of all VDS) (29 and 30 April 2017).</p> Full article ">Figure 12
<p>Traffic speed: MAE, MAPE, and RMSE evaluation metrics.</p> Full article ">Figure 13
<p>Traffic occupancy: Actual and predicted values for a single VDS.</p> Full article ">Figure 14
<p>Traffic occupancy: Actual and predicted values (all 26 VDS) (peak hour).</p> Full article ">Figure 15
<p>Traffic occupancy: Actual and predicted values (average of all VDS) (48 h).</p> Full article ">Figure 16
<p>Traffic occupancy: MAE, MAPE, and RMSE evaluation metrics.</p> Full article ">Figure 17
<p>MAPE calculated by comparing actual and predicted vehicles flow on weekends.</p> Full article ">Figure 18
<p>Comparison of original and predicted flow values on weekends.</p> Full article ">Figure 19
<p>MAPE calculated by comparing actual and predicted vehicles flow on morning peak hours.</p> Full article ">Figure 20
<p>Comparison of original and predicted flow values on morning peak hours.</p> Full article ">Figure 21
<p>MAPE calculated by comparing actual and predicted vehicles flow on evening peak hours.</p> Full article ">Figure 22
<p>Comparison of original and predicted flow values on evening peak hours.</p> Full article ">Figure 23
<p>Model execution time while making predictions using the pre-trained models.</p> Full article ">Figure 24
<p>Comparison of deep model execution time on CPUs and GPUs.</p> Full article ">
16 pages, 2801 KiB  
Article
An Evaluation of Gearbox Condition Monitoring Using Infrared Thermal Images Applied with Convolutional Neural Networks
by Yongbo Li, James Xi Gu, Dong Zhen, Minqiang Xu and Andrew Ball
Sensors 2019, 19(9), 2205; https://doi.org/10.3390/s19092205 - 13 May 2019
Cited by 43 | Viewed by 6154
Abstract
As an important machine component, the gearbox is widely used in industry for power transmission. Condition monitoring (CM) of a gearbox is critical to provide timely information for undertaking necessary maintenance actions. Massive research efforts have been made in the last two decades [...] Read more.
As an important machine component, the gearbox is widely used in industry for power transmission. Condition monitoring (CM) of a gearbox is critical to provide timely information for undertaking necessary maintenance actions. Massive research efforts have been made in the last two decades to develop vibration-based techniques. However, vibration-based methods usually include several inherent shortages including contact measurement, localized information, noise contamination, and high computation costs, making it difficult to be a cost-effective CM technique. In this paper, infrared thermal (IRT) images, which can contain information covering a large area and acquired remotely, are based on developing a cost-effective CM method. Moreover, a convolutional neural network (CNN) is employed to automatically process the raw IRT images for attaining more comprehensive feature parameters, which avoids the deficiency of incomplete information caused by various feature-extraction methods in vibration analysis. Thus, an IRT–CNN method is developed to achieve online remote monitoring of a gearbox. The performance evaluation based on a bevel gearbox shows that the proposed method can achieve nearly 100% correctness in identifying several common gear faults such as tooth pitting, cracks, and breakages and their compounds. It is also especially robust to ambient temperature changes. In addition, IRT also significantly outperforms its vibration-based counterparts. Full article
(This article belongs to the Special Issue Sensors Fusion in Non-Destructive Testing Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Architecture of the convolutional neural networks.</p> Full article ">Figure 2
<p>SoftMax regression model.</p> Full article ">Figure 3
<p>Flowchart of the proposed IRT–CNN method for fault diagnosis of gearbox: (<b>a</b>) illustrative diagrams, and (<b>b</b>) flowchart of training process.</p> Full article ">Figure 4
<p>(<b>a</b>) The machinery fault simulator system, (<b>b</b>) the schematic diagram of the test rig and thermal camera setup.</p> Full article ">Figure 5
<p>The designed gear faults: (<b>a</b>) Pitting in the driving tooth; (<b>b</b>) A broken tooth of driving gear; (<b>c</b>) A missing tooth of driving gear; (<b>d</b>) A cracked tooth of follower gear.</p> Full article ">Figure 6
<p>IRT images extracted from an infrared thermal video: (<b>a</b>) marked meshing area for reference (<b>b</b>) the raw IRT images.</p> Full article ">Figure 7
<p>Thermal images under eight gears with different health conditions: (<b>a</b>) healthy condition (<b>b</b>) pitting in the driving tooth (<b>c</b>) a broken tooth in the driving gear (<b>d</b>) a missing tooth in the driving gear (<b>e</b>) crack in the follower tooth (<b>f</b>) pitting in the driving gear with crack in the follower tooth pitting teeth (<b>g</b>) broken tooth in the driving gear with crack in the follower tooth (<b>h</b>) missing tooth in the driving tooth with crack in the follower tooth.</p> Full article ">Figure 8
<p>The classification results of the IRT–CNN method.</p> Full article ">Figure 9
<p>The classification results of the IRT–CNN trained by different percentages of data samples.</p> Full article ">Figure 10
<p>The vibration signals for eight gear-health conditions: (<b>a</b>) healthy condition (<b>b</b>) pitting in the driving tooth (<b>c</b>) a broken tooth in the driving gear (<b>d</b>) a missing tooth in the driving gear (<b>e</b>) crack in the follower tooth (<b>f</b>) pitting in the driving gear with crack in the follower tooth (<b>g</b>) broken tooth in the driving gear with crack in the follower tooth (<b>h</b>) missing tooth in the driving tooth with crack in the follower tooth.</p> Full article ">Figure 11
<p>Envelope spectra for eight gear-health conditions: (<b>a</b>) healthy condition (<b>b</b>) pitting in the driving tooth (<b>c</b>) a broken tooth in the driving gear (<b>d</b>) a missing tooth in the driving gear (<b>e</b>) crack in the follower tooth (<b>f</b>) pitting in the driving gear with crack in the follower tooth (<b>g</b>) broken tooth in the driving gear with crack in the follower tooth (<b>h</b>) missing tooth in the driving tooth with crack in the follower tooth.</p> Full article ">Figure 12
<p>The classification results using Vib-CNN method.</p> Full article ">Figure 13
<p>Projections of the features using PCA with different source signals: (<b>a</b>) IRT –CNN method (<b>b</b>) Vib-CNN method.</p> Full article ">Figure 14
<p>Training and testing performance curves in terms of overall accuracy.</p> Full article ">Figure 15
<p>IRT images under normal condition of gearbox in six temperature ranges: (<b>a</b>) 36 °C–39 °C (<b>b</b>) 39 °C–42 °C (<b>c</b>) 42 °C–45 °C (<b>d</b>) 45 °C–48 °C (<b>e</b>) 48 °C–51 °C (<b>f</b>) 51 °C–54 °C.</p> Full article ">Figure 16
<p>Diagnosis results for each temperature range using proposed IRT–CNN method.</p> Full article ">Figure 17
<p>Confusion matrix results with a temperature range of 39 °C–42 °C.</p> Full article ">
15 pages, 2381 KiB  
Article
The Effect of Treadmill Walking on Gait and Upper Trunk through Linear and Nonlinear Analysis Methods
by Liang Shi, Feng Duan, Yikang Yang and Zhe Sun
Sensors 2019, 19(9), 2204; https://doi.org/10.3390/s19092204 - 13 May 2019
Cited by 35 | Viewed by 6485
Abstract
Treadmills are widely used to recover walking function in the rehabilitation field for those patients with gait disorders. Nevertheless, the ultimate goal of walking function recovery is to walk on the ground rather than on the treadmill. This study aims to determine the [...] Read more.
Treadmills are widely used to recover walking function in the rehabilitation field for those patients with gait disorders. Nevertheless, the ultimate goal of walking function recovery is to walk on the ground rather than on the treadmill. This study aims to determine the effect of treadmill walking on gait and upper trunk movement characteristics using wearable sensors. Eight healthy male subjects are recruited to perform 420-m straight overground walking (OW) and 5 min treadmill walking (TW), wearing 3 inertial measurement units and a pair of insole sensors. In addition to common linear features, nonlinear features, which contains sample entropy, maximal Lyapunov exponent and fractal dynamic of stride intervals (detrended fluctuation analysis), are used to compare the difference between TW and OW condition. Canonical correlation analysis is also used to indicate the correlation between upper trunk movement characteristics and gait features in the aspects of spatiotemporal parameters and gait dynamic features. The experimental results show that the treadmill can cause a shorter stride length, less stride time and worsen long-range correlation of stride intervals. And the treadmill can significantly increase the stability for both gait and upper trunk, while it can significantly reduce gait regularity during swing phase. Canonical correlation analysis results show that treadmill can reduce the correlation between gait and upper trunk features. One possible interpretation of these results is that people tend to walk more cautiously to prevent the risk of falling and neglect the coordination between gait and upper trunk when walking on the treadmill. This study can provide fundamental insightful information about the effect of treadmill walking on gait and upper trunk to support future similar studies. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Wearable sensors used in this experiment, including three inertial measurement unit (IMU) and a pair of insole sensors. Two IMU are attached to feet with elastic belt in the position that is closed to toe (IMU Feet). The other is attached to the low back, where it is closed to L3 region (IMU Lumbar). A pair of insole sensors are put into subject’s own shoes and the data logger of insole sensors is fastened in the subject’s body. In addition, mediolateral (ML), vertical (V), anterior-posterior (AP), Pitch, Roll and Yaw directions are shown. (<b>b</b>) Representation of three IMU with elastic belt. (<b>c</b>) A pair of insole sensors (Novel, Germany). (<b>d</b>) A diagram of human body analysis using insole sensors and IMU.</p> Full article ">Figure 2
<p>(<b>a</b>) A diagram of entire gait period for left and right foot, including swing phase and stance phase. (<b>b</b>) The Plantar pressure during entire gait period for analyzing gait features of stance phase. (<b>c</b>) Representation of center of pressure (COP) path cyclogram. CISP, ML, V and AP respectively denote cyclogram intersection point, mediolateral, vertical and anterior-posterior. During stance phase, COP position moves forward under support foot. When support foot becomes the other foot, COP position moves from one foot to the other. (<b>d</b>) The position of COP point, the position of maximum pressure point and C-M distance means the average distance between these two points. (<b>e</b>) IMU is attached to foot for analyzing gait features of swing phase. (<b>f</b>) A diagram of upper trunk movement degree, stability and regularity when walking. <math display="inline"><semantics> <mi>θ</mi> </semantics></math> means upper trunk movement degree in pitch direction.</p> Full article ">Figure 3
<p>(<b>a</b>) Contrast of spatiotemporal parameters between TW and OW condition. (<b>b</b>) Contrast of stride length. (<b>c</b>) Contrast of average stride time. (<b>d</b>) Contrast of CV of stride time. (<b>e</b>) Contrast of fractal dynamic of stride intervals. (<b>f</b>) Contrast of COP path efficiency.</p> Full article ">Figure 4
<p>Comparison of gait features between TW and OW condition: (<b>a</b>) Sample entropy of foot acceleration during swing phase in V direction. (<b>b</b>) The RMS of foot acceleration during swing phase in V direction. (<b>c</b>) The maximal Lyapunov exponent of COP position during stance phase in ML and AP directions.</p> Full article ">Figure 5
<p>Comparison of upper trunk stability between TW and OW condition: (<b>a</b>) The maximal Lyapunov exponent of lumbar acceleration in ML, V and AP directions. (<b>b</b>) The RMS of lumbar acceleration in ML, V and AP directions.</p> Full article ">
14 pages, 2294 KiB  
Article
Directional Forgetting for Stable Co-Adaptation in Myoelectric Control
by Dennis Yeung, Dario Farina and Ivan Vujaklija
Sensors 2019, 19(9), 2203; https://doi.org/10.3390/s19092203 - 13 May 2019
Cited by 14 | Viewed by 4326
Abstract
Conventional myoelectric controllers provide a mapping between electromyographic signals and prosthetic functions. However, due to a number of instabilities continuously challenging this process, an initial mapping may require an extended calibration phase with long periods of user-training in order to ensure satisfactory performance. [...] Read more.
Conventional myoelectric controllers provide a mapping between electromyographic signals and prosthetic functions. However, due to a number of instabilities continuously challenging this process, an initial mapping may require an extended calibration phase with long periods of user-training in order to ensure satisfactory performance. Recently, studies on co-adaptation have highlighted the benefits of concurrent user learning and machine adaptation where systems can cope with deficiencies in the initial model by learning from newly acquired data. However, the success remains highly dependent on careful weighting of these new data. In this study, we proposed a function driven directional forgetting approach to the recursive least-squares algorithm as opposed to the classic exponential forgetting scheme. By only discounting past information in the same direction of the new data, local corrections to the mapping would induce less distortion to other regions. To validate the approach, subjects performed a set of real-time myoelectric tasks over a range of forgetting factors. Results show that directional forgetting with a forgetting factor of 0.995 outperformed exponential forgetting as well as unassisted user learning. Moreover, myoelectric control remained stable after adaptation with directional forgetting over a range of forgetting factors. These results indicate that a directional approach to discounting past training data can improve performance and alleviate sensitivities to parameter selection in recursive adaptation algorithms. Full article
(This article belongs to the Special Issue Assistance Robotics and Biosensors 2019)
►▼ Show Figures

Figure 1

Figure 1
<p>Experimental setup showing electrode position and visual feedback of the virtual task space. The red cursor is controlled via the myoelectric interface while the pink circle represents the target.</p> Full article ">Figure 2
<p>Online estimation and adaptation schematic. During adaptation runs, online adaptation with RLS was triggered if the current target was not reached within 5 s. The dotted diagonal line striking through the controller block (Regression) indicates conditional parameter update.</p> Full article ">Figure 3
<p>Flow diagram of the experimental procedure. The adaptation settings that were tested include: RLS-DF with <math display="inline"><semantics> <mrow> <mi>λ</mi> <mo>=</mo> <mn>0.995</mn> <mo>,</mo> <mn>0.97</mn> <mo>,</mo> <mn>0.95</mn> </mrow> </semantics></math>, and 0.93, and RLS-EF with <math display="inline"><semantics> <mrow> <mi>λ</mi> <mo>=</mo> <mn>0.995</mn> </mrow> </semantics></math>. The order of adaptation settings tested was randomised for each subject to prevent biasing of results.</p> Full article ">Figure 4
<p>Results from the evaluation of the five subjects. * indicates statistical significance detected with Dunn–Bonferroni pairwise testing.</p> Full article ">Figure 5
<p>(<b>a</b>) Completion ratio of evaluation runs. Runs involving machine and user co-adaptation, regardless of algorithm or forgetting factor, had, on average, higher target hit rates compared to initial evaluations with the batch-trained model (Baseline). (<b>b</b>) Averaged dot products between normalised row vectors of weights from batch-trained models and their online-adapted versions. A value of 1 indicates no change to the sensitivities of the mapping while lower values represent larger changes to the model during machine adaptation.</p> Full article ">Figure 6
<p>Cursor trajectories of adaptation and evaluation runs with RLS-DF from Subject 4. Green circles represent targets that were successfully reached, orange circles represent targets that were hit but dwell time was insufficient and red circles represent targets that were not hit within the time limit. Purple circles are only present in adaptation runs and represent targets that were reached with the aid of machine adaptation. Though some forgetting factors performed better than others, it can be observed that the solution space is still navigable after adaptation regardless of <math display="inline"><semantics> <mi>λ</mi> </semantics></math>.</p> Full article ">
10 pages, 2588 KiB  
Article
Miniature Diamond-Based Fiber Optic Pressure Sensor with Dual Polymer-Ceramic Adhesives
by Hyungdae Bae, Ayush Giri, Oluwafikunwa Kolawole, Amin Azimi, Aaron Jackson and Gary Harris
Sensors 2019, 19(9), 2202; https://doi.org/10.3390/s19092202 - 13 May 2019
Cited by 6 | Viewed by 4735
Abstract
Diamond is a good candidate for harsh environment sensing due to its high melting temperature, Young’s modulus, and thermal conductivity. A sensor made of diamond will be even more promising when combined with some advantages of optical sensing (i.e., EMI inertness, high temperature [...] Read more.
Diamond is a good candidate for harsh environment sensing due to its high melting temperature, Young’s modulus, and thermal conductivity. A sensor made of diamond will be even more promising when combined with some advantages of optical sensing (i.e., EMI inertness, high temperature operation, and miniaturization). We present a miniature diamond-based fiber optic pressure sensor fabricated using dual polymer-ceramic adhesives. The UV curable polymer and the heat-curing ceramic adhesive are employed for easy and reliable optical fiber mounting. The usage of the two different adhesives considerably improves the manufacturability and linearity of the sensor, while significantly decreasing the error from the temperature cross-sensitivity. Experimental study shows that the sensor exhibits good linearity over a pressure range of 2.0–9.5 psi with a sensitivity of 18.5 nm/psi (R2 = 0.9979). Around 275 °C of working temperature was achieved by using polymer/ceramic dual adhesives. The sensor can benefit many fronts that require miniature, low-cost, and high-accuracy sensors including biomedical and industrial applications. With an added antioxidation layer on the diamond diaphragm, the sensor can also be applied for harsh environment applications due to the high melting temperature and Young’s modulus of the material. Full article
(This article belongs to the Section Sensor Materials)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) A scanning electron micrograph of chemical vapor deposition (CVD) diamond film and (<b>b</b>) schematic of the diamond-based pressure sensor with a polymer-ceramic hybrid adhesive; (<b>c</b>) microscopic image of a fabricated sensor.</p> Full article ">Figure 2
<p>The fabrication process of the diamond-based pressure sensor (<b>a</b>–<b>f</b>).</p> Full article ">Figure 3
<p>Experiment arrangement for pressure and temperature calibration.</p> Full article ">Figure 4
<p>(<b>a</b>) Experiment arrangement for pressure and temperature calibration; (<b>b</b>) fast Fourier transform (FFT) result from the wavenumber spectrum of the sensor.</p> Full article ">Figure 5
<p>Pressure calibration curves for the sensor with only UV curable adhesive vs. with UV curable and ceramic adhesive at 24.5 °C.</p> Full article ">Figure 6
<p>(<b>a</b>) Pressure calibration curves at five different temperature and (<b>b</b>) a temperature calibration curve at 2 psi.</p> Full article ">Figure 7
<p>A temperature calibration curve (<b>a</b>) at 2 psi with a range of 25 to 65 °C and (<b>b</b>) at atmospheric pressure with a range of 25 to 325 °C.</p> Full article ">
15 pages, 2372 KiB  
Article
Combined Kalman Filter and Multifeature Fusion Siamese Network for Real-Time Visual Tracking
by Lijun Zhou and Jianlin Zhang
Sensors 2019, 19(9), 2201; https://doi.org/10.3390/s19092201 - 13 May 2019
Cited by 12 | Viewed by 4922
Abstract
SiamFC has a simple network structure and can be pretrained offline on a large data set, so it has attracted the attention of many researchers. It has no online learning process at all. Hence, there are no good solutions for some complex tracking [...] Read more.
SiamFC has a simple network structure and can be pretrained offline on a large data set, so it has attracted the attention of many researchers. It has no online learning process at all. Hence, there are no good solutions for some complex tracking scenarios such as occlusion and large target deformation. For this problem, we propose a method using the Kalman filter method and fusion multiresolution features and get multiple response scores. The Kalman filter acquires the target’s trajectory information, which is used to process complex tracking scenes and to change the selection method of the search area. This also enables our tracker to stably track fast moving targets.The introduction of the Kalman filter compensates for the shortcomings that SiamFC can only track offline, and the tracking network has an online learning process. The fusion of multiresolution features to obtain multiple response scores map helps the tracker to obtain robust features that can be adapted to a variety of tracking targets. Our proposed method has reached the state-of-the-art in testing on five data sets and can be run in real time (40 fps), including OTB2013, OTB2015, OTB50, VOT2015 and VOT 2016. Full article
(This article belongs to the Special Issue Intelligent Signal Processing, Data Science and the IoT World)
►▼ Show Figures

Figure 1

Figure 1
<p>The architecture of the proposed Kalman–Siam network. It consists of a Kalman filter motion trajectory estimation module and a tracking network. The input of the motion trajectory prediction module is an image of the current frame and the previous frame, which predicts the position of the target at the current frame and cropping the search area. In the tracking network module, the templates and the different resolution features of the search image are respectively subjected to correlation operations to obtain response score maps of different layer features. The response score maps are finally combined with a certain weight.</p> Full article ">Figure 2
<p>The precision plots and success plots on OTB-2015 benchmarks. The curves and numbers were generated with an visual tracker benchmark(OTB) toolkit.</p> Full article ">Figure 3
<p>The precision plots and success plots on OTB-50 benchmarks. The curves and numbers were generated with an visual tracker benchmark(OTB) toolkit.</p> Full article ">Figure 4
<p>The precision plots and success plots on OTB-2013 benchmarks. Curves and numbers are generated with visual tracker benchmark(OTB) toolkit.</p> Full article ">Figure 5
<p>Qualitative results of our Kalman–Siam, along with state-of-the-art trackers on six challenge sequences. <span class="html-italic">girl2, motorRolling, Human3, tiger1, coupon, skating2</span>. Kalman–Siam tracks accurately and robustly in these hard cases.</p> Full article ">Figure 6
<p>The success plots on the OTB-2015 and OTB-50 dataset in the three scenarios of occlusion, fast motion, and deformation.</p> Full article ">Figure 7
<p>Expected average overlap(EAO) ranking with trackers in VOT2015. The legend shows the results of the top 10 tracker and Kalman–Siam. The yellow horizontal dotted line indicates the VOT2015 state-of-the-art bound.</p> Full article ">Figure 8
<p>Expected average overlap(EAO) ranking with trackers in VOT2016. The legend shows the results of the top 10 tracker and Kalman–Siam.</p> Full article ">
18 pages, 2194 KiB  
Article
Research on Cognitive Marine Radar Based on LFM Waveform Control
by Yi Liu, Shufang Zhang, Jidong Suo, Tingting Yao and Jingbo Zhang
Sensors 2019, 19(9), 2200; https://doi.org/10.3390/s19092200 - 13 May 2019
Cited by 1 | Viewed by 3425
Abstract
In this paper, the method of applying cognitive radar technology to marine radar is studied, and the cognitive marine radar structure and transmitted signal model with three control parameters are constructed. The selection method of waveform control parameters, which is based on the [...] Read more.
In this paper, the method of applying cognitive radar technology to marine radar is studied, and the cognitive marine radar structure and transmitted signal model with three control parameters are constructed. The selection method of waveform control parameters, which is based on the target spatial distribution and the reference target detection effect with the minimum emission energy as the criterion, is given. The transmission signal control selection method given in this paper can flexibly realize different emission signal groups of m × n × p groups by independently setting the values m, n and p of three control parameters. It does not require radar hardware circuit reconstruction to meet the radar waveform changes. This is more convenient for the technical realization of cognitive marine radar. According to the method of this paper, a cognitive marine radar test system was constructed. The experimental results show that the proposed radar could reduce the emission energy by 15.9 dB compared with the traditional fixed-parameter pulse compression marine radar under the experimental conditions. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The basic waveform of the transmitted signal.</p> Full article ">Figure 2
<p>Schematic diagram of three pulses emitted by each pulse group of the radar.</p> Full article ">Figure 3
<p>Echo video signal of the transmitted pulse group and composite video signal after the distance stitching.</p> Full article ">Figure 4
<p>Structure block diagram of cognitive marine radar.</p> Full article ">Figure 5
<p>Baseband signal generator block diagram.</p> Full article ">Figure 6
<p>Statistical characteristics of 54♯ IPIX radar clutter before and after the accumulation and average processing.</p> Full article ">Figure 7
<p>Radar experimental system experimental test site.</p> Full article ">Figure 8
<p>Comparison of radar images during perception and cognitive control.</p> Full article ">Figure 9
<p>Schematic diagram of three pulses emitted by each fire group of the radar.</p> Full article ">
21 pages, 5696 KiB  
Article
Guided Wave Propagation in Detection of Partial Circumferential Debonding in Concrete Structures
by Beata Zima
Sensors 2019, 19(9), 2199; https://doi.org/10.3390/s19092199 - 13 May 2019
Cited by 16 | Viewed by 4517
Abstract
The following article presents results of investigating the damage detection in reinforced concrete beams with artificially introduced debonding between the rod and cover, using a non-destructive method based on elastic waves propagation. The primary aim of the research was to analyze the possible [...] Read more.
The following article presents results of investigating the damage detection in reinforced concrete beams with artificially introduced debonding between the rod and cover, using a non-destructive method based on elastic waves propagation. The primary aim of the research was to analyze the possible use of guided waves in partial circumferential debonding detection. Guided waves were excited and registered in reinforced concrete specimens with varying extents of debonding damage by piezoelectric sensors attached at both ends of the beams. Experimental results in the form of time–domain signals registered for variable extent of debonding were compared, and the relationships relating to the damage size and time of flight and average wave velocity were proposed. The experimental results were compared with theoretical predictions based on dispersion curves traced for the free rod of circular cross-section and rectangular reinforced concrete cross-section. The high agreement of theoretical and experimental data proved that the proposed method, taking advantage of average wave velocity, can be efficiently used for assessing debonding size in reinforced concrete structures. It was shown that the development of damage size in circumferential direction has a completely different impact on wave velocity than development of debonding length. The article contains a continuation of work previously conducted on the detection of delamination in concrete structures. The proposed relationship is the next essential step for developing a diagnostics method for detecting debondings of any size and orientation. Full article
(This article belongs to the Special Issue Piezoelectric Transducers: Advances in Structural Health Monitoring)
►▼ Show Figures

Figure 1

Figure 1
<p>Partial circumferential debonding caused by corrosion damage in reinforced concrete beam.</p> Full article ">Figure 2
<p>Types of cross-sections distinguished in partially debonded reinforced concrete beam: (<b>a</b>) single, debonded waveguide, (<b>b</b>) rectangular reinforced concrete cross-section, and (<b>c</b>) rectangular cross-section with circumferential debonding between concrete cast and steel rod.</p> Full article ">Figure 3
<p>Group velocity dispersion curves of 2 cm diameter steel bar (E = 207 GPa, v = 0.3, and ρ = 7894 kg/m<sup>3</sup>) obtained analytically (solid lines) and dispersion curves for rectangular reinforced concrete cross-section (steel: E = 207 GPa, v = 0.3, ρ = 7894 kg/m<sup>3</sup>, concrete: E = 29 GPa, v = 0.2, and ρ = 2306 kg/m<sup>3</sup>).</p> Full article ">Figure 4
<p>Experimental specimens<b>:</b> (<b>a</b>) geometry and (<b>b</b>) cross-section of the beam.</p> Full article ">Figure 5
<p>Schemes of investigated specimens: (<b>a</b>) beam #A with healthy bonded rod and beams with circumferential debonding with extent of (<b>b</b>) #B = 90, (<b>c</b>) #C = 180, (<b>d</b>) #D = 270, (<b>e</b>) #E = 360, and (<b>f</b>) a photograph of the rod wrapped in cellophane film.</p> Full article ">Figure 6
<p>Experimental set up: (<b>a</b>) tested reinforced concrete beam and (<b>b</b>) a detail of transducer attached to rod embedded in beam.</p> Full article ">Figure 7
<p>Experimental wave propagation signals registered at the end of the beam with variable circumferential debonding extent: (<b>a</b>) undamaged beam and beam with debonding extent of (<b>b</b>) 90°, (<b>c</b>) 180°, (<b>d</b>) 270°, and (<b>e</b>) 360° for excitation frequency 60 kHz.</p> Full article ">Figure 8
<p>Initial part of the signal registered for the beam #D: determining the time of flight for overlapping wave packets.</p> Full article ">Figure 9
<p>Changes in an average wave velocity depending on debonding size.</p> Full article ">Figure 10
<p>Times of flight of reflections registered at the end of the beams for excitation frequency of (<b>a</b>) 40 kHz and (<b>b</b>) 50 kHz for variable extent of debonding.</p> Full article ">Figure 11
<p>Numerical results of wave propagation in (<b>a</b>) undamaged beam and beams with circumferential deboning of extent of (<b>b</b>) de = 180° (beam #C) and (<b>c</b>) de = 360° (beam #E) for excitation frequency 60 kHz.</p> Full article ">Figure 12
<p>Numerical wave propagation signals registered at the end of the beam with variable circumferential debonding extent: (<b>a</b>) undamaged beam and beam with debonding extent of (<b>b</b>) 90°, (<b>c</b>) 180°, (<b>d</b>) 270°, and (<b>e</b>) 360° for excitation frequency 60 kHz.</p> Full article ">Figure 13
<p>Scheme of specimen with multiple debondings.</p> Full article ">Figure 14
<p>Envelopes of signals obtained (<b>a</b>) experimentally and (<b>b</b>) numerically for varying extent of debonding.</p> Full article ">
16 pages, 11960 KiB  
Article
Delamination Detection in Polymeric Ablative Materials Using Pulse-Compression Thermography and Air-Coupled Ultrasound
by Stefano Laureti, Muhammad Khalid Rizwan, Hamed Malekmohammadi, Pietro Burrascano, Maurizio Natali, Luigi Torre, Marco Rallini, Ivan Puri, David Hutchins and Marco Ricci
Sensors 2019, 19(9), 2198; https://doi.org/10.3390/s19092198 - 13 May 2019
Cited by 25 | Viewed by 5006
Abstract
Ablative materials are used extensively in the aerospace industry for protection against high thermal stresses and temperatures, an example being glass/silicone composites. The extreme conditions faced and the cost-risk related to the production/operating stage of such high-tech materials indicate the importance of detecting [...] Read more.
Ablative materials are used extensively in the aerospace industry for protection against high thermal stresses and temperatures, an example being glass/silicone composites. The extreme conditions faced and the cost-risk related to the production/operating stage of such high-tech materials indicate the importance of detecting any anomaly or defect arising from the manufacturing process. In this paper, two different non-destructive testing techniques, namely active thermography and ultrasonic testing, have been used to detect a delamination in a glass/silicone composite. It is shown that a frequency modulated chirp signal and pulse-compression can successfully be used in active thermography for detecting such a delamination. Moreover, the same type of input signal and post-processing can be used to generate an image using air-coupled ultrasound, and an interesting comparison between the two can be made to further characterise the defect. Full article
(This article belongs to the Special Issue Infrared Thermography and Additional Non-Destructive Testing for Building, Structure and Material Inspections)
►▼ Show Figures

Figure 1

Figure 1
<p>Implementation of the pulse-compression algorithm in air-coupled ultrasonic testing (PuC-ACUT). A coded signal <math display="inline"><semantics> <mrow> <mi>s</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> is input to an ultrasonic transducer, which excites the sample under test (SUT) for a limited time and within a bounded frequency range. The output signal <math display="inline"><semantics> <mrow> <mi>y</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> is then convolved with the matched filter <math display="inline"><semantics> <mrow> <mi mathvariant="sans-serif">Ψ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> to obtain an estimate of the sample impulse response <math display="inline"><semantics> <mrow> <mover accent="true"> <mi>h</mi> <mo>˜</mo> </mover> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math>.</p> Full article ">Figure 2
<p>Implementation of the pulse-compression algorithm in active thermography. A coded signal excites <math display="inline"><semantics> <mrow> <mi>s</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> modulates the on/off state of a heat source, exciting the sample under test (SUT) for a limited time and within a bounded frequency range. The output signal <math display="inline"><semantics> <mrow> <msub> <mi>y</mi> <mrow> <mi>S</mi> <mi>Q</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>+</mo> <mi>y</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> is de-trend from the step heating contribution <math display="inline"><semantics> <mrow> <msub> <mi>y</mi> <mrow> <mi>S</mi> <mi>Q</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>y</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> is obtained. This is convolved (“PuC box”) with the matched filter <math display="inline"><semantics> <mrow> <mi mathvariant="sans-serif">Ψ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> to obtain an estimate of the sample impulse response <math display="inline"><semantics> <mrow> <mover accent="true"> <mi>h</mi> <mo>˜</mo> </mover> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math>.</p> Full article ">Figure 3
<p>(<b>a</b>,<b>b</b>) Photographs of the glass/silicone Thermal Protection Shielding (TPS) sample manufactured by Angeloni Tech Materials (Italy). The green dotted line in (<b>b</b>) highlights the position of the delamination, which can be barely seen by the naked eye.</p> Full article ">Figure 4
<p>Photographs of (<b>a</b>) the complete Pulse-Compression Thermography (PuCT) experimental setup and (<b>b</b>) the LED array rotated on its square gantry allowing the active LED elements to be seen. See the main text (<a href="#sec4dot1-sensors-19-02198" class="html-sec">Section 4.1</a>.) for details.</p> Full article ">Figure 5
<p>Photograph of the pulse-compression air-coupled ultrasonic setup—(1) transducers, (2) pinhole and (3) sample under test. See the main text (<a href="#sec4dot2-sensors-19-02198" class="html-sec">Section 4.2</a>.) for details.</p> Full article ">Figure 6
<p>(<b>a</b>) Normalised estimated impulse response <math display="inline"><semantics> <mrow> <mover accent="true"> <mi>h</mi> <mo>˜</mo> </mover> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> for a single pixel of the reconstructed thermograms after pulse-compression. (<b>b</b>) A series of thermograms obtained by imaging pixelwise the <math display="inline"><semantics> <mrow> <mover accent="true"> <mi>h</mi> <mo>˜</mo> </mover> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math> amplitude (normalised and in dB scale) as time elapses, i.e., at 1, 2, 3 and 4 s. An extended area of higher temperature is noticed, which is likely to be related to the delamination.</p> Full article ">Figure 7
<p>First Principal Component reconstructed image after pulse-compression. The potential delaminated area is showed by the “hot” pixels area, and it was marked with red dots to be easily identified.</p> Full article ">Figure 8
<p>(<b>a</b>) Amplified chirp input signal and (<b>b</b>) acquired data from a single scanning step onto the delaminated part. (<b>c</b>) Obtained impulse response after pulse-compression, together with its envelope.</p> Full article ">Figure 9
<p>Pulse-compression air coupled ultrasonic results. The image shows the <math display="inline"><semantics> <mrow> <mover accent="true"> <mi>h</mi> <mo>˜</mo> </mover> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math>s normalised amplitude (dB) and the presence of an extended delamination is visible as pixel area with higher attenuation.</p> Full article ">Figure 10
<p>Pulse-compression air coupled ultrasonic results. The image shows the <math display="inline"><semantics> <mrow> <mover accent="true"> <mi>h</mi> <mo>˜</mo> </mover> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </semantics></math>s normalised amplitude (dB) but with three different bounded amplitude levels.</p> Full article ">
6 pages, 3699 KiB  
Correction
Correction: Lee, T. R., et al. On the Use of Rotary-Wing Aircraft to Sample Near-Surface Thermodynamic Fields: Results from Recent Field Campaigns. Sensors 2019, 19(1), 10
by Temple R. Lee, Michael Buban, Edward Dumas and C. Bruce Baker
Sensors 2019, 19(9), 2197; https://doi.org/10.3390/s19092197 - 13 May 2019
Cited by 2 | Viewed by 2547
Abstract
The authors wish to make the following correction to this paper [...] Full article
►▼ Show Figures

Figure 1

Figure 1
Full article ">Figure 2
Full article ">Figure 3
Full article ">Figure 4
Full article ">Figure 5
Full article ">Figure 6
Full article ">Figure 7
Full article ">Figure 8
Full article ">
11 pages, 1923 KiB  
Article
A Comparison Study of Fatigue Behavior of Hard and Soft Piezoelectric Single Crystal Macro-Fiber Composites for Vibration Energy Harvesting
by Mahesh Peddigari, Ga-Yeon Kim, Chan Hee Park, Yuho Min, Jong-Woo Kim, Cheol-Woo Ahn, Jong-Jin Choi, Byung-Dong Hahn, Joon-Hwan Choi, Dong-Soo Park, Jae-Keun Hong, Jong-Taek Yeom, Kwi-Il Park, Dae-Yong Jeong, Woon-Ha Yoon, Jungho Ryu and Geon-Tae Hwang
Sensors 2019, 19(9), 2196; https://doi.org/10.3390/s19092196 - 13 May 2019
Cited by 39 | Viewed by 5598
Abstract
Designing a piezoelectric energy harvester (PEH) with high power density and high fatigue resistance is essential for the successful replacement of the currently using batteries in structural health monitoring (SHM) systems. Among the various designs, the PEH comprising of a cantilever structure as [...] Read more.
Designing a piezoelectric energy harvester (PEH) with high power density and high fatigue resistance is essential for the successful replacement of the currently using batteries in structural health monitoring (SHM) systems. Among the various designs, the PEH comprising of a cantilever structure as a passive layer and piezoelectric single crystal-based fiber composites (SFC) as an active layer showed excellent performance due to its high electromechanical properties and dynamic flexibilities that are suitable for low frequency vibrations. In the present study, an effort was made to investigate the reliable performance of hard and soft SFC based PEHs. The base acceleration of both PEHs is held at 7 m/s2 and the frequency of excitation is tuned to their resonant frequency (fr) and then the output power (Prms) is monitored for 107 fatigue cycles. The effect of fatigue cycles on the output voltage, vibration displacement, dielectric, and ferroelectric properties of PEHs was analyzed. It was noticed that fatigue-induced performance degradation is more prominent in soft SFC-based PEH (SS-PEH) than in hard SFC-based PEH (HS-PEH). The HS-PEH showed a slight degradation in the output power due to a shift in fr, however, no degradation in the maximum power was noticed, in fact, dielectric and ferroelectric properties were improved even after 107 vibration cycles. In this context, the present study provides a pathway to consider the fatigue life of piezoelectric material for the designing of PEH to be used at resonant conditions for long-term operation. Full article
(This article belongs to the Special Issue Flexible and Stretchable Piezoelectric Devices for Mechanical Sensing and Energy Harvesting)
►▼ Show Figures

Figure 1

Figure 1
<p>Photographs of (<b>a</b>) flexible single crystal-based fiber composites (SFC) and (<b>b</b>) experimental setup used for investigating the fatigue behavior of hard and soft SFC-based PEHs. The inset shows an image of soft and hard SFCs after attaching to the Ti-alloy elastic layer.</p> Full article ">Figure 2
<p>(<b>a</b>) and (<b>c</b>), (<b>b</b>) and (<b>d</b>) are the RMS voltage and RMS power curves for hard and soft SFC based piezoelectric energy harvesters (PEHs) measured as a function of load resistances at different excitation frequencies.</p> Full article ">Figure 3
<p>Variation in the <span class="html-italic">P</span><sub>rms</sub> response of (<b>a</b>) hard SFC-based PEH (HS-PEH) and (<b>b</b>) soft SFC-based PEH (SS-PEH) as a function of vibration cycles measured at 7 m/s<sup>2</sup>.</p> Full article ">Figure 4
<p>Comparison of power frequency response curves for (<b>a</b>) HS-PEH and (<b>b</b>) SS-PEHs measured at base acceleration of 7 m/s<sup>2</sup>.</p> Full article ">Figure 5
<p>The comparison graphs of output voltage, vibration displacement, capacitance, and polarization changes for (<b>a</b>–<b>d</b>) HS-PEH and (<b>e</b>–<b>h</b>) SS-PEHs measured before and after fatigue test at 7 m/s<sup>2</sup>.</p> Full article ">
11 pages, 2471 KiB  
Article
Single Transparent Piezoelectric Detector for Optoacoustic Sensing—Design and Signal Processing
by Elias Blumenröther, Oliver Melchert, Jonas Kanngießer, Merve Wollweber and Bernhard Roth
Sensors 2019, 19(9), 2195; https://doi.org/10.3390/s19092195 - 12 May 2019
Cited by 17 | Viewed by 4667
Abstract
In this article, we present a simple and intuitive approach to create a handheld optoacoustic setup for near field measurements. A single piezoelectric transducer glued in between two sheets of polymethyl methacrylate (PMMA) facilitates nearfield depth profiling of layered media. The detector electrodes [...] Read more.
In this article, we present a simple and intuitive approach to create a handheld optoacoustic setup for near field measurements. A single piezoelectric transducer glued in between two sheets of polymethyl methacrylate (PMMA) facilitates nearfield depth profiling of layered media. The detector electrodes are made of indium tin oxide (ITO) which is both electrically conducting as well as optically transparent, enabling an on-axis illumination through the detector. By mapping the active detector area, we show that it matches the design form precisely. We also present a straightforward approach to determine the instrument response function, which allows to obtain the original pressure profile arriving at the detector. To demonstrate the validity of this approach, the measurement on a simple test sample is deconvolved with the instrument response function and compared to simulation results. Except for the sputter instrumentation, all required materials and instruments as well as the tools needed to create such a setup are available to standard scientific laboratories. Full article
(This article belongs to the Special Issue State-of-the-Art Sensors Technology in Germany)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Sketch of the optoacoustics (OA) setup. DAC: Data acquisition card, Preamp: Preamplifier. (<b>b</b>) Photograph of the probe. The laser injects 532 nm pulses into a fiber with a transparent jacket. A calibrated diode placed next to the fiber monitors the energy. The transparent detector is placed directly on the sample. Illumination is centered through the detector. The polymethyl methacrylate (PMMA) backing layer is 5 mm and the fronting layer 0.5 mm thick.</p> Full article ">Figure 2
<p>(<b>a</b>) Transmission spectra of the polyvinylidene (PVDF) film with and without indium tin oxide (ITO) electrodes. ITO reduces transparency drastically below 400 nm, at 600 nm transmission is highest (80%) and decreases slowly towards higher wavelengths. At 532 nm transmission is still above 70%. Oscillations are due to interferometric effects of the thin film. (<b>b</b>) Photo of detector film on scale paper with 1 mm between lines. Contrast is enhanced for visibility. The active area of the detector, where the two electrodes overlap, shows as a slightly darker spot at the apex of the curved electrode section.</p> Full article ">Figure 3
<p>(<b>a</b>) Microscopic view of the sputtered indium tin oxide (ITO) electrodes. The electrode sputtered on top of the polyvinylidene (PVDF) film (left) and the electrode at the bottom (right) overlap in a nearly circular region with a diameter of roughly 1 mm. This is expected to be the active area of the piezoelectric sensor. (<b>b</b>) Spatial pyroelectric response of the PVDF-sensor. The sensor film has been illuminated through a 150 µm aperture. The absorption within the detector material gives rise to a transient pyroelectric signal. By scanning the illuminated area in x<sub>1</sub> and x<sub>2</sub> direction over the sensor film, the energy-normalized amplitude of the pyroelectric signal at different positions of the detector was acquired. The shape and size of the acquired spatial response nicely resembles the area where the sputtered electrodes overlap. Outside of the displayed region no significant pyroelectric signal could be observed.</p> Full article ">Figure 4
<p>Removal of the pyroelectric signal from the measurement of the instrument response function (IRF). Pyroelectrical signal begins at the time of the laser excitation at 1.12 µs. Primary OA signal arrives at 1.33 µs.</p> Full article ">Figure 5
<p>The optoacoustics (OA) signal of black plastic (blue) shows unwanted reflections. Butterworth filter as used on OA signal (orange). Resulting instrument response functions (IRF), only featuring the delta peak response (yellow).</p> Full article ">Figure 6
<p>Visualization of post-processing. (<b>a</b>) optoacoustic (OA) signal of the instrument response functions (IRF) and ink-on-glass after removal of the pyroelectric signal (green and red). IRF was zero padded to prevent transformation artifacts. (<b>b</b>) Frequency spectra of IRF (green) and the ink-on glass-signal with 20 MHz cut-off (red), the Butterworth filter for the 20 MHz cut-off as applied on the signal (turquoise), and the resulting signal after deconvolution (blue). (<b>c</b>) Back transformed signal (blue) and 1D simulation (brown) agree well. All features of the signal can be identified.</p> Full article ">
20 pages, 7711 KiB  
Article
ANTspin: Efficient Absolute Localization Method of RFID Tags via Spinning Antenna
by Leixian Shen, Qingyun Zhang, Jiayi Pang, He Xu, Peng Li and Donghui Xue
Sensors 2019, 19(9), 2194; https://doi.org/10.3390/s19092194 - 12 May 2019
Cited by 35 | Viewed by 5264
Abstract
The Global Positioning System (GPS) has been widely applied in outdoor positioning, but it cannot meet the accuracy requirements of indoor positioning. Comprising an important part of the Internet of Things perception layer, Radio Frequency Identification (RFID) plays an important role in indoor [...] Read more.
The Global Positioning System (GPS) has been widely applied in outdoor positioning, but it cannot meet the accuracy requirements of indoor positioning. Comprising an important part of the Internet of Things perception layer, Radio Frequency Identification (RFID) plays an important role in indoor positioning. We propose a novel localization scheme aiming at the defects of existing RFID localization technology in localization accuracy and deployment cost, called ANTspin: Efficient Absolute Localization Method of RFID Tags via Spinning Antenna, which introduces a rotary table in the experiment. The reader antenna is fixed on the rotary table to continuously collect dynamic data. When compared with static acquisition, there is more information for localization. After that, the relative incident angle and distance between tags and the antenna can be analyzed for localization with characteristics of Received Signal Strength Indication (RSSI) data. We implement ANTspin using COTS RFID devices and the experimental results show that it achieves a mean accuracy of 9.34 cm in 2D and mean accuracy of 13.01 cm in three-dimensions (3D) with high efficiency and low deployment cost. Full article
(This article belongs to the Special Issue Emerging Applications and Challenges of RFID for the Internet of Things)
►▼ Show Figures

Figure 1

Figure 1
<p>Static data fluctuation. (<b>a</b>) Received Signal Strength Indication (RSSI); (<b>b</b>) Phase; and, (<b>c</b>) Doppler Frequency.</p> Full article ">Figure 2
<p>Experimental scenario. (<b>a</b>) Regional division scenario; and, (<b>b</b>) Bookshelf test scenario</p> Full article ">Figure 3
<p>Principle analysis.</p> Full article ">Figure 4
<p>Principle analysis.</p> Full article ">Figure 5
<p>Two-dimensional (2D) deployment.</p> Full article ">Figure 6
<p>Three-dimensional (3D) deployment.</p> Full article ">Figure 7
<p>Data of Phase and Doppler Frequency.</p> Full article ">Figure 8
<p>RSSI data changes. (<b>a</b>) In one row; and, (<b>b</b>) in one column.</p> Full article ">Figure 9
<p>Data segmentation and interception. (<b>a</b>) Raw data; (<b>b</b>) After segmentation; and, (<b>c</b>) After interception.</p> Full article ">Figure 10
<p>Wavelet filtering.</p> Full article ">Figure 11
<p>Least squares fitting.</p> Full article ">Figure 12
<p>Relative incident angle analysis.</p> Full article ">Figure 13
<p>2D localization principle analysis. (<b>a</b>) Top view; (<b>b</b>) Sectional view.</p> Full article ">Figure 14
<p>3D localization principle analysis. (<b>a</b>) Top view; and, (<b>b</b>) Sectional view.</p> Full article ">Figure 15
<p>Experimental scenarios. (<b>a</b>) two-dimensional (2D); (<b>b</b>) three-dimensional (3D); and, (<b>c</b>) three-dimensional (3D).</p> Full article ">Figure 16
<p>Localization error in 2D. (<b>a</b>) Relative incident angle error in 2D; (<b>b</b>) Distance calculation error in 2D; and, (<b>c</b>) Localization error in 2D.</p> Full article ">Figure 17
<p>Localization error in 3D. (<b>a</b>) Relative incident angle error in 3D; (<b>b</b>) Distance calculation error in 3D; and, (<b>c</b>) Localization error in 3D.</p> Full article ">Figure 18
<p>Distance between tags vs. Error.</p> Full article ">Figure 19
<p>Distance between antenna and tags array vs. Error.</p> Full article ">Figure 20
<p>Distance between reference tags vs. Error.</p> Full article ">Figure 21
<p>Express post stations. (<b>a</b>) Large express parcels situation; (<b>b</b>) Small express parcels situation.</p> Full article ">
8 pages, 2232 KiB  
Article
Demonstration of a Low-Cost and Portable Optical Cavity-Based Sensor through Refractive Index Measurements
by Donggee Rho, Caitlyn Breaux and Seunghyun Kim
Sensors 2019, 19(9), 2193; https://doi.org/10.3390/s19092193 - 12 May 2019
Cited by 4 | Viewed by 4005
Abstract
An optical cavity-based sensor using a differential detection method has been proposed for point-of-care diagnostics. We developed a low-cost and portable optical cavity-based sensor system using a 3D printer and off-the-shelf optical components. In this paper, we demonstrate the sensing capability of the [...] Read more.
An optical cavity-based sensor using a differential detection method has been proposed for point-of-care diagnostics. We developed a low-cost and portable optical cavity-based sensor system using a 3D printer and off-the-shelf optical components. In this paper, we demonstrate the sensing capability of the portable system through refractive index measurements. Fabricated optical cavity samples were tested using the portable system and compared to simulation results. A referencing technique and digital low pass filtering were applied to reduce the noise of the portable system. The measurement results match the simulation results well and show the improved linearity and sensitivity by employing the differential detection method. The limit of detection achieved was 1.73 × 10−5 Refractive Index Unit (RIU), which is comparable to other methods for refractive index sensing. Full article
(This article belongs to the Section Optical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Schematic diagram of the optical cavity-based sensor using 830 nm and 880 nm laser diodes. (<b>b</b>) Cross-sectional view of the optical cavity structure.</p> Full article ">Figure 2
<p>(<b>a</b>) Simulation results showing efficiencies of 830 nm (blue dashed line) and 880 nm (red dotted line) and differential value (green solid line) versus the refractive index inside the optical cavity in the range between 1.328 and 1.338. (<b>b</b>) Simulation results as shown in <a href="#sensors-19-02193-f002" class="html-fig">Figure 2</a>a with the range of refractive index between 1.3329 and 1.3338.</p> Full article ">Figure 3
<p>(<b>a</b>) Fabricated optical cavity sample including 6 fluidic channels. 3D printed adapters are attached to inlets and outlets. (<b>b</b>) Prototype of portable optical cavity-based sensor. (<b>c</b>) Optical components mounted on the middle level plate of the portable system. (<b>d</b>) Schematic of servo motors (blue parts) with blocking plates (yellow parts) to block laser diodes alternately.</p> Full article ">Figure 4
<p>The frequency response of the digital low-pass filter (LPF).</p> Full article ">Figure 5
<p>Measurement results showing the average pixel intensities for 830 nm (blue dashed line) and 880 nm (red dotted line) and differential value (green solid line) versus the refractive indices in the same range (1.3329–1.3338) as shown in <a href="#sensors-19-02193-f002" class="html-fig">Figure 2</a>b.</p> Full article ">
11 pages, 3681 KiB  
Article
Active Hyperspectral Sensor Based on MEMS Fabry-Pérot Interferometer
by Teemu Kääriäinen, Priit Jaanson, Aigar Vaigu, Rami Mannila and Albert Manninen
Sensors 2019, 19(9), 2192; https://doi.org/10.3390/s19092192 - 12 May 2019
Cited by 14 | Viewed by 6659
Abstract
An active hyperspectral sensor (AHS) was developed for target detection and classification applications. AHS measures light scattered from a target, illuminated by a broadband near-infrared supercontinuum (SC) light source. Spectral discrimination is based on a voltage-tunable MEMS Fabry-Pérot Interferometer (FPI). The broadband light [...] Read more.
An active hyperspectral sensor (AHS) was developed for target detection and classification applications. AHS measures light scattered from a target, illuminated by a broadband near-infrared supercontinuum (SC) light source. Spectral discrimination is based on a voltage-tunable MEMS Fabry-Pérot Interferometer (FPI). The broadband light is filtered by the FPI prior to transmitting, allowing for a high spectral-power density within the eye-safety limits. The approach also allows for a cost-efficient correction of the SC instability, employing a non-dispersive reference detector. A precision of 0.1% and long-term stability better than 0.5% were demonstrated in laboratory tests. The prototype was mounted on a car for field measurements. Several road types and objects were distinguished based on the spectral response of the sensor targeted in front of the car. Full article
(This article belongs to the Special Issue Sensors In Target Detection)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic of the active hyperspectral sensor (AHS) instrument consisting of supercontinuum (SC), transmitter and receiver modules (<b>a</b>). PL = pump laser, SCF = supercontinuum fiber, L1 = short focal length lens for fiber coupling, L2 = short focal lens for focusing light through the Fabry-Pérot interferometer (FPI), LPF = long pass filter, TD = trigger detector, RD = reference detector, M1-M2 = convex and concave mirrors for beam expansion, L3 = light collecting lens, DM = dichroic mirror, SD = signal detector, L4 = focusing lens. CAD drawing of the prototype (<b>b</b>). The transmitter module is shown in blue.</p> Full article ">Figure 2
<p>Spectral power density of the generated SC. The generation of longer wavelengths is more efficient with longer fibers, however, excessively long fiber results in a power-loss due to intrinsic fiber absorption. Short fiber lengths result in an inefficient spectral broadening.</p> Full article ">Figure 3
<p>Tuning characteristics of the FPI. (<b>a</b>) The central wavelength and full-width at half maximum at a given voltage. The red curve shows the relationship between FPI voltage and FWHM. The blue curve shows the relationship between FPI voltage and the central bandpass wavelength. (<b>b</b>) Spectral transmission measured with a constant voltage interval, indicated by different colors.</p> Full article ">Figure 4
<p>(<b>a</b>) Measured widths for single (1.5 μm) wavelength and a Gaussian fit Extrapolated 1/e<sup>2</sup> diameter at 100 m and (<b>b</b>) the fitted M<sup>2</sup> in function of wavelength.</p> Full article ">Figure 5
<p>(<b>a</b>) Allan-Werle deviation of the measured pulse intensity with and without the reference signal. The center wavelength of the channel shown is 1500 ± 2.5 nm. Spectralon was used as the target at 5 m distance. The data was collected in 14 h. (<b>b</b>) The Allan-Werle deviation for averaged 1, 10 and 100 pulses as well as 7-hour values are shown for center wavelengths between 1350 and 1650 nm with 10 nm increments.</p> Full article ">Figure 6
<p>(<b>a</b>) Total intensity measured with three different gain levels and 5 levels of signal obstruction. (<b>b</b>) 1σ Standard deviation of intensity-normalized spectra.</p> Full article ">Figure 7
<p>The sensor mounted on the test vehicle.</p> Full article ">Figure 8
<p>Selected spectra of road surfaces from the field-trial. Each spectrum is an average over 10 spectral measurements with a 100 Hz measurement rate. The standard error of the mean of the 10 spectra are shown as error bars. The test vehicle was driven with a speed ranging between 10–40 km/h.</p> Full article ">Figure 9
<p>Selected spectra of objects from the field-trial. Each spectrum is an average over 10 spectral measurements with a 100 Hz measurement rate. The standard error of the mean of the 10 spectra are shown as error bars. The test vehicle was driven with a speed ranging between 0–10 km/h.</p> Full article ">
24 pages, 4729 KiB  
Article
Plummeting Broadcast Storm Problem in Highways by Clustering Vehicles Using Dominating Set and Set Cover
by S. Kamakshi and V. S. Shankar Sriram
Sensors 2019, 19(9), 2191; https://doi.org/10.3390/s19092191 - 12 May 2019
Cited by 9 | Viewed by 4123
Abstract
“Vehicular Ad-hoc Networks” (VANETs): As an active research area in the field of wireless sensor networks, they ensure road safety by exchanging alert messages about unexpected events in a decentralized manner. One of the significant challenges in the design of an efficient dissemination [...] Read more.
“Vehicular Ad-hoc Networks” (VANETs): As an active research area in the field of wireless sensor networks, they ensure road safety by exchanging alert messages about unexpected events in a decentralized manner. One of the significant challenges in the design of an efficient dissemination protocol for VANETs is the broadcast storm problem, owing to the large number of rebroadcasts. A generic solution to prevent the broadcast storm problem is to cluster the vehicles based on topology, density, distance, speed, or location in such a manner that only a fewer number of vehicles will rebroadcast the alert message to the next group. However, the selection of cluster heads and gateways of the clusters are the key factors that need to be optimized in order to limit the number of rebroadcasts. Hence, to address the aforementioned issues, this paper presents a novel distributed algorithm CDS_SC: Connected Dominating Set and Set Cover for cluster formation that employs a dominating set to choose cluster heads and set covering to select cluster gateways. The CDS_SC is unique among state-of-the-art algorithms, as it relies on local neighborhood information and constructs clusters incrementally. Hence, the proposed method can be implemented in a distributed manner as an event-triggered protocol. Also, the stability of cluster formation is increased along with a reduction in rebroadcasting by allowing a cluster head to be passive when all its cluster members can receive the message from the gateway vehicles. The simulation was carried out in dense, average, and sparse traffic scenarios by varying the number of vehicles injected per second per lane. Besides, the speed of each individual vehicle in each scenario was varied to test the degree of cohesion between vehicles with different speeds. The simulation results confirmed that the proposed algorithm achieved 99% to 100% reachability of alert messages with only 6% to 10% of rebroadcasting vehicles in average and dense traffic scenarios. Full article
(This article belongs to the Special Issue Advances in Sustainable Computing for Wireless Sensor Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Network model: simulation of highway with 4 lanes in each direction.</p> Full article ">Figure 2
<p>The vertex subsets {1, 6, 8}, {1, 4, 9}, {3, 8}, {3, 5, 9}, and {3, 7, 10} are few dominating sets of <math display="inline"><semantics> <mi>G</mi> </semantics></math> where as {3, 8} is a minimal dominating set. Hence <math display="inline"><semantics> <mrow> <mi>γ</mi> <mrow> <mo stretchy="false">(</mo> <mi>G</mi> <mo stretchy="false">)</mo> </mrow> <mo>=</mo> <mn>2.</mn> </mrow> </semantics></math> The vertex sets {3, 4, 5, 8}, {3, 4, 7, 8}, and {3, 6, 7, 8} are few minimum connected dominating sets Hence <math display="inline"><semantics> <mrow> <msub> <mi>γ</mi> <mi>c</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi>G</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </semantics></math> = 4.</p> Full article ">Figure 3
<p>Cluster formation and gateway selection.</p> Full article ">Figure 4
<p>State transition in vehicles.</p> Full article ">Figure 5
<p>Cluster formation algorithm.</p> Full article ">Figure 6
<p>Gateway selection algorithm.</p> Full article ">Figure 7
<p>Clustering of vehicles in an average traffic scenario.</p> Full article ">Figure 8
<p>Clustering of vehicles in dense traffic scenario.</p> Full article ">Figure 9
<p>Clustering of vehicles in sparse traffic scenario.</p> Full article ">Figure 10
<p>Cluster stability.</p> Full article ">Figure 11
<p>Average number of members per cluster.</p> Full article ">Figure 12
<p>(<b>a</b>) Cluster head density (<b>b</b>) Cluster state distribution.</p> Full article ">Figure 12 Cont.
<p>(<b>a</b>) Cluster head density (<b>b</b>) Cluster state distribution.</p> Full article ">Figure 13
<p>Clustering performance.</p> Full article ">Figure 14
<p>Packet delivery ratio versus vehicle density.</p> Full article ">Figure 15
<p>End-to-end delay versus vehicle speed.</p> Full article ">Figure 16
<p>Packet collision ratio versus vehicle density.</p> Full article ">Figure 17
<p>Packet loss ratio versus vehicle density.</p> Full article ">
24 pages, 5954 KiB  
Article
A Trajectory Privacy Preserving Scheme in the CANNQ Service for IoT
by Lin Zhang, Chao Jin, Hai-ping Huang, Xiong Fu and Ru-chuan Wang
Sensors 2019, 19(9), 2190; https://doi.org/10.3390/s19092190 - 12 May 2019
Cited by 6 | Viewed by 3519
Abstract
Nowadays, anyone carrying a mobile device can enjoy the various location-based services provided by the Internet of Things (IoT). ‘Aggregate nearest neighbor query’ is a new type of location-based query which asks the question, ‘what is the best location for a given group [...] Read more.
Nowadays, anyone carrying a mobile device can enjoy the various location-based services provided by the Internet of Things (IoT). ‘Aggregate nearest neighbor query’ is a new type of location-based query which asks the question, ‘what is the best location for a given group of people to gather?’ There are numerous, promising applications for this type of query, but it needs to be done in a secure and private way. Therefore, a trajectory privacy-preserving scheme, based on a trusted anonymous server (TAS) is proposed. Specifically, in the snapshot queries, the TAS generates a group request that satisfies the spatial K-anonymity for the group of users—to prevent the location-based service provider (LSP) from an inference attack—and in continuous queries, the TAS determines whether the group request needs to be resent by detecting whether the users will leave their secure areas, so as to reduce the probability that the LSP reconstructs the users’ real trajectories. Furthermore, an aggregate nearest neighbor query algorithm based on strategy optimization, is adopted, to minimize the overhead of the LSP. The response speed of the results is improved by narrowing the search scope of the points of interest (POIs) and speeding up the prune of the non-nearest neighbors. The security analysis and simulation results demonstrated that our proposed scheme could protect the users’ location and trajectory privacy, and the response speed and communication overhead of the service, were superior to other peer algorithms, both in the snapshot and continuous queries. Full article
(This article belongs to the Special Issue Security and Privacy in Internet of Things)
►▼ Show Figures

Figure 1

Figure 1
<p>The architecture based on a trusted anonymous server.</p> Full article ">Figure 2
<p>Original query and K-anonymity query of single user.</p> Full article ">Figure 3
<p>Example of a continuous aggregate nearest neighbor queries (CANNQ).</p> Full article ">Figure 4
<p>Details of our system architecture.</p> Full article ">Figure 5
<p>Anonymous process.</p> Full article ">Figure 6
<p>K-anonymity query and PCANNQ.</p> Full article ">Figure 7
<p>Secure areas and dominated distances.</p> Full article ">Figure 8
<p>Circular secure areas.</p> Full article ">Figure 9
<p>Aggregate subgroup.</p> Full article ">Figure 10
<p>Towards the centroid.</p> Full article ">Figure 11
<p>Security of locations.</p> Full article ">Figure 12
<p>Security of trajectories.</p> Full article ">Figure 13
<p>The effect of points of interest (POIs) distribution on η and Secure areas.</p> Full article ">Figure 14
<p>The effect of moving speed on communication frequency.</p> Full article ">Figure 15
<p>Service processing time proportion.</p> Full article ">Figure 16
<p>The effect of number of users on query response time.</p> Full article ">Figure 17
<p>The effect of number of users on the searching set size.</p> Full article ">Figure 18
<p>The effect of user distribution on query performance.</p> Full article ">
17 pages, 8616 KiB  
Article
Precise Point Positioning Using Dual-Frequency GNSS Observations on Smartphone
by Qiong Wu, Mengfei Sun, Changjie Zhou and Peng Zhang
Sensors 2019, 19(9), 2189; https://doi.org/10.3390/s19092189 - 11 May 2019
Cited by 130 | Viewed by 9349
Abstract
The update of the Android system and the emergence of the dual-frequency GNSS chips enable smartphones to acquire dual-frequency GNSS observations. In this paper, the GPS L1/L5 and Galileo E1/E5a dual-frequency PPP (precise point positioning) algorithm based on RTKLIB and GAMP was applied [...] Read more.
The update of the Android system and the emergence of the dual-frequency GNSS chips enable smartphones to acquire dual-frequency GNSS observations. In this paper, the GPS L1/L5 and Galileo E1/E5a dual-frequency PPP (precise point positioning) algorithm based on RTKLIB and GAMP was applied to analyze the positioning performance of the Xiaomi Mi 8 dual-frequency smartphone in static and kinematic modes. The results showed that in the static mode, the RMS position errors of the dual-frequency smartphone PPP solutions in the E, N, and U directions were 21.8 cm, 4.1 cm, and 11.0 cm, respectively, after convergence to 1 m within 102 min. The PPP of dual-frequency smartphone showed similar accuracy with geodetic receiver in single-frequency mode, while geodetic receiver in dual-frequency mode has higher accuracy. In the kinematic mode, the positioning track of the smartphone dual-frequency data had severe fluctuations, the positioning tracks derived from the smartphone and the geodetic receiver showed approximately difference of 3–5 m. Full article
(This article belongs to the Special Issue Smart Mobile and Sensor Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>The cycle slip rate of 30-minute observations before and after turning on “duty cycle” (data collected by Mi 8 on October 7th, 2018 in the basketball court, Jilin University).</p> Full article ">Figure 2
<p>View of data collection at roof of Geological Palace Museum ((<b>a</b>) under the GNSS antenna; (<b>b</b>) on the edge of the roof).</p> Full article ">Figure 3
<p>Horizontal position errors of four data sets in PPP kinematic mode.</p> Full article ">Figure 4
<p>The number of satellites with simultaneous L1 and L5 frequency data record and PDOP.</p> Full article ">Figure 5
<p>Position errors of smartphone and receiver ((<b>a</b>): east position errors; (<b>b</b>): north position errors; (<b>c</b>): up position errors).</p> Full article ">Figure 6
<p>Carrier phase residuals of smartphone and receiver in static mode ((<b>a</b>) smartphone; (<b>b</b>) receiver).</p> Full article ">Figure 7
<p>Kinematic positioning track in Google Earth.</p> Full article ">Figure 8
<p>Horizontal position errors of four data sets in kinematic mode.</p> Full article ">Figure 9
<p><math display="inline"><semantics> <mrow> <mi mathvariant="normal">C</mi> <mo stretchy="false">/</mo> <msub> <mi>N</mi> <mn>0</mn> </msub> </mrow> </semantics></math> of GPS, Galileo and GLONASS satellites (Receiver: red lines; Smartphone: black lines).</p> Full article ">Figure 10
<p>The number of satellites observed by smartphone.</p> Full article ">
16 pages, 3198 KiB  
Article
Highly Enhanced Inductance Sensing Performance of Dual-Quartz Crystal Converter
by Vojko Matko and Miro Milanovic
Sensors 2019, 19(9), 2188; https://doi.org/10.3390/s19092188 - 11 May 2019
Cited by 4 | Viewed by 3627
Abstract
This paper presents ways of inductance sensitivity improvement in a quartz crystal converter for low inductance measurement. To improve the converter’s sensitivity, two quartz crystals that were connected in parallel and additional capacitance connected to the two quartz crystals in the oscillator’s circuit [...] Read more.
This paper presents ways of inductance sensitivity improvement in a quartz crystal converter for low inductance measurement. To improve the converter’s sensitivity, two quartz crystals that were connected in parallel and additional capacitance connected to the two quartz crystals in the oscillator’s circuit are used. The new approach uses a converter with special switchable oscillator and multiplexer switches to compensate for the crystal’s natural temperature-frequency characteristics and any other influences, such as parasitic capacitances and parasitic inductances, which reduce them to a minimum. The experimental results demonstrate improved sensitivity and well-compensated dynamic temperature influence on the converter’s output frequency. The fundamental quartz crystal frequency-temperature characteristics in the temperature range between 0–40 °C are simultaneously compensated. Furthermore, the converter enables the measurement of the influence of its own hysteresis at different values of inductances at the selected sensitivity by parallel capacitances connected either to the single- or dual-quartz crystal unit. The results show that the converter converting inductances in the range between 85–100 μH to a frequency range between 1–150 kHz only has ±0.05 ppm frequency instability (during the temperature change between 0–40 °C), which gives the converter a resolution of 1 pH. As a result, the converter can be applied where low inductance measurement, nondestructive testing, impedance change measurement, and magnetic material properties measurement are important. Full article
(This article belongs to the Special Issue Integrated Magnetic Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Oscillator’s switching principle.</p> Full article ">Figure 2
<p>(<b>a</b>) Equivalent single-quartz crystal electrical circuit and (<b>b</b>) additional connection of a second crystal in parallel to the first one. In both cases, the same impedance <span class="html-italic">Z</span> is connected in series.</p> Full article ">Figure 3
<p>Automated inductance converter principle.</p> Full article ">Figure 4
<p>Inductance-to-frequency characteristics of the converter for inductance <span class="html-italic">L</span><sub>L</sub> settings in steps from 48.004 μH to 100.002 μH (without capacitance C<sub>p</sub> – the dotted lines and for the different values of capacitance <span class="html-italic">C</span><sub>p</sub>, and for the single- and dual-quartz units at <span class="html-italic">T</span> = 25 °C).</p> Full article ">Figure 5
<p>The relative frequency differences <span class="html-italic">f</span><sub>01u</sub> − <span class="html-italic">f</span><sub>01d</sub>/<span class="html-italic">f</span><sub>0</sub> depending on the connection of the single- or dual-quartz crystal units in the oscillator without the capacitance C<sub>p</sub> and for different capacitance values C<sub>p</sub> = 1–5 pF.</p> Full article ">Figure 6
<p>(<b>a</b>,<b>d</b>) Demonstration of the dynamic temperature change from 0–40 °C and back to 0 °C, (<b>b</b>,<b>e</b>) The change of the output frequency differences (<span class="html-italic">f</span><sub>01</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) and (<span class="html-italic">f</span><sub>02</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) during the temperature change, (<b>c</b>) The zoomed interval (<a href="#sensors-19-02188-f006" class="html-fig">Figure 6</a>b) between eight to 12 minutes for a single measurement of the frequency difference (<span class="html-italic">f</span><sub>01</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) and (<span class="html-italic">f</span><sub>02</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) (<b>f</b>) The zoomed interval (<a href="#sensors-19-02188-f006" class="html-fig">Figure 6</a>e) between 8 to 12 minutes for four measurements of the frequency difference (<span class="html-italic">f</span><sub>01</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) and (<span class="html-italic">f</span><sub>02</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)).</p> Full article ">Figure 7
<p>(<b>a</b>,<b>d</b>) Dynamic temperature change from 0 °C to 40 °C and back to 0 °C, (<b>b</b>,<b>e</b>) The change of the output frequency differences (<span class="html-italic">f</span><sub>01</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) and (<span class="html-italic">f</span><sub>02</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) during the temperature change, (<b>c</b>) The zoomed interval (<a href="#sensors-19-02188-f007" class="html-fig">Figure 7</a>b) between eight and 12 minutes for one measurement of frequency differences (<span class="html-italic">f</span><sub>01</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) and (<span class="html-italic">f</span><sub>02</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) (<b>f</b>) The zoomed interval (<a href="#sensors-19-02188-f007" class="html-fig">Figure 7</a>e) between 8 and 12 minutes for four measurements of frequency differences (<span class="html-italic">f</span><sub>01</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)) and (<span class="html-italic">f</span><sub>02</sub>(t) − <span class="html-italic">f</span><sub>r</sub>(t)).</p> Full article ">
19 pages, 8230 KiB  
Article
A Novel Method to Enable the Awareness Ability of Non-V2V-Equipped Vehicles in Vehicular Networks
by Jian Wang, Qiang Zheng, Fang Mei, Weiwen Deng and Yuming Ge
Sensors 2019, 19(9), 2187; https://doi.org/10.3390/s19092187 - 11 May 2019
Cited by 1 | Viewed by 3667
Abstract
Autonomous vehicles need to have sufficient perception of the surrounding environment to produce appropriate driving behavior. The Vehicle-to-Vehicle (V2V) communication technology can exchange the speed, position, direction, and other information between autonomous vehicles to improve the sensing ability of the traditional on-board sensors. [...] Read more.
Autonomous vehicles need to have sufficient perception of the surrounding environment to produce appropriate driving behavior. The Vehicle-to-Vehicle (V2V) communication technology can exchange the speed, position, direction, and other information between autonomous vehicles to improve the sensing ability of the traditional on-board sensors. For example, V2V communication technology does not have a blind spot like a conventional on-board sensor, and V2V communication is not easily affected by weather conditions. However, it is almost impossible to make every vehicle a V2V-equipped vehicle in the real environment due to reasons such as policy and user choice. Low penetration of V2V-equipped vehicles greatly reduces the performance of the traditional V2V system. In this paper, however, we propose a novel method that can extend the awareness ability of the traditional V2V system without adding much extra investment. In the traditional V2V system, only a V2V-equipped vehicle can broadcast its own location information. However, the situation is somewhat different in our V2V system. Although non-V2V-equipped vehicles cannot broadcast their own location information, we can let V2V-equipped vehicle with radar and other sensors detect the location information of the surrounding non-V2V-equipped vehicles and then broadcast it out. Therefore, we think that a non-V2V-equipped vehicle can also broadcast its own location information. In this way, we greatly extend the awareness ability of the traditional V2V system. The proposed method is validated by real experiments and simulation experiments. Full article
(This article belongs to the Special Issue Internet of Vehicles)
►▼ Show Figures

Figure 1

Figure 1
<p>Our V2V system can extend the awareness ability of the traditional V2V system. (<b>a</b>) V2V-equipped Vehicle A cannot detect the non-V2V-equipped Vehicles C, D, F, and H in the traditional V2V system. (<b>b</b>) V2V-equipped Vehicle A can detect the non-V2V-equipped Vehicles C, D, F, and H in our V2V system.</p> Full article ">Figure 2
<p>Experimental vehicle.</p> Full article ">Figure 3
<p>Hardware of the system.</p> Full article ">Figure 4
<p>The coordinate system of the experimental car.</p> Full article ">Figure 5
<p>The process of our system.</p> Full article ">Figure 6
<p>Open convex and closed convex.</p> Full article ">Figure 7
<p>License recognition.</p> Full article ">Figure 8
<p>Scenario of method feasibility simulation.</p> Full article ">Figure 9
<p>The influence of V2VRate on sysDetectAbility.</p> Full article ">Figure 10
<p>The influence of vehicleRate on sysDetectAbility.</p> Full article ">Figure 11
<p>The influence of detectR2 on sysDetectAbility.</p> Full article ">Figure 12
<p>The influence of vehicle density, detectRate, and packet payload size on the packet delivery ratio.</p> Full article ">Figure 13
<p>The influence of vehicle density and detectRate on delay.</p> Full article ">Figure 14
<p>The influence of vehicle density, detectRate, and packet payload size on throughput.</p> Full article ">Figure 15
<p>Scenario of our experiment. (<b>a</b>) Scenario of the Forward Collision Warning (FCW) experiment. (<b>b</b>) Scenario of the Collision Warning At Crossroads (CWAC) experiment.</p> Full article ">Figure 16
<p>Result of FCW Experiment 1. (<b>a</b>) Initial FCW distance is 50 m, and initial velocity is 10 m/s. (<b>b</b>) Initial FCW distance is 50 m, and initial velocity is 15 m/s.</p> Full article ">Figure 17
<p>Result of FCW Experiment 2. (<b>a</b>) Initial FCW distance is 25 m, and initial velocity is 10 m/s. (<b>b</b>) Initial FCW distance is 25 m, and initial velocity is 15 m/s.</p> Full article ">Figure 18
<p>Result of the CWAC experiment.</p> Full article ">
12 pages, 1557 KiB  
Article
Effects of Frequency Filtering on Intensity and Noise in Accelerometer-Based Physical Activity Measurements
by Jonatan Fridolfsson, Mats Börjesson, Christoph Buck, Örjan Ekblom, Elin Ekblom-Bak, Monica Hunsberger, Lauren Lissner and Daniel Arvidsson
Sensors 2019, 19(9), 2186; https://doi.org/10.3390/s19092186 - 11 May 2019
Cited by 50 | Viewed by 9124
Abstract
In objective physical activity (PA) measurements, applying wider frequency filters than the most commonly used ActiGraph (AG) filter may be beneficial when processing accelerometry data. However, the vulnerability of wider filters to noise has not been investigated previously. This study explored the effect [...] Read more.
In objective physical activity (PA) measurements, applying wider frequency filters than the most commonly used ActiGraph (AG) filter may be beneficial when processing accelerometry data. However, the vulnerability of wider filters to noise has not been investigated previously. This study explored the effect of wider frequency filters on measurements of PA, sedentary behavior (SED), and capturing of noise. Apart from the standard AG band-pass filter (0.29–1.63 Hz), modified filters with low-pass component cutoffs at 4 Hz, 10 Hz, or removed were analyzed. Calibrations against energy expenditure were performed with lab data from children and adults to generate filter-specific intensity cut-points. Free-living accelerometer data from children and adults were processed using the different filters and intensity cut-points. There was a contribution of acceleration related to PA at frequencies up to 10 Hz. The contribution was more pronounced at moderate and vigorous PA levels, although additional acceleration also occurred at SED. The classification discrepancy between AG and the wider filters was small at SED (1–2%) but very large at the highest intensities (>90%). The present study suggests an optimal low-pass frequency filter with a cutoff at 10 Hz to include all acceleration relevant to PA with minimal effect of noise. Full article
(This article belongs to the Collection Sensors for Globalized Healthy Living and Wellbeing)
►▼ Show Figures

Figure 1

Figure 1
<p>Description of the original ActiGraph and the modified processing methods. * Truncation to 6 g with the modified methods was only performed with the calibration data. ** Converting acceleration in g to counts with a 2.13 g to 8-bit resolution was only performed with the output presented in counts.</p> Full article ">Figure 2
<p>MET values plotted against acceleration output (mg, counts) using different filters for all locomotion speeds and subjects as well as fitted smoothing splines. Dotted lines represent cut-points at 1.5 (LPA), 3 (MPA), 6 (VPA), and 9 (VVPA) METs, respectively.</p> Full article ">Figure 3
<p>(<b>a</b>) Absolute aggregated three-second samples of acceleration from the sub-bands. (<b>b</b>) Aggregated acceleration relative to the ActiGraph filter.</p> Full article ">Figure 4
<p>Row-normalized confusion charts comparing ActiGraph-filtered output with output from the modified filters epoch by epoch. Numbers on the diagonal from the upper left to lower right corner with 100% would indicate a perfect agreement between the filters whereas numbers in the lower or upper triangles from the diagonal indicate that the modified filters classify the activity intensity as lower or higher, respectively.</p> Full article ">
15 pages, 3881 KiB  
Article
Characterization of Ultrasonic Energy Diffusion in a Steel Alloy Sample with Tensile Force Using PZT Transducers
by Guangtao Lu, Tao Wang, Mingle Zhou and Yourong Li
Sensors 2019, 19(9), 2185; https://doi.org/10.3390/s19092185 - 11 May 2019
Cited by 8 | Viewed by 4235
Abstract
During the propagation of ultrasound in a polycrystalline material, ultrasonic energy losses due to the scattering at the boundaries between grains is usually described by the ultrasonic energy diffusion equation, and the boundaries of the grains in the material are influenced by the [...] Read more.
During the propagation of ultrasound in a polycrystalline material, ultrasonic energy losses due to the scattering at the boundaries between grains is usually described by the ultrasonic energy diffusion equation, and the boundaries of the grains in the material are influenced by the structural load. The aim of this research is to investigate the characterization of ultrasonic energy diffusion in a steel alloy sample under structural load by using lead zirconate titanate (PZT) transducers. To investigate the influence of structural load on ultrasonic energy diffusion, an experimental setup of a steel alloy plate under different tensile forces is designed and four samples with similar dimensions are fabricated. The experimental results of the four samples reveal that, during the loading process, the normalized ultrasonic energy diffusion coefficient fluctuates firstly, then decreases and at last increases as the tensile force increases. The proposed tensile force index shows a similar changing trend to the recorded displacement of the sample. Moreover, when the tensile force is less than the lower yield point or the sample deforms elastically, the index can be approximated by a cubic model. Therefore, the proposed tensile force index can be used to monitor the tensile force in the elastic deformation stage. Moreover, based on these findings, some force evaluation methods and their potential applications, such as the preloading detection of bolts, can be developed based on the linear relationships between the proposed index and the applied force. Full article
(This article belongs to the Special Issue Ultrasound Transducers)
►▼ Show Figures

Figure 1

Figure 1
<p>Experimental setup for the data acquisition system.</p> Full article ">Figure 2
<p>Tensile samples with two PZTs: (<b>a</b>) Before the test; (<b>b</b>) After the destructive tests.</p> Full article ">Figure 3
<p>Dimensions of the sample and the PZTs (unit: mm).</p> Full article ">Figure 4
<p>Normalized displacement and group velocity of different modes of Lamb wave excited in the sample: (<b>a</b>) Normalized displacement; (<b>b</b>) Group velocity.</p> Full article ">Figure 5
<p>Curves of the excitation pulse in time and frequency domains: (<b>a</b>) Time domain; (<b>b</b>) Frequency domain.</p> Full article ">Figure 6
<p>Response of the structure to the excitation pulse.</p> Full article ">Figure 7
<p>Ultrasonic energy density curve with a tensile force 48 kN.</p> Full article ">Figure 8
<p>Deformation curve of sample L4 under tensile force. (A—Elastic deformation stage; B—Plastic deformation stage; C—Necking stage).</p> Full article ">Figure 9
<p>Normalized ultrasonic energy coefficients of three samples under tensile forces: (<b>a</b>) Sample L1; (<b>b</b>) Sample L2; (<b>c</b>) Sample L3.</p> Full article ">Figure 10
<p>Tensile force index versus the tensile force for sample L1. <b>A</b>—Elastic deformation stage; <b>B</b>—Plastic deformation stage; <b>C</b>—Necking stage.</p> Full article ">Figure 11
<p>Tensile force index of the three samples during the elastic deformation stage: (<b>a</b>) Sample L1; (<b>b</b>) Sample L2; (<b>c</b>) Sample L3.</p> Full article ">
26 pages, 641 KiB  
Review
Multi-Phase Flow Metering in Offshore Oil and Gas Transportation Pipelines: Trends and Perspectives
by Lærke Skov Hansen, Simon Pedersen and Petar Durdevic
Sensors 2019, 19(9), 2184; https://doi.org/10.3390/s19092184 - 11 May 2019
Cited by 93 | Viewed by 13641
Abstract
Multi-phase flow meters are of huge importance to the offshore oil and gas industry. Unreliable measurements can lead to many disadvantages and even wrong decision-making. It is especially important for mature reservoirs as the gas volume fraction and water cut is increasing during [...] Read more.
Multi-phase flow meters are of huge importance to the offshore oil and gas industry. Unreliable measurements can lead to many disadvantages and even wrong decision-making. It is especially important for mature reservoirs as the gas volume fraction and water cut is increasing during the lifetime of a well. Hence, it is essential to accurately monitor the multi-phase flow of oil, water and gas inside the transportation pipelines. The objective of this review paper is to present the current trends and technologies within multi-phase flow measurements and to introduce the most promising methods based on parameters such as accuracy, footprint, safety, maintenance and calibration. Typical meters, such as tomography, gamma densitometry and virtual flow meters are described and compared based on their performance with respect to multi-phase flow measurements. Both experimental prototypes and commercial solutions are presented and evaluated. For a non-intrusive, non-invasive and inexpensive meter solution, this review paper predicts a progress for virtual flow meters in the near future. The application of multi-phase flows meters are expected to further expand in the future as fields are maturing, thus, efficient utilization of existing fields are in focus, to decide if a field is still financially profitable. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Subsea manifold and transportation pipelines to separation platform.</p> Full article ">Figure 2
<p>Produced water and discharged dispersed oil from the Danish platforms in the North Sea. The platforms include: Dan, Gorm, Halfdan, Tyra, Syd Arne and Siri. The fields are operated by Total E&amp;P Danmark A/S, Hess Danmark ApS and INEOS Oil &amp; Gas. The blue graph illustrates the total amount of produced water from the fields. The red graph illustrates the discharged dispersed oil in the PW.</p> Full article ">Figure 3
<p>Oil production system with test separator and 1st stage separator. The flow inside the pipe is denoted as either M for multi-phase flow or S for single phase flow. After the test separator each phase flow is ideally measured by a single-phase flow meter (FM).</p> Full article ">Figure 4
<p>Multi-phase flow measurement using tomography imaging process.</p> Full article ">Figure 5
<p>ECT sensor with 8 electrodes around the pipe. One electrode is excited and the capacitance is measured.</p> Full article ">Figure 6
<p>Principle of an orifice plate. Interruption of the flow inside a pipe due to an orifice plate. DP transmitters are measuring the pressure difference at a point before and after the orifice plate and the velocity of the fluid is hereby obtained by Bernoulli’s equation.</p> Full article ">Figure 7
<p>Principle of a venturi meter. The DP transmitters are located before the pipe is converging (<math display="inline"><semantics> <msub> <mi>d</mi> <mn>1</mn> </msub> </semantics></math>) and when the pipe is most converged (<math display="inline"><semantics> <msub> <mi>d</mi> <mn>2</mn> </msub> </semantics></math>).</p> Full article ">Figure 8
<p>A PT sensor placed on a pipe. The PT sensor is replaced without causing any affection on the oil production due to the location of the valve.</p> Full article ">Figure 9
<p>A coriolis meter placed inline of the pipe. Replacement of the meter will cause a shut down in the oil production in the given location.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-257
Previous Issue
Next Issue
Back to TopTop