[go: up one dir, main page]
Next Issue
Volume 18, March
Previous Issue
Volume 18, January
 
 
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 18
Issue 2
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 18, Issue 2 (February 2018) – 362 articles

Cover Story (view full-size image): Pec and co-workers report a random access preamble (RAP) design technique for underwater acoustic cellular systems. After showing that the conventional RAP used in long term evolution (LTE) systems is not appropriate for underwater acoustic cellular systems, two different types of RAPs (RAP 1 and RAP 2) are proposed to detect the identity of underwater equipment/nodes (UE) and estimate the time delay between a UE and an underwater base station at the physical layer. The performance of RAP detection is investigated by analyzing the detection probabilities and false alarm probabilities of RAP 1 and RAP 2 in a Doppler environment. By evaluating the performances of RAP 1 and RAP 2 in various situations, it was concluded that RAP 2 is more suitable for underwater acoustic cellular systems. The AF and CAF analytically obtained in this paper are shown to be similar to those obtained using experimental data. [...] Read more.
  • 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:
22 pages, 2627 KiB  
Article
Comparison of Feature Learning Methods for Human Activity Recognition Using Wearable Sensors
by Frédéric Li, Kimiaki Shirahama, Muhammad Adeel Nisar, Lukas Köping and Marcin Grzegorzek
Sensors 2018, 18(2), 679; https://doi.org/10.3390/s18020679 - 24 Feb 2018
Cited by 241 | Viewed by 15021
Abstract
Getting a good feature representation of data is paramount for Human Activity Recognition (HAR) using wearable sensors. An increasing number of feature learning approaches—in particular deep-learning based—have been proposed to extract an effective feature representation by analyzing large amounts of data. However, getting [...] Read more.
Getting a good feature representation of data is paramount for Human Activity Recognition (HAR) using wearable sensors. An increasing number of feature learning approaches—in particular deep-learning based—have been proposed to extract an effective feature representation by analyzing large amounts of data. However, getting an objective interpretation of their performances faces two problems: the lack of a baseline evaluation setup, which makes a strict comparison between them impossible, and the insufficiency of implementation details, which can hinder their use. In this paper, we attempt to address both issues: we firstly propose an evaluation framework allowing a rigorous comparison of features extracted by different methods, and use it to carry out extensive experiments with state-of-the-art feature learning approaches. We then provide all the codes and implementation details to make both the reproduction of the results reported in this paper and the re-use of our framework easier for other researchers. Our studies carried out on the OPPORTUNITY and UniMiB-SHAR datasets highlight the effectiveness of hybrid deep-learning architectures involving convolutional and Long-Short-Term-Memory (LSTM) to obtain features characterising both short- and long-term time dependencies in the data. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Activity Recognition Chain (ARC). Raw data are firstly acquired from sensors. After pre-processing, segments of data are extracted (Segmentation) and values relevant to the recognition problem are computed from them (Feature extraction). A classifier is then trained and evaluated using those features (Classification). In our framework, all steps of the ARC—except the feature extraction part—are fixed, and a Support-Vector-Machine (SVM) classifier is used for the classification stage.</p> Full article ">Figure 2
<p>An overview of the codebook approach.</p> Full article ">Figure 3
<p>Architecture of a MLP model with <span class="html-italic">H</span> hidden layers for sensor-based HAR. Input data from the different sensor channels are first flattened into a <math display="inline"> <semantics> <mrow> <mo stretchy="false">(</mo> <mi>T</mi> <mo>×</mo> <mi>S</mi> <mo stretchy="false">)</mo> </mrow> </semantics> </math>-dimensional vector and then fed to the hidden layers. All layers are fully-connected. The numbers in parenthesis indicate the number of neurons per layer. <span class="html-italic">T</span>, <span class="html-italic">S</span> and <span class="html-italic">N</span> designate the time length of the input data, the number of sensor channels, and the number of classes, respectively.</p> Full article ">Figure 4
<p>Architecture of a CNN model for sensor-based HAR. <span class="html-italic">T</span>, <span class="html-italic">S</span> and <math display="inline"> <semantics> <msub> <mi>n</mi> <mi>k</mi> </msub> </semantics> </math> designate the time length of the input data, the number of sensor channels, and the number of convolutional kernels of the <span class="html-italic">k</span>th layer, respectively. The convolutional and pooling kernels of the <span class="html-italic">k</span>th layer process patches of the input data of sizes <math display="inline"> <semantics> <mrow> <mo stretchy="false">(</mo> <msubsup> <mi>f</mi> <mi>T</mi> <mrow> <mo stretchy="false">(</mo> <mi>k</mi> <mo stretchy="false">)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>f</mi> <mi>S</mi> <mrow> <mo stretchy="false">(</mo> <mi>k</mi> <mo stretchy="false">)</mo> </mrow> </msubsup> <mo stretchy="false">)</mo> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <mo stretchy="false">(</mo> <msubsup> <mi>p</mi> <mi>T</mi> <mrow> <mo stretchy="false">(</mo> <mi>k</mi> <mo stretchy="false">)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>p</mi> <mi>S</mi> <mrow> <mo stretchy="false">(</mo> <mi>k</mi> <mo stretchy="false">)</mo> </mrow> </msubsup> <mo stretchy="false">)</mo> </mrow> </semantics> </math>, respectively. Neurons of intermediate convolutional layers perform convolution products across all convolutional maps of the previous layer.</p> Full article ">Figure 5
<p>Architecture of a LSTM network for sensor-based HAR (left). The intermediate and last LSTM layers are organized in a many-to-many and many-to-one layout, respectively. Each LSTM layer is composed of LSTM cells (right). <math display="inline"> <semantics> <msub> <mi mathvariant="bold-italic">x</mi> <mi>t</mi> </msub> </semantics> </math>, <math display="inline"> <semantics> <msub> <mi mathvariant="bold-italic">m</mi> <mi>t</mi> </msub> </semantics> </math> and <math display="inline"> <semantics> <msub> <mi mathvariant="bold-italic">c</mi> <mi>t</mi> </msub> </semantics> </math> represent the cell input, memory, and output at time <span class="html-italic">t</span>, respectively. In addition, ⨁ and ⨂ refer to the element-wise addition and multiplication, respectively. <math display="inline"> <semantics> <msub> <mi mathvariant="bold-italic">W</mi> <mo>*</mo> </msub> </semantics> </math>, <math display="inline"> <semantics> <msub> <mi mathvariant="bold-italic">U</mi> <mo>*</mo> </msub> </semantics> </math>, <math display="inline"> <semantics> <msub> <mi mathvariant="bold-italic">b</mi> <mo>*</mo> </msub> </semantics> </math> and <math display="inline"> <semantics> <msub> <mi>σ</mi> <mo>*</mo> </msub> </semantics> </math> designate internal weight matrices, bias vector and activation function (for gate <math display="inline"> <semantics> <mrow> <mo>*</mo> <mo>∈</mo> <mo>{</mo> <mi>i</mi> <mo>,</mo> <mi>o</mi> <mo>,</mo> <mi>f</mi> <mo>}</mo> </mrow> </semantics> </math>). <math display="inline"> <semantics> <msub> <mi>σ</mi> <mn>1</mn> </msub> </semantics> </math> and <math display="inline"> <semantics> <msub> <mi>σ</mi> <mn>2</mn> </msub> </semantics> </math> are internal activation functions applied on the input and memory of the cell, respectively.</p> Full article ">Figure 6
<p>Architecture of a hybrid CNN+LSTM model for sensor-based HAR. Each slice along the time dimension of the output of the convolutional block(s) (in blue) is fed to one LSTM cell. All LSTM layers are organized in a many-to-many pattern, except the last which follows a many-to-one scheme.</p> Full article ">Figure 7
<p>Architecture of an autoencoder model for sensor-based HAR. The numbers in parenthesis refer to the number of neurons per layer. <span class="html-italic">H</span> denotes the number of hidden layers in both the encoder and decoder.</p> Full article ">Figure 8
<p>Average F1-scores of the feature learning methods on the OPPORTUNITY dataset, using different numbers of sensor channels (ranked by decreasing variance).</p> Full article ">
21 pages, 8022 KiB  
Article
Low Computational Signal Acquisition for GNSS Receivers Using a Resampling Strategy and Variable Circular Correlation Time
by Yeqing Zhang, Meiling Wang and Yafeng Li
Sensors 2018, 18(2), 678; https://doi.org/10.3390/s18020678 - 24 Feb 2018
Cited by 11 | Viewed by 6083
Abstract
For the objective of essentially decreasing computational complexity and time consumption of signal acquisition, this paper explores a resampling strategy and variable circular correlation time strategy specific to broadband multi-frequency GNSS receivers. In broadband GNSS receivers, the resampling strategy is established to work [...] Read more.
For the objective of essentially decreasing computational complexity and time consumption of signal acquisition, this paper explores a resampling strategy and variable circular correlation time strategy specific to broadband multi-frequency GNSS receivers. In broadband GNSS receivers, the resampling strategy is established to work on conventional acquisition algorithms by resampling the main lobe of received broadband signals with a much lower frequency. Variable circular correlation time is designed to adapt to different signal strength conditions and thereby increase the operation flexibility of GNSS signal acquisition. The acquisition threshold is defined as the ratio of the highest and second highest correlation results in the search space of carrier frequency and code phase. Moreover, computational complexity of signal acquisition is formulated by amounts of multiplication and summation operations in the acquisition process. Comparative experiments and performance analysis are conducted on four sets of real GPS L2C signals with different sampling frequencies. The results indicate that the resampling strategy can effectively decrease computation and time cost by nearly 90–94% with just slight loss of acquisition sensitivity. With circular correlation time varying from 10 ms to 20 ms, the time cost of signal acquisition has increased by about 2.7–5.6% per millisecond, with most satellites acquired successfully. Full article
(This article belongs to the Special Issue Sensor Signal and Information Processing)
►▼ Show Figures

Figure 1

Figure 1
<p>The signal processing framework of GNSS receivers. The modules marked as red are the improved strategies of this paper, including the resampling strategy, variable circular correlation time and acquisition with pilot channel.</p> Full article ">Figure 2
<p>Acceptable sampling frequency (cyan areas) based on the bandpass sampling theory. Blue and green lines are lower and upper boundaries of the acceptable sampling frequency. The red line indicates the resampling frequency of the proposed strategy.</p> Full article ">Figure 3
<p>Signal flow chart by applying the resampling strategy and the convolutional method to the received broadband satellite signal. The green, dark brown, and orange represent frequency spectra of GPS L2C, P(Y), and M code signals, respectively.</p> Full article ">Figure 4
<p>Circular correlation results of the baseband signal <math display="inline"> <semantics> <mrow> <msub> <mi>S</mi> <mi>w</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> and the local zero-padding code <math display="inline"> <semantics> <mrow> <msub> <mi>C</mi> <mi>r</mi> </msub> <mrow> <mo>(</mo> <mi>p</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math>: (<b>a</b>) code offset between <math display="inline"> <semantics> <mrow> <msub> <mi>S</mi> <mi>w</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>C</mi> <mi>r</mi> </msub> <mrow> <mo>(</mo> <mi>p</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> is 0 (aligned); (<b>b</b>) code offset between <math display="inline"> <semantics> <mrow> <msub> <mi>S</mi> <mi>w</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>C</mi> <mi>r</mi> </msub> <mrow> <mo>(</mo> <mi>p</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> is less than <math display="inline"> <semantics> <mrow> <mo stretchy="false">(</mo> <msub> <mi>L</mi> <mi>c</mi> </msub> <mo>−</mo> <msub> <mi>L</mi> <mi>x</mi> </msub> <mo stretchy="false">)</mo> </mrow> </semantics> </math> samples; (<b>c</b>) code offset between <math display="inline"> <semantics> <mrow> <msub> <mi>S</mi> <mi>w</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>C</mi> <mi>r</mi> </msub> <mrow> <mo>(</mo> <mi>p</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> is more than <math display="inline"> <semantics> <mrow> <mo stretchy="false">(</mo> <msub> <mi>L</mi> <mi>c</mi> </msub> <mo>−</mo> <msub> <mi>L</mi> <mi>x</mi> </msub> <mo stretchy="false">)</mo> </mrow> </semantics> </math> samples.</p> Full article ">Figure 5
<p>The received broadband signal of Dataset 1 in the frequency and time domains, and the amplitude distribution. The bandwidth of the main lobe signal is <math display="inline"> <semantics> <mrow> <mn>2.046</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math> and the intermediate frequency is <math display="inline"> <semantics> <mrow> <mn>7.4</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math>.</p> Full article ">Figure 6
<p>The main lobe signal filtered from the received broadband signal of Dataset 1 in the frequency domain. The bandwidth of the main lobe signal is <math display="inline"> <semantics> <mrow> <mn>2.046</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math>, the intermediate frequency is <math display="inline"> <semantics> <mrow> <mn>7.4</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math> and the sampling frequency is <math display="inline"> <semantics> <mrow> <mn>53</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math>.</p> Full article ">Figure 7
<p>The resampled signal by applying the resampling strategy to the main lobe signal of Dataset 1 in the frequency domain. The bandwidth of the resampled signal is <math display="inline"> <semantics> <mrow> <mn>2.046</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math>, the intermediate frequency is <math display="inline"> <semantics> <mrow> <mn>1.43</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math> and the sampling frequency is <math display="inline"> <semantics> <mrow> <mn>5.97</mn> <mo> </mo> <mi>MHz</mi> </mrow> </semantics> </math>.</p> Full article ">Figure 8
<p>Acquisition results for GPS L2C satellites without/with the resampling strategy.</p> Full article ">Figure 9
<p>Correlation ratios of Satellite <math display="inline"> <semantics> <mrow> <mi>PRN</mi> <mn>12</mn> </mrow> </semantics> </math> acquired using the resampling strategy.</p> Full article ">Figure 10
<p>Sensitivity of the resampling strategy for weak signals. The red plot is the detection probability of signal acquisition with the resampling strategy, and the blue one is that of the conventional acquisition algorithm (without the resampling strategy).</p> Full article ">Figure 11
<p>Computation of signal acquisition without/with the resampling strategy for all experimental datasets: (<b>a</b>) Multiplication computation in linear axis; (<b>b</b>) Multiplication computation in log axis; (<b>c</b>) Summation computation in linear axis; (<b>d</b>) Summation computation in log axis.</p> Full article ">Figure 12
<p>Number of satellites acquired without/with the resampling strategy for variable circular correlation time. The green circles indicate the difference of acquisition results without/with the resampling strategy; the cyan circles indicate the incomplete acquisition results with too short circular correlation time.</p> Full article ">Figure 13
<p>Time cost of signal acquisition with/without the resampling strategy for variable circular correlation time. Red bars and plot represent acquisition results with the resampling strategy, while blue ones are for that of the conventional acquisition algorithm.</p> Full article ">
16 pages, 4167 KiB  
Article
Quadrature Errors and DC Offsets Calibration of Analog Complex Cross-Correlator for Interferometric Passive Millimeter-Wave Imaging Applications
by Chao Wang, Xin Xin, Bingyuan Liang, Zhiping Li and Jungang Miao
Sensors 2018, 18(2), 677; https://doi.org/10.3390/s18020677 - 24 Feb 2018
Cited by 14 | Viewed by 5442
Abstract
The design and calibration of the cross-correlator are crucial issues for interferometric imaging systems. In this paper, an analog complex cross-correlator with output DC offsets and amplitudes calibration capability is proposed for interferometric passive millimeter-wave security sensing applications. By employing digital potentiometers in [...] Read more.
The design and calibration of the cross-correlator are crucial issues for interferometric imaging systems. In this paper, an analog complex cross-correlator with output DC offsets and amplitudes calibration capability is proposed for interferometric passive millimeter-wave security sensing applications. By employing digital potentiometers in the low frequency amplification circuits of the correlator, the outputs characteristics of the correlator could be digitally controlled. A measurement system and a corresponding calibration scheme were developed in order to eliminate the output DC offsets and the quadrature amplitude error between the in-phase and the quadrature correlating subunits of the complex correlator. By using vector modulators to provide phase controllable correlated noise signals, the measurement system was capable of obtaining the output correlation circle of the correlator. When injected with −18 dBm correlated noise signals, the calibrated quadrature amplitude error was 0.041 dB and the calibrated DC offsets were under 26 mV, which was only 7.1% of the uncalibrated value. Furthermore, we also described a quadrature errors calibration algorithm in order to estimate the quadrature phase error and in order to improve the output phase accuracy of the correlator. After applying this calibration, we were able to reduce the output phase error of the correlator to 0.3°. Full article
(This article belongs to the Special Issue Sensors for Microwave Imaging and Detection)
►▼ Show Figures

Figure 1

Figure 1
<p>A block diagram of the diode-based analog complex cross-correlator.</p> Full article ">Figure 2
<p>The schematic of the digital programmable low frequency amplification circuit of the in-phase correlating subunit.</p> Full article ">Figure 3
<p>The correlator’s outputs DC offsets and quadrature errors calibration measurement test bench.</p> Full article ">Figure 4
<p>The output correlation circle of a complex correlator.</p> Full article ">Figure 5
<p>The flowchart for the DC offsets and the quadrature amplitude calibration.</p> Full article ">Figure 6
<p>Photographs of the experimental analog complex cross-correlator: (<b>a</b>) Top layer of a correlator board; (<b>b</b>) bottom layer of a correlator board; and (<b>c</b>) eight-channel correlator module.</p> Full article ">Figure 7
<p>Photograph of the main parts in the experimental system.</p> Full article ">Figure 8
<p>(<b>a</b>) The measured output correlation circle without hardware calibration; (<b>b</b>) the measured output correlation circle with hardware calibration; and (<b>c</b>) the output correlation circle after the quadrature errors calibration.</p> Full article ">Figure 9
<p>(<b>a</b>) The uncalibrated phase error at different input power levels; and (<b>b</b>) the calibrated phase error at different input power levels.</p> Full article ">
28 pages, 6824 KiB  
Article
An Automatic Gait Feature Extraction Method for Identifying Gait Asymmetry Using Wearable Sensors
by Arif Reza Anwary, Hongnian Yu and Michael Vassallo
Sensors 2018, 18(2), 676; https://doi.org/10.3390/s18020676 - 24 Feb 2018
Cited by 75 | Viewed by 12713
Abstract
This paper aims to assess the use of Inertial Measurement Unit (IMU) sensors to identify gait asymmetry by extracting automatic gait features. We design and develop an android app to collect real time synchronous IMU data from legs. The results from our method [...] Read more.
This paper aims to assess the use of Inertial Measurement Unit (IMU) sensors to identify gait asymmetry by extracting automatic gait features. We design and develop an android app to collect real time synchronous IMU data from legs. The results from our method are validated using a Qualisys Motion Capture System. The data are collected from 10 young and 10 older subjects. Each performed a trial in a straight corridor comprising 15 strides of normal walking, a turn around and another 15 strides. We analyse the data for total distance, total time, total velocity, stride, step, cadence, step ratio, stance, and swing. The accuracy of detecting the stride number using the proposed method is 100% for young and 92.67% for older subjects. The accuracy of estimating travelled distance using the proposed method for young subjects is 97.73% and 98.82% for right and left legs; and for the older, is 88.71% and 89.88% for right and left legs. The average travelled distance is 37.77 (95% CI ± 3.57) meters for young subjects and is 22.50 (95% CI ± 2.34) meters for older subjects. The average travelled time for young subjects is 51.85 (95% CI ± 3.08) seconds and for older subjects is 84.02 (95% CI ± 9.98) seconds. The results show that wearable sensors can be used for identifying gait asymmetry without the requirement and expense of an elaborate laboratory setup. This can serve as a tool in diagnosing gait abnormalities in individuals and opens the possibilities for home based self-gait asymmetry assessment. Full article
(This article belongs to the Special Issue Sensors for Gait, Posture, and Health Monitoring)
►▼ Show Figures

Figure 1

Figure 1
<p>IMU sensors placement in right and left metatarsal foot locations of the barefoot.</p> Full article ">Figure 2
<p>Proposed MetaWear casing, Velcro elastic belt, buckles and IMU sensor: (1) Buckle and Elastic Belt: the buckle is sewn onto an elastic belt for fastening to Velcro; (2) Bottom case which keeps the sensor safe from pressure, temperature and water; (3) Lock Open Edge which helps to open the cover from bottom case; (4) Sensor Lock Mechanism: The four locks keep the sensor sideways movement and orientation; (5) Cover Lock Mechanism which tightly locks with the case; (6) Velcro-Elastic Joint: The elastic belt is sewed with Velcro; (7) Velcro which adjusts and tighten when the sensor is attached; and (8) IMU sensor and battery.</p> Full article ">Figure 3
<p>Proposed android app to collect data from MetaWear CPro.</p> Full article ">Figure 4
<p>Raw accelerometer and gyroscope data from right and left feet of older subject 1.</p> Full article ">Figure 5
<p>(<b>a</b>) Sensor frame and earth frame of accelerometer and gyroscope axes; (<b>b</b>) The orientation of frame <span class="html-italic">E</span> is achieved by a rotation, from alignment with frame <span class="html-italic">S</span>, of angle of <span class="html-italic">φ</span>, <span class="html-italic">θ</span>, and <span class="html-italic">ψ</span> around the axis <span class="html-italic">S<sub>xyz</sub></span>.</p> Full article ">Figure 6
<p>The process diagram of the complete orientation filter for an IMU.</p> Full article ">Figure 7
<p>Acceleration due to user movement <span class="html-italic">AM<sub>xyz</sub></span> after removing gravity component.</p> Full article ">Figure 7 Cont.
<p>Acceleration due to user movement <span class="html-italic">AM<sub>xyz</sub></span> after removing gravity component.</p> Full article ">Figure 8
<p>The total acceleration <span class="html-italic">AT<sub>xyz</sub></span> and gyroscope <span class="html-italic">GT<sub>xyz</sub></span>.</p> Full article ">Figure 9
<p>Normal human gait phases [<a href="#B53-sensors-18-00676" class="html-bibr">53</a>].</p> Full article ">Figure 10
<p>Eight different phases of a gait cycle from accelerometer data.</p> Full article ">Figure 11
<p>Peaks magnitude variation from <a href="#sensors-18-00676-f008" class="html-fig">Figure 8</a>.</p> Full article ">Figure 12
<p>The proposed step detection technique.</p> Full article ">Figure 13
<p>Proposed stance and swing detection technique.</p> Full article ">Figure 14
<p>Result of stride, stance and swing event detection using proposed method.</p> Full article ">Figure 15
<p>Result of step event detection using proposed method.</p> Full article ">Figure 16
<p>Zero-velocity update (ZUPT) from <math display="inline"> <semantics> <mrow> <mrow> <mo>|</mo> <mrow> <mi>A</mi> <msub> <mi>T</mi> <mrow> <mi>x</mi> <mi>y</mi> <msub> <mi>z</mi> <mi>i</mi> </msub> </mrow> </msub> </mrow> <mo>|</mo> </mrow> </mrow> </semantics> </math>.</p> Full article ">Figure 17
<p>Proposed method for estimating travelled distance.</p> Full article ">Figure 18
<p>First integral operation to get velocity <span class="html-italic">V<sub>i</sub></span>(<span class="html-italic">t</span>).</p> Full article ">Figure 19
<p>2nd integral operation to get distance <span class="html-italic">D<sub>i</sub></span>(<span class="html-italic">t</span>).</p> Full article ">Figure 20
<p>Proposed variability monitoring for GA.</p> Full article ">Figure 21
<p>Stride, stance and swing information of right and left legs.</p> Full article ">Figure 22
<p>Stride asymmetry information of right and left legs.</p> Full article ">Figure 23
<p>Step asymmetry estimation of right and left legs.</p> Full article ">Figure 24
<p>BoxPlot of stride and step asymmetry in distances from right and left legs.</p> Full article ">Figure 25
<p>Box Plot of stride and step asymmetry in times from right and left legs.</p> Full article ">
21 pages, 3774 KiB  
Review
Integrated Affinity Biosensing Platforms on Screen-Printed Electrodes Electrografted with Diazonium Salts
by Paloma Yáñez-Sedeño, Susana Campuzano and José M. Pingarrón
Sensors 2018, 18(2), 675; https://doi.org/10.3390/s18020675 - 24 Feb 2018
Cited by 50 | Viewed by 12202
Abstract
Adequate selection of the electrode surface and the strategies for its modification to enable subsequent immobilization of biomolecules and/or nanomaterials integration play a major role in the performance of electrochemical affinity biosensors. Because of the simplicity, rapidity and versatility, electrografting using diazonium salt [...] Read more.
Adequate selection of the electrode surface and the strategies for its modification to enable subsequent immobilization of biomolecules and/or nanomaterials integration play a major role in the performance of electrochemical affinity biosensors. Because of the simplicity, rapidity and versatility, electrografting using diazonium salt reduction is among the most currently used functionalization methods to provide the attachment of an organic layer to a conductive substrate. This particular chemistry has demonstrated to be a powerful tool to covalently immobilize in a stable and reproducible way a wide range of biomolecules or nanomaterials onto different electrode surfaces. Considering the great progress and interesting features arisen in the last years, this paper outlines the potential of diazonium chemistry to prepare single or multianalyte electrochemical affinity biosensors on screen-printed electrodes (SPEs) and points out the existing challenges and future directions in this field. Full article
(This article belongs to the Special Issue Screen-Printed Electrodes)
►▼ Show Figures

Figure 1

Figure 1
<p>Functionalization of different carbon nanomaterial-modified SPEs (<b>a</b>) via diazonium salt reduction and affinity biosensor fabrication (<b>b</b>). Reprinted and adapted from [<a href="#B33-sensors-18-00675" class="html-bibr">33</a>] with permission.</p> Full article ">Figure 2
<p>Preparation of a diazonium-modified antibody electrode: carboxyl diazonium is covalently attached to antibody by EDC/NHS (1) and diazonium–antibody is deposited onto an electrode by cyclic voltammetry (2). Reprinted and adapted from [<a href="#B26-sensors-18-00675" class="html-bibr">26</a>] with permission.</p> Full article ">Figure 3
<p>Schematic display of the working principle of the immunosensor based on a direct competitive format and SWVs of the immunosensor before the competition step (1) and after incubation with different concentrations of OA: 0.00 (2), 1.00 (3), 10.0 (4), 100 (5) and 1000 (6) ng L<sup>−1</sup>. Reprinted and adapted from [<a href="#B59-sensors-18-00675" class="html-bibr">59</a>] with permission.</p> Full article ">Figure 4
<p>Electrochemical immunosensor developed for pSA determination onto a CNF SPE modified with a 4-carboxyphenyl layer (<b>a</b>) and comparison of the DPV responses provided by the immunosensor to 100 pg mL<sup>−1</sup> of serum albumin from porcine, bovine, rabbit and albumin in chicken egg (<b>b</b>). Reprinted and adapted from [<a href="#B61-sensors-18-00675" class="html-bibr">61</a>] with permission.</p> Full article ">Figure 5
<p>Schematic illustration of the different steps and protocols involved in the preparation and functioning of the dual GHRL and PYY immunosensor. Reprinted from [<a href="#B63-sensors-18-00675" class="html-bibr">63</a>] with permission.</p> Full article ">Figure 6
<p>Schematic display of the different steps involved in the construction of an amperometric immunosensor for APN involving grafted DWCNTs and oriented immobilization of anti-APN by using the metallic-complex chelating polymer Mix &amp; Go. Reprinted from [<a href="#B86-sensors-18-00675" class="html-bibr">86</a>] with permission.</p> Full article ">Figure 7
<p>Schematic display of the working principle of the impedimetric aptasensor for the determination of OTA. Reprinted from [<a href="#B21-sensors-18-00675" class="html-bibr">21</a>] with permission.</p> Full article ">Figure 8
<p>Impedimetric aptasensor developed for Lys determination on a 4-ABA-modified SPCE. Reprinted from [<a href="#B69-sensors-18-00675" class="html-bibr">69</a>] with permission.</p> Full article ">Figure 9
<p>Overview of the preparation of the <span class="html-italic">S. typhimurium</span> aptasensor. Reprinted from [<a href="#B95-sensors-18-00675" class="html-bibr">95</a>] with permission.</p> Full article ">Scheme 1
<p>Reaction mechanisms proposed for the preparation of single (<b>a</b>) or multiple (<b>b</b>) layers onto electrode surfaces through electroreductive electrografting of aryldiazonium salts. In (<b>b</b>) R<sub>1</sub> and R<sub>2</sub> represent two different substituents. Reprinted from [<a href="#B30-sensors-18-00675" class="html-bibr">30</a>] with permission.</p> Full article ">
28 pages, 1132 KiB  
Article
Towards an Iterated Game Model with Multiple Adversaries in Smart-World Systems
by Xiaofei He, Xinyu Yang, Wei Yu, Jie Lin and Qingyu Yang
Sensors 2018, 18(2), 674; https://doi.org/10.3390/s18020674 - 24 Feb 2018
Cited by 4 | Viewed by 4350
Abstract
Diverse and varied cyber-attacks challenge the operation of the smart-world system that is supported by Internet-of-Things (IoT) (smart cities, smart grid, smart transportation, etc.) and must be carefully and thoughtfully addressed before widespread adoption of the smart-world system can be fully realized. Although [...] Read more.
Diverse and varied cyber-attacks challenge the operation of the smart-world system that is supported by Internet-of-Things (IoT) (smart cities, smart grid, smart transportation, etc.) and must be carefully and thoughtfully addressed before widespread adoption of the smart-world system can be fully realized. Although a number of research efforts have been devoted to defending against these threats, a majority of existing schemes focus on the development of a specific defensive strategy to deal with specific, often singular threats. In this paper, we address the issue of coalitional attacks, which can be launched by multiple adversaries cooperatively against the smart-world system such as smart cities. Particularly, we propose a game-theory based model to capture the interaction among multiple adversaries, and quantify the capacity of the defender based on the extended Iterated Public Goods Game (IPGG) model. In the formalized game model, in each round of the attack, a participant can either cooperate by participating in the coalitional attack, or defect by standing aside. In our work, we consider the generic defensive strategy that has a probability to detect the coalitional attack. When the coalitional attack is detected, all participating adversaries are penalized. The expected payoff of each participant is derived through the equalizer strategy that provides participants with competitive benefits. The multiple adversaries with the collusive strategy are also considered. Via a combination of theoretical analysis and experimentation, our results show that no matter which strategies the adversaries choose (random strategy, win-stay-lose-shift strategy, or even the adaptive equalizer strategy), our formalized game model is capable of enabling the defender to greatly reduce the maximum value of the expected average payoff to the adversaries via provisioning sufficient defensive resources, which is reflected by setting a proper penalty factor against the adversaries. In addition, we extend our game model and analyze the extortion strategy, which can enable one participant to obtain more payoff by extorting his/her opponents. The evaluation results show that the defender can combat this strategy by encouraging competition among the adversaries, and significantly suppress the total payoff of the adversaries via setting the proper penalty factor. Full article
(This article belongs to the Special Issue The Architectures, Systems, and Applications of Internet of Things for Smart Cities)
►▼ Show Figures

Figure 1

Figure 1
<p>The iterated processing in the game model.</p> Full article ">Figure 2
<p>The feasible region for the equalizer strategy [<a href="#B26-sensors-18-00674" class="html-bibr">26</a>] (reproduced with permission from Xinyu Yang, Xiaofei He, Jie Lin, Wei Yu, Qingyu Yang, A Game-Theoretic Model on Coalitional Attacks in Smart Grid; published by IEEE, 2016). (<b>a</b>) Case 1: <math display="inline"> <semantics> <mrow> <mi>r</mi> <mo>&lt;</mo> <mfrac> <mi>N</mi> <mrow> <mi>N</mi> <mo>−</mo> <mn>1</mn> </mrow> </mfrac> </mrow> </semantics> </math>; (<b>b</b>) Case 2: <math display="inline"> <semantics> <mrow> <mfrac> <mi>N</mi> <mrow> <mi>N</mi> <mo>−</mo> <mn>1</mn> </mrow> </mfrac> <mo>&lt;</mo> <mi>r</mi> <mo>&lt;</mo> <mfrac> <mrow> <mi>N</mi> <mo>+</mo> <mi>β</mi> </mrow> <mrow> <mi>N</mi> <mo>−</mo> <mn>1</mn> </mrow> </mfrac> </mrow> </semantics> </math>; (<b>c</b>) Case 3: <math display="inline"> <semantics> <mrow> <mi>r</mi> <mo>&gt;</mo> <mfrac> <mrow> <mi>N</mi> <mo>+</mo> <mi>β</mi> </mrow> <mrow> <mi>N</mi> <mo>−</mo> <mn>1</mn> </mrow> </mfrac> </mrow> </semantics> </math>.</p> Full article ">Figure 3
<p>The payoff of the some typical strategies [<a href="#B26-sensors-18-00674" class="html-bibr">26</a>] (reproduced with permission from Xinyu Yang, Xiaofei He, Jie Lin, Wei Yu, Qingyu Yang, A Game-Theoretic Model on Coalitional Attacks in Smart Grid; published by IEEE, 2016). (<b>a</b>) Win-Stay-Lose-Shift (WSLS) versus Random Strategy; (<b>b</b>) Equalizer versus Random Strategy; (<b>c</b>) all with Equalizer Strategy.</p> Full article ">Figure 4
<p>The payoff of the Adaptive Equalizer (AE) strategy versus other strategies. (<b>a</b>) AE Strategy versus WSLS Strategy; (<b>b</b>) AE Strategy versus Random Strategy; (<b>c</b>) all with AE Strategy.</p> Full article ">Figure 5
<p>The upper bound of <math display="inline"> <semantics> <mi>χ</mi> </semantics> </math> of the extortion strategy. (<b>a</b>) upper bound of <math display="inline"> <semantics> <mi>χ</mi> </semantics> </math> vs. <span class="html-italic">r</span> (<math display="inline"> <semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>0.15</mn> </mrow> </semantics> </math>); (<b>b</b>) upper bound of <math display="inline"> <semantics> <mi>χ</mi> </semantics> </math> vs. <span class="html-italic">r</span> (<math display="inline"> <semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>3</mn> </mrow> </semantics> </math>); (<b>c</b>) upper bound of <math display="inline"> <semantics> <mi>χ</mi> </semantics> </math> vs. <span class="html-italic">N</span> (<math display="inline"> <semantics> <mrow> <mi>r</mi> <mo>=</mo> <mn>1.6</mn> </mrow> </semantics> </math>).</p> Full article ">Figure 6
<p>The payoff of the extortion strategy versus other strategies. (<b>a</b>) Extortion versus WSLS Strategy; (<b>b</b>) Extortion versus Random Strategy; (<b>c</b>) Extortion versus Equalizer Strategy.</p> Full article ">
15 pages, 2196 KiB  
Article
A Feasibility Study on the Use of a Structured Light Depth-Camera for Three-Dimensional Body Measurements of Dairy Cows in Free-Stall Barns
by Andrea Pezzuolo, Marcella Guarino, Luigi Sartori and Francesco Marinello
Sensors 2018, 18(2), 673; https://doi.org/10.3390/s18020673 - 24 Feb 2018
Cited by 68 | Viewed by 8779
Abstract
Frequent checks on livestock’s body growth can help reducing problems related to cow infertility or other welfare implications, and recognizing health’s anomalies. In the last ten years, optical methods have been proposed to extract information on various parameters while avoiding direct contact with [...] Read more.
Frequent checks on livestock’s body growth can help reducing problems related to cow infertility or other welfare implications, and recognizing health’s anomalies. In the last ten years, optical methods have been proposed to extract information on various parameters while avoiding direct contact with animals’ body, generally causes stress. This research aims to evaluate a new monitoring system, which is suitable to frequently check calves and cow’s growth through a three-dimensional analysis of their bodies’ portions. The innovative system is based on multiple acquisitions from a low cost Structured Light Depth-Camera (Microsoft Kinect™ v1). The metrological performance of the instrument is proved through an uncertainty analysis and a proper calibration procedure. The paper reports application of the depth camera for extraction of different body parameters. Expanded uncertainty ranging between 3 and 15 mm is reported in the case of ten repeated measurements. Coefficients of determination R² > 0.84 and deviations lower than 6% from manual measurements where in general detected in the case of head size, hips distance, withers to tail length, chest girth, hips, and withers height. Conversely, lower performances where recognized in the case of animal depth (R² = 0.74) and back slope (R² = 0.12). Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Perspective view of the animal, with relative positions of the three-dimensional (3D) optical sensors.</p> Full article ">Figure 2
<p>The system was implemented in order to allow for a fast extraction and monitoring of different body parameters, and in particular, hip and withers height, back slope, depth, hip distance, head size, and chest girth. Relevant points positions are indicated by red circles.</p> Full article ">Figure 3
<p>In the first row a flow chart representation of the applied methodology. In the bottom series of images three example of data extraction, respectively, for the head size, the hip distance, and chest girth, as indicated by green arrows.</p> Full article ">Figure 4
<p>Assessment between different cow body parameters: (<b>A</b>) hip distance, (<b>B</b>) body length, (<b>C</b>) average height, (<b>D</b>) slope, (<b>E</b>) chest girth, (<b>F</b>) depth, and (<b>G</b>) head length. In the axis labels, “Manual” and “Kinect” refer, respectively, to manual and Kinect measurement.</p> Full article ">Figure 4 Cont.
<p>Assessment between different cow body parameters: (<b>A</b>) hip distance, (<b>B</b>) body length, (<b>C</b>) average height, (<b>D</b>) slope, (<b>E</b>) chest girth, (<b>F</b>) depth, and (<b>G</b>) head length. In the axis labels, “Manual” and “Kinect” refer, respectively, to manual and Kinect measurement.</p> Full article ">Figure 4 Cont.
<p>Assessment between different cow body parameters: (<b>A</b>) hip distance, (<b>B</b>) body length, (<b>C</b>) average height, (<b>D</b>) slope, (<b>E</b>) chest girth, (<b>F</b>) depth, and (<b>G</b>) head length. In the axis labels, “Manual” and “Kinect” refer, respectively, to manual and Kinect measurement.</p> Full article ">
22 pages, 28018 KiB  
Article
An Automatic and Novel SAR Image Registration Algorithm: A Case Study of the Chinese GF-3 Satellite
by Yuming Xiang, Feng Wang and Hongjian You
Sensors 2018, 18(2), 672; https://doi.org/10.3390/s18020672 - 24 Feb 2018
Cited by 27 | Viewed by 5928
Abstract
The Chinese GF-3 satellite launched in August 2016 is a Synthetic Aperture Radar (SAR) satellite that has the largest number of imaging modes in the world. It achieves a free switch in the spotlight, stripmap, scanSAR, wave, global observation and other imaging modes. [...] Read more.
The Chinese GF-3 satellite launched in August 2016 is a Synthetic Aperture Radar (SAR) satellite that has the largest number of imaging modes in the world. It achieves a free switch in the spotlight, stripmap, scanSAR, wave, global observation and other imaging modes. In order to further utilize GF-3 SAR images, an automatic and fast image registration procedure needs to be done. In this paper, we propose a novel image registration technique for GF-3 images of different imaging modes. The proposed algorithm consists of two stages: coarse registration and fine registration. In the first stage, we combine an adaptive sampling method with the SAR-SIFT algorithm to efficiently eliminate obvious translation, rotation and scale differences between the reference and sensed images. In the second stage, uniformly-distributed control points are extracted, then the fast normalized cross-correlation of an improved phase congruency model is utilized as a new similarity metric to match the reference image and the coarse-registered image in a local search region. Moreover, a selection strategy is used to remove outliers. Experimental results on several GF-3 SAR images of different imaging modes show that the proposed algorithm gives a robust, efficient and precise registration performance, compared with other state-of-the-art algorithms for SAR image registration. Full article
(This article belongs to the Special Issue First Experiences with Chinese Gaofen-3 SAR Sensor)
►▼ Show Figures

Figure 1

Figure 1
<p>Four GF-3 images of different imaging modes. (<b>a</b>) Spotlight (SL); (<b>b</b>) Ultra-Fine Stripmap (UFSM); (<b>c</b>) Fine Stripmap One (FSMI; (<b>d</b>) Quadrupolarization Stripmap One (QPSMI).</p> Full article ">Figure 2
<p>The flowchart of the proposed algorithm. PC, Phase Congruency; NCC, Normalized Cross-Correlation.</p> Full article ">Figure 3
<p>Three processing window. (<b>a</b>) Circle; (<b>b</b>) vertical; (<b>c</b>) horizontal.</p> Full article ">Figure 4
<p>Comparison of the similarity curves. (<b>a</b>) The reference image; (<b>b</b>) MI; (<b>c</b>) fast NCC (fNCC); (<b>d</b>) the sensed image; (<b>e</b>) ISPCC; (<b>f</b>) LSCC.</p> Full article ">Figure 5
<p>Registration results of the first image pair. (<b>a</b>) The reference image; (<b>b</b>) the sensed image; (<b>c</b>) the coarse registration result; (<b>d</b>) the fine registration result.</p> Full article ">Figure 6
<p>Registration results of the second image pair. (<b>a</b>) The reference image; (<b>b</b>) the sensed image; (<b>c</b>) the coarse registration result; (<b>d</b>) the fine registration result.</p> Full article ">Figure 7
<p>Registration results of the third image pair. (<b>a</b>) The reference image; (<b>b</b>) the sensed image; (<b>c</b>) the coarse registration result; (<b>d</b>) the fine registration result.</p> Full article ">Figure 8
<p>The enlarged subimages of the first image pair. (<b>a</b>) First subimage of coarse registration; (<b>b</b>) first subimage of fine registration; (<b>c</b>) second subimage of coarse registration; (<b>d</b>) second subimage of fine registration.</p> Full article ">Figure 9
<p>The enlarged subimages of the second image pair. (<b>a</b>) First subimage of coarse registration; (<b>b</b>) first subimage of fine registration; (<b>c</b>) second subimage of coarse registration; (<b>d</b>) second subimage of fine registration.</p> Full article ">Figure 10
<p>The enlarged subimages of the third image pair. (<b>a</b>) First subimage of coarse registration; (<b>b</b>) first subimage of fine registration; (<b>c</b>) second subimage of coarse registration; (<b>d</b>) second subimage of fine registration.</p> Full article ">Figure 11
<p>The comparative results of the second image pair. (<b>a</b>) The proposed method; (<b>b</b>) SAR-SIFT; (<b>c</b>) BFSIFT; (<b>d</b>) SIFT; (<b>e</b>–<b>h</b>) the first enlarged images of the proposed method, SAR-SIFT, BFSIFT, SIFT; (<b>i</b>–<b>l</b>) the second enlarged images.</p> Full article ">Figure 11 Cont.
<p>The comparative results of the second image pair. (<b>a</b>) The proposed method; (<b>b</b>) SAR-SIFT; (<b>c</b>) BFSIFT; (<b>d</b>) SIFT; (<b>e</b>–<b>h</b>) the first enlarged images of the proposed method, SAR-SIFT, BFSIFT, SIFT; (<b>i</b>–<b>l</b>) the second enlarged images.</p> Full article ">Figure 12
<p>Detection results of three algorithms on the FSMI image. (<b>a</b>) The raw image; (<b>b</b>) IS-PC; (<b>c</b>) PC; (<b>d</b>) PC on the logarithm of the raw image.</p> Full article ">Figure 12 Cont.
<p>Detection results of three algorithms on the FSMI image. (<b>a</b>) The raw image; (<b>b</b>) IS-PC; (<b>c</b>) PC; (<b>d</b>) PC on the logarithm of the raw image.</p> Full article ">Figure 13
<p>Comparisons of NCC, MI, LSCC and ISPCC on the second image pair. (<b>a</b>) Correctly Matching Rate (CMR) versus template size; (<b>b</b>) running time versus template size.</p> Full article ">
17 pages, 10768 KiB  
Article
Study of Impact Damage in PVA-ECC Beam under Low-Velocity Impact Loading Using Piezoceramic Transducers and PVDF Thin-Film Transducers
by Baoxin Qi, Qingzhao Kong, Hui Qian, Devendra Patil, Ing Lim, Mo Li, Dong Liu and Gangbing Song
Sensors 2018, 18(2), 671; https://doi.org/10.3390/s18020671 - 24 Feb 2018
Cited by 45 | Viewed by 5473
Abstract
Compared to conventional concrete, polyvinyl alcohol fiber reinforced engineering cementitious composite (PVA-ECC) offers high-strength, ductility, formability, and excellent fatigue resistance. However, impact-induced structural damage is a major concern and has not been previously characterized in PVA-ECC structures. We investigate the damage of PVA-ECC [...] Read more.
Compared to conventional concrete, polyvinyl alcohol fiber reinforced engineering cementitious composite (PVA-ECC) offers high-strength, ductility, formability, and excellent fatigue resistance. However, impact-induced structural damage is a major concern and has not been previously characterized in PVA-ECC structures. We investigate the damage of PVA-ECC beams under low-velocity impact loading. A series of ball-drop impact tests were performed at different drop weights and heights to simulate various impact energies. The impact results of PVA-ECC beams were compared with mortar beams. A combination of polyvinylidene fluoride (PVDF) thin-film sensors and piezoceramic-based smart aggregate were used for impact monitoring, which included impact initiation and crack evolution. Short-time Fourier transform (STFT) of the signal received by PVDF thin-film sensors was performed to identify impact events, while active-sensing approach was utilized to detect impact-induced crack evolution by the attenuation of a propagated guided wave. Wavelet packet-based energy analysis was performed to quantify failure development under repeated impact tests. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Embedded polyvinylidene fluoride (PVDF) thin-film sensor: (<b>a</b>) photo and (<b>b</b>) internal structure.</p> Full article ">Figure 2
<p>Smart aggregate (<b>a</b>) photo and (<b>b</b>) internal structure.</p> Full article ">Figure 3
<p>Detailed sensor location (unit: mm).</p> Full article ">Figure 4
<p>Specimen preparation. (<b>a</b>) molds with pre-installed PVDF sensors and smart aggregates (<b>b</b>) casted PVA-ECC and mortar beams.</p> Full article ">Figure 5
<p>Low-velocity impact test setup (<b>a</b>) test setup; (<b>b</b>) data acquisition system.</p> Full article ">Figure 6
<p>Short-time Fourier transform (STFT) of PVDF sensor signal for PVA-ECC 1.</p> Full article ">Figure 7
<p>STFT of PVDF sensor signal for PVA-ECC 2.</p> Full article ">Figure 8
<p>STFT of PVDF sensor signal for PVA-ECC 3.</p> Full article ">Figure 9
<p>STFT of PVDF sensor signal for PVA-ECC 3.</p> Full article ">Figure 10
<p>Results of crack detection in PVA-ECC-1 with high-speed video.</p> Full article ">Figure 11
<p>Time-domain signal received by smart aggregate (SA) sensor in PVA-ECC 1.</p> Full article ">Figure 12
<p>Wavelet packet-based energy plots of the received signal in PVA-ECC 1.</p> Full article ">Figure 13
<p>Results of crack detection in PVA-ECC-2 with high-speed video.</p> Full article ">Figure 14
<p>Time-domain signal received by SA sensor in PVA-ECC-2.</p> Full article ">Figure 15
<p>Wavelet packet-based energy plots of the received signal in PVA-ECC-2.</p> Full article ">Figure 16
<p>Results of crack detection in PVA-ECC-3 with high-speed video.</p> Full article ">Figure 17
<p>Time-domain signal received by SA sensor in PVA-ECC-3.</p> Full article ">Figure 18
<p>Wavelet packet-based energy plots of the received signal in PVA-ECC-3.</p> Full article ">Figure 19
<p>Results of crack detection of Mortar-1 with high-speed video.</p> Full article ">Figure 20
<p>Time-domain signal received by SA sensor in Mortar-1.</p> Full article ">Figure 21
<p>Wavelet packet-based energy plots of the received signal in Mortar-1.</p> Full article ">Figure 22
<p>Results of crack detection of Mortar-2 with high-speed video.</p> Full article ">Figure 23
<p>Time-domain signal received by SA sensor in Mortar-2.</p> Full article ">Figure 24
<p>Wavelet packet-based energy plots of the received signal in Mortar-2.</p> Full article ">Figure 25
<p>Results of crack detection of Mortar-3 with high-speed video.</p> Full article ">Figure 26
<p>Time-domain signal received by SA sensor in Mortar-2.</p> Full article ">Figure 27
<p>Wavelet packet-based energy plots of the received signal in Mortar-2.</p> Full article ">
12 pages, 2428 KiB  
Article
Error Analysis of the K-Rb-21Ne Comagnetometer Space-Stable Inertial Navigation System
by Qingzhong Cai, Gongliu Yang, Wei Quan, Ningfang Song, Yongqiang Tu and Yiliang Liu
Sensors 2018, 18(2), 670; https://doi.org/10.3390/s18020670 - 24 Feb 2018
Cited by 10 | Viewed by 5636
Abstract
According to the application characteristics of the K-Rb-21Ne comagnetometer, a space-stable navigation mechanization is designed and the requirements of the comagnetometer prototype are presented. By analysing the error propagation rule of the space-stable Inertial Navigation System (INS), the three biases, the [...] Read more.
According to the application characteristics of the K-Rb-21Ne comagnetometer, a space-stable navigation mechanization is designed and the requirements of the comagnetometer prototype are presented. By analysing the error propagation rule of the space-stable Inertial Navigation System (INS), the three biases, the scale factor of the z-axis, and the misalignment of the x- and y-axis non-orthogonal with the z-axis, are confirmed to be the main error source. A numerical simulation of the mathematical model for each single error verified the theoretical analysis result of the system’s error propagation rule. Thus, numerical simulation based on the semi-physical data result proves the feasibility of the navigation scheme proposed in this paper. Full article
(This article belongs to the Special Issue Inertial Sensors for Positioning and Navigation)
►▼ Show Figures

Figure 1

Figure 1
<p>Structure of the space-stable platform.</p> Full article ">Figure 2
<p>Mechanization of space-stable platform Inertial Navigation System (INS).</p> Full article ">Figure 3
<p>Single error test result in a numerical simulation of the mathematical model.</p> Full article ">Figure 4
<p>Numerical simulation based on semi-physical data.</p> Full article ">Figure 5
<p>Raw data of the comagnetometer.</p> Full article ">Figure 6
<p>Position error in numerical simulation based on semi-physical data.</p> Full article ">
10 pages, 5972 KiB  
Article
The Design of a Single-Bit CMOS Image Sensor for Iris Recognition Applications
by Keunyeol Park, Minkyu Song and Soo Youn Kim
Sensors 2018, 18(2), 669; https://doi.org/10.3390/s18020669 - 24 Feb 2018
Cited by 12 | Viewed by 5846
Abstract
This paper presents a single-bit CMOS image sensor (CIS) that uses a data processing technique with an edge detection block for simple iris segmentation. In order to recognize the iris image, the image sensor conventionally captures high-resolution image data in digital code, extracts [...] Read more.
This paper presents a single-bit CMOS image sensor (CIS) that uses a data processing technique with an edge detection block for simple iris segmentation. In order to recognize the iris image, the image sensor conventionally captures high-resolution image data in digital code, extracts the iris data, and then compares it with a reference image through a recognition algorithm. However, in this case, the frame rate decreases by the time required for digital signal conversion of multi-bit digital data through the analog-to-digital converter (ADC) in the CIS. In order to reduce the overall processing time as well as the power consumption, we propose a data processing technique with an exclusive OR (XOR) logic gate to obtain single-bit and edge detection image data instead of multi-bit image data through the ADC. In addition, we propose a logarithmic counter to efficiently measure single-bit image data that can be applied to the iris recognition algorithm. The effective area of the proposed single-bit image sensor (174 × 144 pixel) is 2.84 mm2 with a 0.18 μm 1-poly 4-metal CMOS image sensor process. The power consumption of the proposed single-bit CIS is 2.8 mW with a 3.3 V of supply voltage and 520 frame/s of the maximum frame rates. The error rate of the ADC is 0.24 least significant bit (LSB) on an 8-bit ADC basis at a 50 MHz sampling frequency. Full article
(This article belongs to the Special Issue Image Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>A conventional iris recognition process, adopted from [<a href="#B4-sensors-18-00669" class="html-bibr">4</a>].</p> Full article ">Figure 2
<p>The proposed CMOS image sensor (CIS) with an iris recognition algorithm.</p> Full article ">Figure 3
<p>A block diagram of the proposed iris recognition sensor.</p> Full article ">Figure 4
<p>A single column schematic of the proposed CIS.</p> Full article ">Figure 5
<p>Simulation results of the 1-bit clocked comparator with V<sub>PP</sub> = 1 V, 50 MHz of sampling frequency.</p> Full article ">Figure 6
<p>Example of static random access memory (SRAM) word-line signals.</p> Full article ">Figure 7
<p>XOR output of two different images from dual-SRAM.</p> Full article ">Figure 8
<p>Edge (boundary) detection with the XOR gate.</p> Full article ">Figure 9
<p>Pixel data distribution at an 8-bit image.</p> Full article ">Figure 10
<p>Iris region at the ramp signal (V<sub>REF</sub>).</p> Full article ">Figure 11
<p>SRAM data region when pulse signal gap decreases exponentially.</p> Full article ">Figure 12
<p>Layout of the proposed CIS.</p> Full article ">Figure 13
<p>Fabricated chip on a PCB board.</p> Full article ">Figure 14
<p>Iris recognition processing using the proposed CIS.</p> Full article ">Figure 15
<p>Images based on the conventional iris recognition algorithms and the proposed CIS.</p> Full article ">
13 pages, 1160 KiB  
Article
Mn4+-Doped Magnesium Titanate—A Promising Phosphor for Self-Referenced Optical Temperature Sensing
by Francesca Venturini, Michael Baumgartner and Sergey M. Borisov
Sensors 2018, 18(2), 668; https://doi.org/10.3390/s18020668 - 24 Feb 2018
Cited by 18 | Viewed by 5910
Abstract
Phosphors based on magnesium titanate activated with Mn 4 + ions are of great interest because, when excited with blue light, they display a strong red-emitting luminescence. They are characterized by a luminescence decay which is strongly temperature dependent in the range from [...] Read more.
Phosphors based on magnesium titanate activated with Mn 4 + ions are of great interest because, when excited with blue light, they display a strong red-emitting luminescence. They are characterized by a luminescence decay which is strongly temperature dependent in the range from −50 ∘ C to 150 ∘ C, making these materials very promising for temperature sensing in the biochemical field. In this work, the optical and thermal properties of the luminescence of Mg 2 TiO 4 are investigated for different Mn 4 + doping concentrations. The potential of this material for temperature sensing is demonstrated by fabricating a fiber optic temperature microsensor and by comparing its performance against a standard resistance thermometer. The response of the fiber optic sensor is exceptionally fast, with a response time below 1 s in the liquid phase and below 1.1 s in the gas phase. Full article
(This article belongs to the Special Issue Lab on Fiber Optrodes for Chemical and Biological Sensing: Recent Trends and Advances)
►▼ Show Figures

Figure 1

Figure 1
<p>Scheme of the optical experimental setup. Blue is the excitation, red is the luminescence optical path. PMT: photomultiplier.</p> Full article ">Figure 2
<p>Left: Photographic images of the compact phase fluorometer from PyroScience. Right: from left to right: photographic image of resistance thermometer PT-100, fiber optic temperature microsensor and 1 euro cent for size comparison.</p> Full article ">Figure 3
<p>Emission spectrum of the sample with a Mn<math display="inline"> <semantics> <msup> <mrow/> <mrow> <mn>4</mn> <mo>+</mo> </mrow> </msup> </semantics> </math> doping concentration of 0.40% at different temperatures.</p> Full article ">Figure 4
<p>Normalized luminescence intensity decay at 20 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C for samples with different doping concentrations.</p> Full article ">Figure 5
<p>Upper panel: Time-domain intensity decay time at 20 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C for the sample with 0.20% doping. The red solid line (<b>a</b>) shows the monoexponential function fit with <math display="inline"> <semantics> <msubsup> <mi>χ</mi> <mi>R</mi> <mn>2</mn> </msubsup> </semantics> </math> = 1.97. The red green line (<b>b</b>) shows the double exponential function fit with <math display="inline"> <semantics> <msubsup> <mi>χ</mi> <mi>R</mi> <mn>2</mn> </msubsup> </semantics> </math> = 1.23. Lower panels: deviation plots for the mono- (<b>a</b>) and double (<b>b</b>) exponential fits.</p> Full article ">Figure 6
<p>Luminescence intensity decay of the sample with a Mn<math display="inline"> <semantics> <msup> <mrow/> <mrow> <mn>4</mn> <mo>+</mo> </mrow> </msup> </semantics> </math> doping concentration of 0.40% for temperatures the 4 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C, 12 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C, 29 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C, 45 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C, 61 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C and 77 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C. The direction of the increasing temperatures is also shown.</p> Full article ">Figure 7
<p>(<b>A</b>) Temperature dependence of the monoexponential lifetime for samples with different Mn<math display="inline"> <semantics> <msup> <mrow/> <mrow> <mn>4</mn> <mo>+</mo> </mrow> </msup> </semantics> </math> doping concentrations; (<b>B</b>) Temperature dependence of a sample with 0.40% doping concentration in a broader temperature range obtained in the frequency domain. Points: Measurements, solid lines: linear fit. In B, the linear fit is performed in the temperature range from −30 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C to 80 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C.</p> Full article ">Figure 8
<p>Temperature dependence of the relative sensitivity <span class="html-italic">s</span> for different Mn<math display="inline"> <semantics> <msup> <mrow/> <mrow> <mn>4</mn> <mo>+</mo> </mrow> </msup> </semantics> </math> doping concentrations.</p> Full article ">Figure 9
<p>Response of the luminescence decay time of the fiber optic temperature microsensor (black trace). The temperature in the calibration chamber is measured by PT-100 resistance thermometer (red trace).</p> Full article ">Figure 10
<p>Temperature dependency of the luminescence decay time for the fiber optic temperature microsensor. The insert shows the relative sensitivity.</p> Full article ">Figure 11
<p>Dynamic response of the temperature resistance thermometer PT-100 (<b>A</b>) and fiber optic microsensor (<b>B</b>,<b>C</b>) to temperature changes between the beakers with water kept at 55 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C, 22 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C and ∼1.3 <math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math>C (ice water). C is the zoom-in for the interval 195–210 s corresponding to the curve B.</p> Full article ">
8 pages, 4338 KiB  
Article
Second Generation Small Pixel Technology Using Hybrid Bond Stacking
by Vincent C. Venezia, Alan Chih-Wei Hsiung, Wu-Zang Yang, Yuying Zhang, Cheng Zhao, Zhiqiang Lin and Lindsay A. Grant
Sensors 2018, 18(2), 667; https://doi.org/10.3390/s18020667 - 24 Feb 2018
Cited by 21 | Viewed by 9700
Abstract
In this work, OmniVision’s second generation (Gen2) of small-pixel BSI stacking technologies is reviewed. The key features of this technology are hybrid-bond stacking, deeper back-side, deep-trench isolation, new back-side composite metal-oxide grid, and improved gate oxide quality. This Gen2 technology achieves state-of-the-art low-light [...] Read more.
In this work, OmniVision’s second generation (Gen2) of small-pixel BSI stacking technologies is reviewed. The key features of this technology are hybrid-bond stacking, deeper back-side, deep-trench isolation, new back-side composite metal-oxide grid, and improved gate oxide quality. This Gen2 technology achieves state-of-the-art low-light image-sensor performance for 1.1, 1.0, and 0.9 µm pixel products. Additional improvements on this technology include less than 100 ppm white-pixel process and a high near-infrared (NIR) QE technology. Full article
(This article belongs to the Special Issue Special Issue on the 2017 International Image Sensor Workshop (IISW))
►▼ Show Figures

Figure 1

Figure 1
<p>First-generation BSI-CIS stacking with oxide-oxide bonding and TSV.</p> Full article ">Figure 2
<p>Hybrid bond-stacking schematic, showing the Cu-Cu interconnects outside the array (<b>a</b>) and within the array (<b>b</b>).</p> Full article ">Figure 3
<p>Gen1 (<b>a</b>) and Gen2 (<b>b</b>) schematic; Gen2 BSI stack has thicker silicon, narrower and deeper BS-DTI, and a composite metal-oxide, back-side grid.</p> Full article ">Figure 4
<p>QE comparing Gen1 and Gen2 1 µm, 16 MP technologies.</p> Full article ">Figure 5
<p>(<b>a</b>) Crosstalk vs angle, calculated from QE curves measured in a fix ROI along a diagonal as shown in the (<b>b</b>). Angles are the CRA of the module lens at each ROI.</p> Full article ">Figure 6
<p>The QE from 1 µm and 0.9 µm pixel products, both using the Gen2 BSI-stacking technology.</p> Full article ">Figure 7
<p>FDTD simulation comparing the light confinement of Gen2 (<b>left</b>) and Gen1 (<b>right</b>). Simulations used a monochromatic plan wave at, 630 nm TE mode, incident on the Green and Red pixel of each structure. Structure features are labeled and highlighted.</p> Full article ">Figure 8
<p>Two-frame difference image, noise histogram comparing the 1 µm, 16 MP images sensors that use the Gen1 and Gen2 technology.</p> Full article ">Figure 9
<p>Plot of 1-cumulative density function (1-CDF) comparing the dark image from low white-pixel, PD process. 1 µm, 16 MP (2 fps, 60 C).</p> Full article ">Figure 10
<p>QE comparison from a 1 µm, 16MP Gen2 image sensor to one with an additional process for NIR enhancement.</p> Full article ">Figure 11
<p>(<b>a</b>) Combined Macbeth chart images from 16 MP, 1 µm pixel Gen1 and Gen2 products to show the Gen2 low light (5 Lux) sensitivity improvement. (<b>b</b>) Gen2 SNR improvement vs. lux level, including measured data and simulated results (F2.0, 15 fps).</p> Full article ">
16 pages, 1399 KiB  
Article
SmartVeh: Secure and Efficient Message Access Control and Authentication for Vehicular Cloud Computing
by Qinlong Huang, Yixian Yang and Yuxiang Shi
Sensors 2018, 18(2), 666; https://doi.org/10.3390/s18020666 - 24 Feb 2018
Cited by 18 | Viewed by 5569
Abstract
With the growing number of vehicles and popularity of various services in vehicular cloud computing (VCC), message exchanging among vehicles under traffic conditions and in emergency situations is one of the most pressing demands, and has attracted significant attention. However, it is an [...] Read more.
With the growing number of vehicles and popularity of various services in vehicular cloud computing (VCC), message exchanging among vehicles under traffic conditions and in emergency situations is one of the most pressing demands, and has attracted significant attention. However, it is an important challenge to authenticate the legitimate sources of broadcast messages and achieve fine-grained message access control. In this work, we propose SmartVeh, a secure and efficient message access control and authentication scheme in VCC. A hierarchical, attribute-based encryption technique is utilized to achieve fine-grained and flexible message sharing, which ensures that vehicles whose persistent or dynamic attributes satisfy the access policies can access the broadcast message with equipped on-board units (OBUs). Message authentication is enforced by integrating an attribute-based signature, which achieves message authentication and maintains the anonymity of the vehicles. In order to reduce the computations of the OBUs in the vehicles, we outsource the heavy computations of encryption, decryption and signing to a cloud server and road-side units. The theoretical analysis and simulation results reveal that our secure and efficient scheme is suitable for VCC. Full article
(This article belongs to the Special Issue Security, Trust and Privacy for Sensor Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>System framework of SmartVeh.</p> Full article ">Figure 2
<p>Computation cost of key generation on attribute authority.</p> Full article ">Figure 3
<p>Computation cost of message broadcasting for on-board unit.</p> Full article ">Figure 4
<p>Computation cost of message decryption for on-board unit.</p> Full article ">
9 pages, 2998 KiB  
Article
High-Speed Interrogation for Large-Scale Fiber Bragg Grating Sensing
by Chenyuan Hu and Wei Bai
Sensors 2018, 18(2), 665; https://doi.org/10.3390/s18020665 - 24 Feb 2018
Cited by 13 | Viewed by 4762
Abstract
A high-speed interrogation scheme for large-scale fiber Bragg grating (FBG) sensing arrays is presented. This technique employs parallel computing and pipeline control to modulate incident light and demodulate the reflected sensing signal. One Electro-optic modulator (EOM) and one semiconductor optical amplifier (SOA) were [...] Read more.
A high-speed interrogation scheme for large-scale fiber Bragg grating (FBG) sensing arrays is presented. This technique employs parallel computing and pipeline control to modulate incident light and demodulate the reflected sensing signal. One Electro-optic modulator (EOM) and one semiconductor optical amplifier (SOA) were used to generate a phase delay to filter reflected spectrum form multiple candidate FBGs with the same optical path difference (OPD). Experimental results showed that the fastest interrogation delay time for the proposed method was only about 27.2 us for a single FBG interrogation, and the system scanning period was only limited by the optical transmission delay in the sensing fiber owing to the multiple simultaneous central wavelength calculations. Furthermore, the proposed FPGA-based technique had a verified FBG wavelength demodulation stability of ±1 pm without average processing. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Illustration of interrogation system.</p> Full article ">Figure 2
<p>Timing of InGaAs LIS detector and A/D sampling.</p> Full article ">Figure 3
<p>Reflected spectrum of weak FBG.</p> Full article ">Figure 3 Cont.
<p>Reflected spectrum of weak FBG.</p> Full article ">Figure 4
<p>Central wavelength demodulation when temperature changes.</p> Full article ">Figure 5
<p>Logarithm calculation result.</p> Full article ">Figure 6
<p>Executive timing of central wavelength demodulation.</p> Full article ">Figure 7
<p>Scanning speed of each sensor on FBG array.</p> Full article ">Figure 8
<p>Temperature measurement of <math display="inline"> <semantics> <mrow> <msub> <mrow> <mi>FBG</mi> </mrow> <mrow> <mn>1</mn> <mo>,</mo> <mn>143</mn> </mrow> </msub> <mo>.</mo> </mrow> </semantics> </math></p> Full article ">Figure 9
<p>Central wavelength shift vs. temperature change of <math display="inline"> <semantics> <mrow> <msub> <mrow> <mi>FBG</mi> </mrow> <mrow> <mn>1</mn> <mo>,</mo> <mn>143</mn> </mrow> </msub> <mo>.</mo> </mrow> </semantics> </math></p> Full article ">Figure 10
<p>Logarithm calculation error.</p> Full article ">
14 pages, 2392 KiB  
Article
User Access Management Based on Network Pricing for Social Network Applications
by Fuhong Lin, Zhibo Pang, Xingmin Ma and Qing Gu
Sensors 2018, 18(2), 664; https://doi.org/10.3390/s18020664 - 24 Feb 2018
Cited by 3 | Viewed by 4124
Abstract
Social applications play a very important role in people’s lives, as users communicate with each other through social networks on a daily basis. This presents a challenge: How does one receive high-quality service from social networks at a low cost? Users can access [...] Read more.
Social applications play a very important role in people’s lives, as users communicate with each other through social networks on a daily basis. This presents a challenge: How does one receive high-quality service from social networks at a low cost? Users can access different kinds of wireless networks from various locations. This paper proposes a user access management strategy based on network pricing such that networks can increase its income and improve service quality. Firstly, network price is treated as an optimizing access parameter, and an unascertained membership algorithm is used to make pricing decisions. Secondly, network price is adjusted dynamically in real time according to network load. Finally, selecting a network is managed and controlled in terms of the market economy. Simulation results show that the proposed scheme can effectively balance network load, reduce network congestion, improve the user's quality of service (QoS) requirements, and increase the network’s income. Full article
(This article belongs to the Special Issue Sensor Intelligent Data Analysis for Social Networks: Theory and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>The integrated network model.</p> Full article ">Figure 2
<p>Grades of unascertained membership.</p> Full article ">Figure 3
<p>The uncertainty membership processing procedures.</p> Full article ">Figure 4
<p>(<b>a</b>) Cellular network signals membership function; (<b>b</b>) WLAN (wireless local area network) receiving signals membership function; (<b>c</b>) network traffic membership function; (<b>d</b>) collect fees membership function; (<b>e</b>) cellular network load membership function; (<b>f</b>) WLAN load membership function.</p> Full article ">Figure 5
<p>The network pricing controlling process.</p> Full article ">Figure 6
<p>The integrated network congestion rate with smaller wireless network bandwidth.</p> Full article ">Figure 7
<p>The integrated network congestion rate with larger wireless network bandwidth.</p> Full article ">Figure 8
<p>The integrated network income.</p> Full article ">Figure 9
<p>The ratio of congestion rate and network income.</p> Full article ">Figure 10
<p>The ratio of congestion rate and network income.</p> Full article ">
15 pages, 5821 KiB  
Article
Estimation of Handgrip Force from SEMG Based on Wavelet Scale Selection
by Kai Wang, Xianmin Zhang, Jun Ota and Yanjiang Huang
Sensors 2018, 18(2), 663; https://doi.org/10.3390/s18020663 - 24 Feb 2018
Cited by 23 | Viewed by 5390
Abstract
This paper proposes a nonlinear correlation-based wavelet scale selection technology to select the effective wavelet scales for the estimation of handgrip force from surface electromyograms (SEMG). The SEMG signal corresponding to gripping force was collected from extensor and flexor forearm muscles during the [...] Read more.
This paper proposes a nonlinear correlation-based wavelet scale selection technology to select the effective wavelet scales for the estimation of handgrip force from surface electromyograms (SEMG). The SEMG signal corresponding to gripping force was collected from extensor and flexor forearm muscles during the force-varying analysis task. We performed a computational sensitivity analysis on the initial nonlinear SEMG-handgrip force model. To explore the nonlinear correlation between ten wavelet scales and handgrip force, a large-scale iteration based on the Monte Carlo simulation was conducted. To choose a suitable combination of scales, we proposed a rule to combine wavelet scales based on the sensitivity of each scale and selected the appropriate combination of wavelet scales based on sequence combination analysis (SCA). The results of SCA indicated that the scale combination VI is suitable for estimating force from the extensors and the combination V is suitable for the flexors. The proposed method was compared to two former methods through prolonged static and force-varying contraction tasks. The experiment results showed that the root mean square errors derived by the proposed method for both static and force-varying contraction tasks were less than 20%. The accuracy and robustness of the handgrip force derived by the proposed method is better than that obtained by the former methods. Full article
(This article belongs to the Special Issue Sensors for Gait, Posture, and Health Monitoring)
►▼ Show Figures

Figure 1

Figure 1
<p>Experimental setup. SEMG: surface electromyogram.</p> Full article ">Figure 2
<p>The process of the force-varying analysis task.</p> Full article ">Figure 3
<p>The process of the force-varying validation task.</p> Full article ">Figure 4
<p>The results of the convergence experiment for determining the number of random variables.</p> Full article ">Figure 5
<p>Sensitivity distribution of the ten wavelet scales to handgrip force.</p> Full article ">Figure 6
<p>Sequence combination of wavelet scales based on the sensitivity value.</p> Full article ">Figure 7
<p>Varying RMSE process of each wavelet scale combination (WSC) in the force-varying validation task.</p> Full article ">Figure 8
<p>Comparison of the ten WSCs.</p> Full article ">Figure 9
<p>Results of the force-varying validation tasks.</p> Full article ">Figure 10
<p>Results of the static validation tasks.</p> Full article ">
24 pages, 5120 KiB  
Article
Leader-Follower Formation Control of UUVs with Model Uncertainties, Current Disturbances, and Unstable Communication
by Zheping Yan, Da Xu, Tao Chen, Wei Zhang and Yibo Liu
Sensors 2018, 18(2), 662; https://doi.org/10.3390/s18020662 - 23 Feb 2018
Cited by 45 | Viewed by 5970
Abstract
Unmanned underwater vehicles (UUVs) have rapidly developed as mobile sensor networks recently in the investigation, survey, and exploration of the underwater environment. The goal of this paper is to develop a practical and efficient formation control method to improve work efficiency of multi-UUV [...] Read more.
Unmanned underwater vehicles (UUVs) have rapidly developed as mobile sensor networks recently in the investigation, survey, and exploration of the underwater environment. The goal of this paper is to develop a practical and efficient formation control method to improve work efficiency of multi-UUV sensor networks. Distributed leader-follower formation controllers are designed based on a state feedback and consensus algorithm. Considering that each vehicle is subject to model uncertainties and current disturbances, a second-order integral UUV model with a nonlinear function is established using the state feedback linearized method under current disturbances. For unstable communication among UUVs, communication failure and acoustic link noise interference are considered. Two-layer random switching communication topologies are proposed to solve the problem of communication failure. For acoustic link noise interference, accurate representation of valid communication information and noise stripping when designing controllers is necessary. Effective communication topology weights are designed to represent the validity of communication information interfered by noise. Utilizing state feedback and noise stripping, sufficient conditions for design formation controllers are proposed to ensure UUV formation achieves consensus under model uncertainties, current disturbances, and unstable communication. The stability of formation controllers is proven by the Lyapunov-Razumikhin theorem, and the validity is verified by simulation results. Full article
(This article belongs to the Special Issue Sensors, Wireless Connectivity and Systems for Autonomous Vehicles and Smart Mobility)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The multi-UUV sensor network.</p> Full article ">Figure 2
<p>Model uncertainties, current disturbances, and unstable communication for UUV formation.</p> Full article ">Figure 3
<p>The leader-follower UUV formation.</p> Full article ">Figure 4
<p>Switching topologies: (<b>a</b>) the two-layer random switching topology set; and (<b>b</b>) the Markov random states in switching topology.</p> Full article ">Figure 5
<p>3D trajectory of the leader-follower UUV formation.</p> Full article ">Figure 6
<p>The trajectory of leader and follower UUVs in 2D with the desired triangle formation structure.</p> Full article ">Figure 7
<p>Position and attitude states of the UUVs: (<b>a</b>) state <math display="inline"> <semantics> <mi>x</mi> </semantics> </math> of each UUV; (<b>b</b>) state <math display="inline"> <semantics> <mi>y</mi> </semantics> </math> of each UUV; (<b>c</b>) state <math display="inline"> <semantics> <mi>z</mi> </semantics> </math> of each UUV; (<b>d</b>) pitch <math display="inline"> <semantics> <mi>θ</mi> </semantics> </math> of each UUV; and (<b>e</b>) heading <math display="inline"> <semantics> <mi>ψ</mi> </semantics> </math> of each UUV.</p> Full article ">Figure 8
<p>Velocity states of the UUVs: (<b>a</b>) velocity <math display="inline"> <semantics> <mi>u</mi> </semantics> </math> of each UUV; (<b>b</b>) velocity <math display="inline"> <semantics> <mi>v</mi> </semantics> </math> of each UUV; (<b>c</b>) velocity <math display="inline"> <semantics> <mi>w</mi> </semantics> </math> of each UUV; (<b>d</b>) angular velocity <math display="inline"> <semantics> <mi>q</mi> </semantics> </math> of each UUV; and (<b>e</b>) angular velocity <math display="inline"> <semantics> <mi>r</mi> </semantics> </math> of each UUV.</p> Full article ">
12 pages, 5171 KiB  
Article
Metal Nanoparticles/Porous Silicon Microcavity Enhanced Surface Plasmon Resonance Fluorescence for the Detection of DNA
by Jiajia Wang and Zhenhong Jia
Sensors 2018, 18(2), 661; https://doi.org/10.3390/s18020661 - 23 Feb 2018
Cited by 23 | Viewed by 5550
Abstract
A porous silicon microcavity (PSiMC) with resonant peak wavelength of 635 nm was fabricated by electrochemical etching. Metal nanoparticles (NPs)/PSiMC enhanced fluorescence substrates were prepared by the electrostatic adherence of Au NPs that were distributed in PSiMC. The Au NPs/PSiMC device was used [...] Read more.
A porous silicon microcavity (PSiMC) with resonant peak wavelength of 635 nm was fabricated by electrochemical etching. Metal nanoparticles (NPs)/PSiMC enhanced fluorescence substrates were prepared by the electrostatic adherence of Au NPs that were distributed in PSiMC. The Au NPs/PSiMC device was used to characterize the target DNA immobilization and hybridization with its complementary DNA sequences marked with Rhodamine red (RRA). Fluorescence enhancement was observed on the Au NPs/PSiMC device substrate; and the minimum detection concentration of DNA ran up to 10 pM. The surface plasmon resonance (SPR) of the MC substrate; which is so well-positioned to improve fluorescence enhancement rather the fluorescence enhancement of the high reflection band of the Bragg reflector; would welcome such a highly sensitive in biosensor. Full article
(This article belongs to the Special Issue Recent Advances in Nucleic Acid Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>The flow-process diagram of DNA detection on metal NPs /PSiMC device. (I) Au NPs adsorbed on functionalized PSiMC; (II) the immobilization of target DNA on Au NPs /PSiMC device substrate; (III) the hybridization of probe RRA-DNA sequences with target DNA.</p> Full article ">Figure 2
<p>The cross-section image of PSiMC.</p> Full article ">Figure 3
<p>AFM image of PSi layer.</p> Full article ">Figure 4
<p>(<b>a</b>) The UV-Vis spectra of Au NPs, and (<b>b</b>) surface morphology of PSiMC after deposition of Au NPs.</p> Full article ">Figure 5
<p>The UV-Vis absorption spectra of RRA-DNA.</p> Full article ">Figure 6
<p>The absorption spectrum and emission spectrum of RRA-DNA fragment.</p> Full article ">Figure 7
<p>The reflection spectra vary with the preparation process of (<b>a</b>) PSiMC biosensor, and (<b>b</b>) Bragg reflector sensor.</p> Full article ">Figure 8
<p>The fluorescence spectra of RRA-DNA obtained on Au NPs/PSiMC sensor substrate and PSiMC substrate without Au NPs (the concentration of RRA-DNA is 1 µM). The inset map shows (<b>a</b>) the absorption spectrum of Au NPs and (<b>b</b>) the emission spectrum of RRA.</p> Full article ">Figure 9
<p>The movement of the reflection spectra from the probe DNA binds to the sensor substrates. The first type sensor substrate (S1) functional with glutaraldehyde, is easy to connect with amino-modified DNA fragments; the second type sensor substrate (S2) is readily connected sulfhydryl -modified DNA by adsorbed Au NPs.</p> Full article ">Figure 10
<p>The fluorescence spectra of RRA-DNA detected on PSi biosensor. The concentration of RRA-DNA is 1 µM. There are two spectra from two different samples for one kind of sensor. One of the spectra is the fluorescence spectrum for the detection of complementary probe DNA (the solid square), and the other one is the spectrum for the detection of non complementary probe DNA (the solid regular triangle).</p> Full article ">Figure 11
<p>The fluorescence spectra of RRA-DNA with varying concentrations.</p> Full article ">Figure 12
<p>The fitting graph of the RRA-DNA concentration and fluorescence intensity.</p> Full article ">
10 pages, 4465 KiB  
Article
Enhanced Humidity Sensitivity with Silicon Nanopillar Array by UV Light
by Wei Li, Chao Ding, Yun Cai, Juyan Liu, Linlin Wang, Qingying Ren and Jie Xu
Sensors 2018, 18(2), 660; https://doi.org/10.3390/s18020660 - 23 Feb 2018
Cited by 8 | Viewed by 3984
Abstract
The sensitivity of silicon nanopillar array for relative humidity (RH) with UV illumination was investigated in this work. The silicon nanopillar array was prepared by nanosphere lithography. Electrical measurements were performed on its sensing performance with and without UV irradiation. It was found [...] Read more.
The sensitivity of silicon nanopillar array for relative humidity (RH) with UV illumination was investigated in this work. The silicon nanopillar array was prepared by nanosphere lithography. Electrical measurements were performed on its sensing performance with and without UV irradiation. It was found that UV light improved the humidity sensitivity with different UV light wavelengths and power. The sensor response and recovery time were reduced. Furthermore, the turn-on threshold voltage and the operating voltage both decreased. These sensing characteristics can mainly be attributed to the electron-hole pairs generated by UV light. These electron-hole pairs promote the adsorption and desorption processes. The results indicate that silicon nanopillar array materials with UV irradiation might be competitive as novel sensing materials for fabricating humidity sensors with high performances. Full article
(This article belongs to the Section Biosensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) The schematic diagram of the designed sensors. (<b>b</b>) A photo of the actual sample.</p> Full article ">Figure 2
<p>Measurement system.</p> Full article ">Figure 3
<p>The morphology of the polystyrene (PS) monolayer and Si nanopillar array. (<b>a</b>) SEM image of the PS monolayer; (<b>b</b>) SEM image of the ordered Si nanopillar array; (<b>c</b>) and an atomic force microscope (AFM) image of the ordered Si nanopillar array.</p> Full article ">Figure 4
<p>The current (I)-relative humidity (RH) curves at an applied voltage of 1 V with different UV wavelengths from 260 nm to 360 nm with the same power, 150 μW, and without UV irradiation.</p> Full article ">Figure 5
<p>The I-RH curves at an applied voltage of 1 V with different UV power from 120 μW to 200 μW in the same wavelength, 300 nm, and without UV irradiation.</p> Full article ">Figure 6
<p>I-RH curves at an applied voltage of 1 V with different distances between the UV LED and the sample under UV LED (λ = 300 nm, P =150 μW).</p> Full article ">Figure 7
<p>The reflectance spectrum for flat and Si nanopillar substrates.</p> Full article ">Figure 8
<p>I-V curves measured from 30% RH to 90% RH. (<b>a</b>) Without UV irradiation; (<b>b</b>) With UV irradiation.</p> Full article ">Figure 9
<p>The humidity sensitivity from 30% RH to 90% RH at applied voltages from 1 V (<b>a</b>), 2 V (<b>b</b>), 3 V (<b>c</b>) and 4 V (<b>d</b>), with and without UV irradiation.</p> Full article ">Figure 10
<p>The response and recovery time at an applied voltage of 1 V, with and without UV irradiation: (<b>a</b>) 30%, (<b>b</b>) 50%, (<b>c</b>) 70%, and (<b>d</b>) 90%.</p> Full article ">Figure 11
<p>The humidity hysteresis measured for the sensors: (<b>a</b>) 1 V, (<b>b</b>) 2 V, (<b>c</b>) 3 V, and (<b>d</b>) 4 V.</p> Full article ">Figure 12
<p>The humidity sensor stability after the sensors were exposed to air for 24 weeks with UV and without UV.</p> Full article ">
17 pages, 2676 KiB  
Article
Application of Improved 5th-Cubature Kalman Filter in Initial Strapdown Inertial Navigation System Alignment for Large Misalignment Angles
by Wei Wang and Xiyuan Chen
Sensors 2018, 18(2), 659; https://doi.org/10.3390/s18020659 - 23 Feb 2018
Cited by 31 | Viewed by 3948
Abstract
In view of the fact the accuracy of the third-degree Cubature Kalman Filter (CKF) used for initial alignment under large misalignment angle conditions is insufficient, an improved fifth-degree CKF algorithm is proposed in this paper. In order to make full use of the [...] Read more.
In view of the fact the accuracy of the third-degree Cubature Kalman Filter (CKF) used for initial alignment under large misalignment angle conditions is insufficient, an improved fifth-degree CKF algorithm is proposed in this paper. In order to make full use of the innovation on filtering, the innovation covariance matrix is calculated recursively by an innovative sequence with an exponent fading factor. Then a new adaptive error covariance matrix scaling algorithm is proposed. The Singular Value Decomposition (SVD) method is used for improving the numerical stability of the fifth-degree CKF in this paper. In order to avoid the overshoot caused by excessive scaling of error covariance matrix during the convergence stage, the scaling scheme is terminated when the gradient of azimuth reaches the maximum. The experimental results show that the improved algorithm has better alignment accuracy with large misalignment angles than the traditional algorithm. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>The distribution of weights of innovations under different <math display="inline"> <semantics> <mi>ξ</mi> </semantics> </math>.</p> Full article ">Figure 2
<p>The flow chart of IICKF5.</p> Full article ">Figure 3
<p>The estimated misalignment angles (<math display="inline"> <semantics> <mrow> <mi>ϕ</mi> <mo stretchy="false">(</mo> <mn>0</mn> <mo stretchy="false">)</mo> <mo>=</mo> <mo stretchy="false">[</mo> <mn>20</mn> <mo>°</mo> <mo>,</mo> <mtext> </mtext> <mo>−</mo> <mn>40</mn> <mo>°</mo> <mo>,</mo> <mtext> </mtext> <mn>170</mn> <mo>°</mo> <mo stretchy="false">]</mo> </mrow> </semantics> </math>): (<b>a</b>) Pitching misalignment angle; (<b>b</b>) rolling misalignment angle; (<b>c</b>) azimuth misalignment angle.</p> Full article ">Figure 4
<p>The gradient values of azimuth angle error.</p> Full article ">Figure 5
<p>The estimated azimuth misalignment angles of CKF5, ICKF5 and IICKF5.</p> Full article ">Figure 6
<p>The field test setup.</p> Full article ">Figure 7
<p>The alignment errors of large azimuth misalignment angle of dynamic vehicle experiment: (<b>a</b>) Pitch errors; (<b>b</b>) roll errors; (<b>c</b>) heading errors.</p> Full article ">Figure 8
<p>The gradient values of azimuth angle error (dynamic vehicle experiment).</p> Full article ">Figure 9
<p>The alignment errors of different algorithms: (<b>a</b>) Pitch errors; (<b>b</b>) roll errors; (<b>c</b>) heading errors.</p> Full article ">
11 pages, 9407 KiB  
Article
Methionine-Capped Gold Nanoclusters as a Fluorescence-Enhanced Probe for Cadmium(II) Sensing
by Yan Peng, Maomao Wang, Xiaoxia Wu, Fu Wang and Lang Liu
Sensors 2018, 18(2), 658; https://doi.org/10.3390/s18020658 - 23 Feb 2018
Cited by 25 | Viewed by 6193
Abstract
Gold nanoclusters (Au NCs) have been considered as novel heavy metal ions sensors due to their ultrafine size, photo-stability and excellent fluorescent properties. In this study, a green and facile method was developed for the preparation of fluorescent water-soluble gold nanoclusters with methionine [...] Read more.
Gold nanoclusters (Au NCs) have been considered as novel heavy metal ions sensors due to their ultrafine size, photo-stability and excellent fluorescent properties. In this study, a green and facile method was developed for the preparation of fluorescent water-soluble gold nanoclusters with methionine as a stabilizer. The nanoclusters emit orange fluorescence with excitation/emission peaks at 420/565 nm and a quantum yield of about 1.46%. The fluorescence of the Au NCs is selectively and sensitively enhanced by addition of Cd(II) ions attributed to the Cd(II) ion-induced aggregation of nanoclusters. This finding was further used to design a fluorometric method for the determination of Cd(II) ions, which had a linear response in the concentration range from 50 nM to 35 μM and a detection limit of 12.25 nM. The practicality of the nanoprobe was validated in various environmental water samples and milk powder samples, with a fairly satisfactory recovery percent. Full article
(This article belongs to the Special Issue Fluorescent Probes and Sensors)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Schematic illustration of the Gold nanoclusters’ (Au NCs) formation and the Cd<sup>2+</sup> induced fluorescence enhancing of Au NCs.</p> Full article ">Figure 2
<p>UV–vis absorption spectra (<b>A</b>) and fluorescence excitation (black) and emission (red) spectra (<b>B</b>) of the as-synthesized Au NCs. The insets of B show photographs of the Au NC aqueous solution in room light (a) and UV light (b), and powder in room light (c) and UV light (d).</p> Full article ">Figure 3
<p>Representative high resolution transmission electron microscopy (HRTEM) images of the as-synthesized luminescent Au NCs (<b>A</b>); the particle-size distribution histogram of Au NCs (<b>B</b>); Au NCs in the presence of 50 µM Cd<sup>2+</sup> (<b>C</b>); and particle-size distribution histogram (<b>D</b>).</p> Full article ">Figure 4
<p>Fluorescence lifetimes of Au NCs in aqueous solution (<b>A</b>); data were collected at 565 nm with excitation at 375 nm. X-ray photoelectron spectroscopy (XPS) spectra of Au 4f of Au NCs (<b>B</b>).</p> Full article ">Figure 5
<p>The effect of pH (<b>A</b>), concentration of NaCl (<b>B</b>) and incubation time (<b>C</b>) on fluorescence intensity of Au NCs upon addition of Cd<sup>2+</sup> ions at different concentrations (35 μM (a); 10 μM (b); and 0 μM (c)), respectively.</p> Full article ">Figure 6
<p>(<b>A</b>) The fluorescence response of the Au NCs in the presence of various concentrations of Cd<sup>2+</sup> (0.0025, 0.005, 0.025, 0.05, 2.5, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 75, 100 µM); (<b>B</b>) Relationship between fluorescence intensity and Cd<sup>2+</sup> concentration. The inset picture shows the linear detection range for 0.05–35 µM of Cd<sup>2+</sup>.</p> Full article ">Figure 7
<p>(<b>A</b>) Selective experiments of Au NCs for other competitive metal ions and anions; (<b>B</b>) Photographs of Au NCs under UV light after being incubated with various ions.</p> Full article ">
34 pages, 6618 KiB  
Article
A Method for Dynamically Selecting the Best Frequency Hopping Technique in Industrial Wireless Sensor Network Applications
by Erlantz Fernández de Gorostiza, Jorge Berzosa, Jon Mabe and Roberto Cortiñas
Sensors 2018, 18(2), 657; https://doi.org/10.3390/s18020657 - 23 Feb 2018
Cited by 19 | Viewed by 6327
Abstract
Industrial wireless applications often share the communication channel with other wireless technologies and communication protocols. This coexistence produces interferences and transmission errors which require appropriate mechanisms to manage retransmissions. Nevertheless, these mechanisms increase the network latency and overhead due to the retransmissions. Thus, [...] Read more.
Industrial wireless applications often share the communication channel with other wireless technologies and communication protocols. This coexistence produces interferences and transmission errors which require appropriate mechanisms to manage retransmissions. Nevertheless, these mechanisms increase the network latency and overhead due to the retransmissions. Thus, the loss of data packets and the measures to handle them produce an undesirable drop in the QoS and hinder the overall robustness and energy efficiency of the network. Interference avoidance mechanisms, such as frequency hopping techniques, reduce the need for retransmissions due to interferences but they are often tailored to specific scenarios and are not easily adapted to other use cases. On the other hand, the total absence of interference avoidance mechanisms introduces a security risk because the communication channel may be intentionally attacked and interfered with to hinder or totally block it. In this paper we propose a method for supporting the design of communication solutions under dynamic channel interference conditions and we implement dynamic management policies for frequency hopping technique and channel selection at runtime. The method considers several standard frequency hopping techniques and quality metrics, and the quality and status of the available frequency channels to propose the best combined solution to minimize the side effects of interferences. A simulation tool has been developed and used in this work to validate the method. Full article
(This article belongs to the Collection Smart Industrial Wireless Sensor Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Example of a Frequency Hopping schedule.</p> Full article ">Figure 2
<p>Example of Received Signal Strength Indicator (RSSI) signal over 1 s for an IEEE 802.15.4 network with 16 channels. The RSSI values lie within −120 dBm and 0 dBm. Higher RSSI values indicate channels interfered to a greater extent, while lower RSSI values indicate channels interfered to a lesser extent.</p> Full article ">Figure 3
<p>Example of linear projection from the mean value of the RSSI signal (blue) to a normalized channel gain, <span class="html-italic">H</span> (red). The dashed red line represents the power metric, <span class="html-italic">Q</span>, derived from the channel gain.</p> Full article ">Figure 4
<p>Different statistical properties obtained from the RSSI values in <a href="#sensors-18-00657-f002" class="html-fig">Figure 2</a>: (<b>a</b>) Mean value; (<b>b</b>) Standard deviation; (<b>c</b>) Skewness; (<b>d</b>) 50% quantile; (<b>e</b>) SOTH with −70 dBm threshold.</p> Full article ">Figure 5
<p>Normalized channel gains from RSSI statistical properties of <a href="#sensors-18-00657-f004" class="html-fig">Figure 4</a>.</p> Full article ">Figure 6
<p>Classification of frequency hopping techniques.</p> Full article ">Figure 7
<p>Channel selection for Matched Frequency Hopping (MFH) technique. The red line represents the channel gain and the blue line represents the cumulative sum of the power metric. The selected channels are represented as vertical black lines. The best channels are selected from the intersections between the cumulative metric and the equally spaced horizontal lines.</p> Full article ">Figure 8
<p>Comparison of channel selection between different frequency hopping techniques: (<b>a</b>) Highest Gain Frequency Hopping (HGFH); (<b>b</b>) Matched Frequency Hopping (MFH); (<b>c</b>) Clipped Matched Frequency Hopping (CMFH) with <span class="html-italic">ξ</span> = 0.1; (<b>d</b>) Advanced Frequency Hopping (AFH) with <span class="html-italic">α</span> = 0.1. The selected channels are represented as vertical black lines. HGFH directly uses the channel gain (red) to select the best channels, while MFH, CMFH and AFH use the cumulative metrics (blue) obtained from the power metrics (green) and their intersections with the equally spaced horizontal lines.</p> Full article ">Figure 9
<p>Channel selection evolution in the Clipped Matched Frequency Hopping (CMFH) technique when varying the threshold parameter: (<b>a</b>) <span class="html-italic">ξ</span> = 0.1; (<b>b</b>) <span class="html-italic">ξ</span> = 0.3; (<b>c</b>) <span class="html-italic">ξ</span> = 0.5.</p> Full article ">Figure 10
<p>Channel selection evolution in the Advanced Frequency Hopping (AFH) technique when varying the α parameter: (<b>a</b>) <span class="html-italic">α</span> = 1; (<b>b</b>) <span class="html-italic">α</span> = 0.1; (<b>c</b>) <span class="html-italic">α</span> = 0.01.</p> Full article ">Figure 11
<p>Comparison of channel selection between different frequency hopping techniques: (<b>a</b>) Random Frequency Hopping (RFH); (<b>b</b>) Matched Frequency Hopping (MFH); (<b>c</b>) Weighted Random Frequency Hopping (WRFH). In the RFH and MFH cases, horizontal lines are equally spaced while in the WRFH case, horizontal lines are randomly spaced.</p> Full article ">Figure 12
<p>Comparison of channel usage between different frequency hopping techniques: (<b>a</b>) WRFH; (<b>b</b>) UBAFH with <math display="inline"> <semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>10</mn> </mrow> </semantics> </math>; (<b>c</b>) UBAFH with <math display="inline"> <semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>100</mn> </mrow> </semantics> </math>; (<b>d</b>) SAFH with <math display="inline"> <semantics> <mrow> <mi>c</mi> <mo>=</mo> <mn>10</mn> </mrow> </semantics> </math>; (<b>e</b>) SAFH with <math display="inline"> <semantics> <mrow> <mi>c</mi> <mo>=</mo> <mn>100</mn> </mrow> </semantics> </math>.</p> Full article ">Figure 13
<p>Representation of observation (blue) and operation (purple) times and example of channel occupancy with two static networks (WLAN and LR-WPAN) and a dynamic network (Bluetooth) that hops over three frequency channels. The coexisting networks will be analysed during the observation time to determine the operation of the network of interest during the operation time.</p> Full article ">Figure 14
<p>Signal (green) and noise (red) overlapping. There are different sources that contribute to noise: static networks and dynamic networks that hop over different frequencies. The noise is analysed during the observation time (blue) to determine the hopping sequence of the signal of interest. The signal of interest is only transmitted during the operation time (purple). The overlapping of signal and noise is represented in black. This overlap only implies time and frequency coexistence; for an error to occur, the SNR must be higher than the receiving sensitivity of the nodes too.</p> Full article ">Figure 15
<p>Different network topologies: (<b>a</b>) Single-hop: each node (black dots) communicates directly to the gateway (red dot); (<b>b</b>) Multi-hop: the nodes communicate to the gateway hopping through other nodes.</p> Full article ">Figure 16
<p>Main interface of the simulation tool. The left panels contain buttons to configure the analysis parameters and select the different plotting options. The right panel is reserved for the graphical representation of the results. The different type of results are represented in their corresponding tab within the graph panel.</p> Full article ">Figure 17
<p>Example of an 802.15.4 network deployment with four nodes (black dots) and a gateway (G). The gateway is surrounded by four interfering networks (coloured markers).</p> Full article ">Figure 18
<p>Interference noise at the gateway location: (<b>a</b>) Frequency pattern. The WLAN network transmits in the 6th channel of the 802.11 standard, centred at 2437 MHz with a bandwidth of 22 MHz (red). The LR-WPAN A network transmits in the 4th channel of the 802.15.4 standard, centred at 2420 MHz with a bandwidth of 3 MHz (yellow). The LR-WPAN B network transmits in the 13th channel, centred at 2465 MHz (green). The Bluetooth network transmits over seven frequency channels with a bandwidth of 1 MHz each, hopping from channel to channel in a random way (blue); (<b>b</b>) Signal strength. The signal strength of each interfering network at the gateway position is calculated according to the free-space path loss model. The four interfering networks are at the same distance from the gateway, but have different signal strengths as they have different transmission powers.</p> Full article ">Figure 19
<p>Channel selection. HGFH technique directly uses the channel gain (red) to select the 10 best channels. WRFH, AFH and SAFH techniques use the cumulative metric (blue) derived from the power metric (green) to select the best channels. In all the case, the selected channels are represented as vertical black lines.</p> Full article ">Figure 20
<p>Frequency pattern of signal and noise for different hopping techniques and quality metrics. The interfering noise is represented in red and the signal of interest in green. The same noise pattern is considered for all the hopping techniques and all the quality metrics. Different hopping techniques and different quality metrics result in different frequency patterns for the signal of interest. The overlapping of the noise and signal of interest is represented in black.</p> Full article ">Figure 21
<p>Strength of signal and noise for different hopping techniques and quality metrics. The interfering noise is represented in red and the signal of interest in green. The noise is the sum of all the interfering networks as received at the gateway. The signal of interest is the signal strength received at the gateway coming from the first node.</p> Full article ">Figure 22
<p>Packet Error Rate for all possible node connections for different quality metrics and hopping techniques. Each element contains the Packet Error Rate (PER) value (percentage) for the specified connection. <math display="inline"> <semantics> <mrow> <mn>1</mn> <mo>&gt;</mo> <mi>G</mi> </mrow> </semantics> </math>, for instance, indicates the connection from node 1 to the gateway.</p> Full article ">Figure 23
<p>Inferred topologies with <math display="inline"> <semantics> <mrow> <msub> <mi>w</mi> <mrow> <mi>P</mi> <mi>E</mi> <mi>R</mi> </mrow> </msub> <mo>=</mo> <mn>1</mn> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>w</mi> <mrow> <mi>D</mi> <mi>I</mi> <mi>S</mi> <mi>T</mi> </mrow> </msub> <mo>=</mo> <mn>0</mn> </mrow> </semantics> </math>. Minimizing the overall PER results in direct connections between nodes and gateway for all the hopping techniques and quality metrics.</p> Full article ">Figure 24
<p>Inferred topologies with <math display="inline"> <semantics> <mrow> <msub> <mi>w</mi> <mrow> <mi>P</mi> <mi>E</mi> <mi>R</mi> </mrow> </msub> <mo>=</mo> <mn>0</mn> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>w</mi> <mrow> <mi>D</mi> <mi>I</mi> <mi>S</mi> <mi>T</mi> </mrow> </msub> <mo>=</mo> <mn>1</mn> </mrow> </semantics> </math>. Minimizing the transmission distances results in multi-hop connections for all the hopping techniques and quality metrics. The overall PER is considerably increased, except for the HGFH technique, which remains the same as with single-hop topology.</p> Full article ">Figure 25
<p>Inferred topologies with <math display="inline"> <semantics> <mrow> <msub> <mi>w</mi> <mrow> <mi>P</mi> <mi>E</mi> <mi>R</mi> </mrow> </msub> <mo>=</mo> <mn>0.5</mn> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>w</mi> <mrow> <mi>D</mi> <mi>I</mi> <mi>S</mi> <mi>T</mi> </mrow> </msub> <mo>=</mo> <mn>0.5</mn> </mrow> </semantics> </math>. Different hopping techniques and different quality metrics result in different topologies.</p> Full article ">
20 pages, 9960 KiB  
Article
Parameter Estimation of the Thermal Network Model of a Machine Tool Spindle by Self-made Bluetooth Temperature Sensor Module
by Yuan-Chieh Lo, Yuh-Chung Hu and Pei-Zen Chang
Sensors 2018, 18(2), 656; https://doi.org/10.3390/s18020656 - 23 Feb 2018
Cited by 12 | Viewed by 5970
Abstract
Thermal characteristic analysis is essential for machine tool spindles because sudden failures may occur due to unexpected thermal issue. This article presents a lumped-parameter Thermal Network Model (TNM) and its parameter estimation scheme, including hardware and software, in order to characterize both the [...] Read more.
Thermal characteristic analysis is essential for machine tool spindles because sudden failures may occur due to unexpected thermal issue. This article presents a lumped-parameter Thermal Network Model (TNM) and its parameter estimation scheme, including hardware and software, in order to characterize both the steady-state and transient thermal behavior of machine tool spindles. For the hardware, the authors develop a Bluetooth Temperature Sensor Module (BTSM) which accompanying with three types of temperature-sensing probes (magnetic, screw, and probe). Its specification, through experimental test, achieves to the precision ±(0.1 + 0.0029|t|) °C, resolution 0.00489 °C, power consumption 7 mW, and size Ø40 mm × 27 mm. For the software, the heat transfer characteristics of the machine tool spindle correlative to rotating speed are derived based on the theory of heat transfer and empirical formula. The predictive TNM of spindles was developed by grey-box estimation and experimental results. Even under such complicated operating conditions as various speeds and different initial conditions, the experiments validate that the present modeling methodology provides a robust and reliable tool for the temperature prediction with normalized mean square error of 99.5% agreement, and the present approach is transferable to the other spindles with a similar structure. For realizing the edge computing in smart manufacturing, a reduced-order TNM is constructed by Model Order Reduction (MOR) technique and implemented into the real-time embedded system. Full article
(This article belongs to the Special Issue Selected Sensor Related Papers from ICI2017)
►▼ Show Figures

Figure 1

Figure 1
<p>System identification procedure.</p> Full article ">Figure 2
<p>(<b>a</b>) 4-node Thermal Network Model (TNM) at operating mode; (<b>b</b>) 4-node TNM at natural cooling mode.</p> Full article ">Figure 3
<p>(<b>a</b>) Circuit diagram of Bluetooth Temperature Sensor Module (BTSM); (<b>b</b>) Printed Circuit Board (PCB) layout of BTSM.</p> Full article ">Figure 4
<p>(<b>a</b>) Evolution of BTSM and three types of temperature probes; (<b>b</b>) Experimental setup.</p> Full article ">Figure 5
<p>Sensor location and TNM representation in machine tool spindle.</p> Full article ">Figure 6
<p>(<b>a</b>) Experimental result of 12 nodes on 6021 rpm; (<b>b</b>) The transition region of reaching thermal equilibrium.</p> Full article ">Figure 7
<p>(<b>a</b>) Simulated heat generation varying with rotational speeds; (<b>b</b>) Estimated convective resistance varying with rotational speeds.</p> Full article ">Figure 8
<p>(<b>a</b>) Self-validation between predicted and measured steady state temperature, <span class="html-italic">T</span><sub>1</sub> to <span class="html-italic">T</span><sub>4</sub> are the positions of the rear bearing A, midpoint of inner housing, front bearing D, and midpoint of shaft, respectively; (<b>b</b>) Steady-state temperature validation.</p> Full article ">Figure 9
<p>Model self-validation at operating mode with various rotation speeds (<b>a</b>) Rear bearing (<span class="html-italic">T</span><sub>1</sub>) temperature validation (<b>b</b>) Midpoint of housing (<span class="html-italic">T</span><sub>2</sub>) temperature validation (<b>c</b>) Front bearing (<span class="html-italic">T</span><sub>3</sub>) temperature validation (<b>d</b>) Midpoint of shaft (<span class="html-italic">T</span><sub>4</sub>) temperature validation.</p> Full article ">Figure 10
<p>Model self-validation at natural cooling mode with various rotation speeds (<b>a</b>) Rear bearing (<span class="html-italic">T</span><sub>1</sub>) temperature validation (<b>b</b>) Midpoint of housing (<span class="html-italic">T</span><sub>2</sub>) temperature validation (<b>c</b>) Front bearing (<span class="html-italic">T</span><sub>3</sub>) temperature validation (<b>d</b>) Midpoint of shaft (<span class="html-italic">T</span><sub>4</sub>) temperature validation.</p> Full article ">Figure 11
<p>External validation between measured temperature and predicted temperature of estimated TNM under stepwise rotational speed (3001, 5018, 7028 rpm).</p> Full article ">Figure 12
<p>(<b>a</b>) Hankel singular values; (<b>b</b>) Short circuit time constant method with C2–C4 are short circuited; (<b>c</b>) Pole location of estimated TNM when comparing with the Model Order Reduction (MOR) model, SCTC model and varying with rotational speed.</p> Full article ">Figure 13
<p>Bode diagram of estimated TNM and 1st-order truncated model based on MOR.</p> Full article ">Figure 14
<p>Schematic summary.</p> Full article ">Figure 15
<p>Demonstration of the thermo-feature identification system.</p> Full article ">
17 pages, 891 KiB  
Review
Microwave Sensors for Breast Cancer Detection
by Lulu Wang
Sensors 2018, 18(2), 655; https://doi.org/10.3390/s18020655 - 23 Feb 2018
Cited by 125 | Viewed by 12773
Abstract
Breast cancer is the leading cause of death among females, early diagnostic methods with suitable treatments improve the 5-year survival rates significantly. Microwave breast imaging has been reported as the most potential to become the alternative or additional tool to the current gold [...] Read more.
Breast cancer is the leading cause of death among females, early diagnostic methods with suitable treatments improve the 5-year survival rates significantly. Microwave breast imaging has been reported as the most potential to become the alternative or additional tool to the current gold standard X-ray mammography for detecting breast cancer. The microwave breast image quality is affected by the microwave sensor, sensor array, the number of sensors in the array and the size of the sensor. In fact, microwave sensor array and sensor play an important role in the microwave breast imaging system. Numerous microwave biosensors have been developed for biomedical applications, with particular focus on breast tumor detection. Compared to the conventional medical imaging and biosensor techniques, these microwave sensors not only enable better cancer detection and improve the image resolution, but also provide attractive features such as label-free detection. This paper aims to provide an overview of recent important achievements in microwave sensors for biomedical imaging applications, with particular focus on breast cancer detection. The electric properties of biological tissues at microwave spectrum, microwave imaging approaches, microwave biosensors, current challenges and future works are also discussed in the manuscript. Full article
(This article belongs to the Special Issue Label-Free Biosensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Microwave imaging measurement system developed by Li et al.; (<b>b</b>) Microwave sensor array configuration; (<b>c</b>) Schematic diagram of experiment. Reprinted with copyright permission from Li et al. [<a href="#B106-sensors-18-00655" class="html-bibr">106</a>].</p> Full article ">Figure 2
<p>(<b>a</b>) Conventional antenna geometry; (<b>b</b>) Modified bow-tie antenna; (<b>c</b>) Simulated and measured return loss; (<b>d</b>) Simulated and measured gain. Reprinted with copyright permission from Ting et al. [<a href="#B107-sensors-18-00655" class="html-bibr">107</a>].</p> Full article ">
13 pages, 2670 KiB  
Article
Using a Second Order Sigma-Delta Control to Improve the Performance of Metal-Oxide Gas Sensors
by Lukasz Kowalski, Joan Pons-Nin, Eric Navarrete, Eduard Llobet and Manuel Domínguez-Pumar
Sensors 2018, 18(2), 654; https://doi.org/10.3390/s18020654 - 23 Feb 2018
Cited by 7 | Viewed by 5777
Abstract
Controls of surface potential have been proposed to accelerate the time response of MOX gas sensors. These controls use temperature modulations and a feedback loop based on first-order sigma-delta modulators to keep constant the surface potential. Changes in the surrounding gases, therefore, must [...] Read more.
Controls of surface potential have been proposed to accelerate the time response of MOX gas sensors. These controls use temperature modulations and a feedback loop based on first-order sigma-delta modulators to keep constant the surface potential. Changes in the surrounding gases, therefore, must be compensated by average temperature produced by the control loop, which is the new output signal. The purpose of this paper is to present a second order sigma-delta control of the surface potential for gas sensors. With this new control strategy, it is possible to obtain a second order zero of the quantization noise in the output signal. This provides a less noisy control of the surface potential, while at the same time some undesired effects of first order modulators, such as the presence of plateaus, are avoided. Experiments proving these performance improvements are presented using a gas sensor made of tungsten oxide nanowires. Plateau avoidance and second order noise shaping is shown with ethanol measurements. Full article
(This article belongs to the Special Issue Gas Sensors based on Semiconducting Metal Oxides)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) First-order sigma-delta modulator topology to control the chemical resistance of the metal-oxide sensing layer. At each sampling time <span class="html-italic">T<sub>S</sub></span>, depending on whether the chemical resistance <span class="html-italic">R<sub>chem</sub></span>, measured at the reference temperature <span class="html-italic">T<sub>high</sub></span>, is below (or above) the desired value <span class="html-italic">R<sub>th</sub></span>, a BIT1 (or BIT0) temperature waveform is applied to the sensor during the next sampling period; (<b>b</b>) parameters of the BIT0 and BIT1 waveforms.</p> Full article ">Figure 2
<p>Block diagram (<b>Top</b>) and equivalent sampled circuit (<b>Bottom</b>) of the 2nd order sigma-delta topology designed to control the chemical resistance of the MOX sensing layer.</p> Full article ">Figure 3
<p>Artistic view inside of the hot wall reactor during the AACVD process. A nitrogen flow carries the aerosol droplets of solvent containing the organic precursor.</p> Full article ">Figure 4
<p>(<b>a</b>) XRD results obtained for typical tungsten oxide nanowire films. Tungsten oxide is single crystalline and belongs to the monoclinic phase; (<b>b</b>) low magnification micrograph showing the AACVD grown film on top of the electrode area of a sensor within the 4-element transducer (<b>Left</b>). Higher magnification micrograph showing the typical microstructure of the AACVD grown tungsten nanowire films (<b>Right</b>).</p> Full article ">Figure 5
<p>Description of the experimental setup.</p> Full article ">Figure 6
<p>Experimental results in which 1st and 2nd order controls are used to obtain a given sequence of target chemical resistances <span class="html-italic">R<sub>th</sub></span>. (<b>a</b>) Time evolution of the chemical resistance (<b>Left</b>) and of the average temperature provided by the 1st order loop (<b>Right</b>); <span class="html-italic">R<sub>th</sub></span> was set to 376, 385, 375, 381, 377, 382, and 380 kΩ in 15 min intervals; <span class="html-italic">T<sub>high</sub></span> = 280 °C, <span class="html-italic">T<sub>low</sub></span> = 200 °C, δ = 25%, and <span class="html-italic">T<sub>S</sub></span> = 1 s; (<b>b</b>) same results when 2nd order control was applied to set R<sub>chem</sub> to 341, 348, 335, 346, 337, 344, and 339 kΩ in 60 min intervals; <span class="html-italic">T<sub>high</sub></span> = 280 °C, <span class="html-italic">T<sub>low</sub></span> = 200 °C, α = 2 kΩ, δ = 20% and <span class="html-italic">T<sub>S</sub></span> = 2 s. In left plots, the grey lines are the raw signals at the sampling frequency, while the green one is the moving average obtained with 200 samples.</p> Full article ">Figure 7
<p>Experimental results in which 1st and 2nd order controls are used to obtain the same sequence of target chemical resistances <span class="html-italic">R<sub>th</sub></span> = 370, 382, 372, 358, 368, 388, 372, and 352 kΩ in 60 min intervals. In both cases, <span class="html-italic">T<sub>high</sub></span> = 280 °C, <span class="html-italic">T<sub>low</sub></span> = 200 °C, δ = 20%, and <span class="html-italic">T<sub>S</sub></span> = 2 s; α = 2 kΩ in the 2nd order case; (<b>a</b>) time evolution of the chemical resistance (<b>Left</b>) and of the average temperature provided by the 1st order loop (<b>Right</b>); (<b>b</b>) same results when 2nd order control was applied. In left plots, the grey lines are the raw signals at the sampling frequency, while the green one is the moving average obtained with 50 samples.</p> Full article ">Figure 8
<p>Experimental results in which 1st and 2nd order controls are applied to set <span class="html-italic">R<sub>chem</sub></span> to 340 kΩ for 3 h. In both cases, <span class="html-italic">T<sub>high</sub></span> = 290 °C, <span class="html-italic">T<sub>low</sub></span> = 200 °C, δ = 25%, and <span class="html-italic">T<sub>S</sub></span> = 0.5 s; α = 1.4 kΩ in the 2nd order case. (<b>a</b>) Chemical resistance of the sensing layer (<b>Top</b>) and averaged temperature provided by the 1st order loop (<b>Bottom</b>); (<b>b</b>) same results as provided by the 2nd order loop; (<b>c</b>) power spectrum densities after 16,384 samples of the bit streams, obtained with standard P-Welch MatLab estimation. In top (<b>a</b>,<b>b</b>) plots, the grey lines are the raw signals at the sampling frequency, while the green one is the moving average obtained with 200 samples.</p> Full article ">Figure 9
<p>Experimental results in which either a 1st or a 2nd order control loop is used to set <span class="html-italic">R<sub>chem</sub></span> to 300 kΩ while a step of ethanol concentration was applied to the sensor. <span class="html-italic">T<sub>high</sub></span> = 290 °C, <span class="html-italic">T<sub>low</sub></span> = 160 °C, δ = 25%, and T<sub>S</sub> = 1 s in both cases; α = 3 kΩ in the 2nd order case. (<b>a</b>) Evolution with time of the ethanol concentration (<b>Top</b>), the average temperature provided by the control loop (<b>Middle</b>), and the chemical resistance (<b>Bottom</b>) when 1st order control was used; (<b>b</b>) same results for 2nd order control. In bottom plots the grey lines are the raw signals at the sampling frequency, while the green one is the moving average obtained with 60 samples.</p> Full article ">Figure 10
<p>Experimental results in which either a 1st or a 2nd order control loop has been used to set <span class="html-italic">R<sub>chem</sub></span> to 155 kΩ when a sequence of small steps of ethanol of 25 ppb concentration was applied to the sensor in 10 min intervals. <span class="html-italic">T<sub>high</sub></span> = 290 °C, <span class="html-italic">T<sub>low</sub></span> = 160 °C, δ = 25%, and <span class="html-italic">T<sub>S</sub></span> = 1 s in both cases; α = 3 kΩ in the 2nd order case. (<b>a</b>) Evolution with time of the ethanol concentration (<b>Top</b>), the averaged bit stream provided by the control loop (<b>Middle</b>), and the chemical resistance (<b>Bottom</b>) when 1st order control was used; (<b>b</b>) same results when 2nd order control was used. In bottom plots the grey lines are the raw signals at the sampling frequency, while the green one is the moving average obtained with 200 samples.</p> Full article ">
15 pages, 4633 KiB  
Article
Adaptive Correlation Model for Visual Tracking Using Keypoints Matching and Deep Convolutional Feature
by Yuankun Li, Tingfa Xu, Honggao Deng, Guokai Shi and Jie Guo
Sensors 2018, 18(2), 653; https://doi.org/10.3390/s18020653 - 23 Feb 2018
Cited by 1 | Viewed by 4548
Abstract
Although correlation filter (CF)-based visual tracking algorithms have achieved appealing results, there are still some problems to be solved. When the target object goes through long-term occlusions or scale variation, the correlation model used in existing CF-based algorithms will inevitably learn some non-target [...] Read more.
Although correlation filter (CF)-based visual tracking algorithms have achieved appealing results, there are still some problems to be solved. When the target object goes through long-term occlusions or scale variation, the correlation model used in existing CF-based algorithms will inevitably learn some non-target information or partial-target information. In order to avoid model contamination and enhance the adaptability of model updating, we introduce the keypoints matching strategy and adjust the model learning rate dynamically according to the matching score. Moreover, the proposed approach extracts convolutional features from a deep convolutional neural network (DCNN) to accurately estimate the position and scale of the target. Experimental results demonstrate that the proposed tracker has achieved satisfactory performance in a wide range of challenging tracking scenarios. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Flow chart of the proposed framework. CF: correlation filter.</p> Full article ">Figure 2
<p>Visualization of input and outputs of different layers. From left to right are the input frame, feature maps of conv2-2, feature maps of conv3-4, feature maps of conv4-4, and feature maps of conv5-4. (<b>a</b>) Bird1; (<b>b</b>) MotorRolling; (<b>c</b>) David.</p> Full article ">Figure 3
<p>Precision and success plots over the OTB-2013 dataset. (<b>a</b>) Precision plot; (<b>b</b>) Success plots.</p> Full article ">Figure 4
<p>Precision and success plots over the OTB-2015 dataset. (<b>a</b>) Precision plot; (<b>b</b>) Success plots.</p> Full article ">Figure 5
<p>Success plots over six tracking challenges of (<b>a</b>) occlusion; (<b>b</b>) background clutter; (<b>c</b>) in-plane rotation; (<b>d</b>) out-of-plane; (<b>e</b>) deformation; and (<b>f</b>) motion blur.</p> Full article ">Figure 6
<p>Tracking results on sequences with attributes of occlusion and background cluster. From top to bottom, the name of the video is (<b>a</b>) Box; (<b>b</b>) Bird1; (<b>c</b>) Soccer; (<b>d</b>) Human3.</p> Full article ">Figure 7
<p>Tracking results on sequences with attributes of occlusion and background cluster. From top to bottom, the name of the video is (<b>a</b>) MotorRolling; (<b>b</b>) Couple; (<b>c</b>) Trellis; (<b>d</b>) Car4.</p> Full article ">Figure 8
<p>Tracking performance of different updating methods. (<b>a</b>) Success plots on OTB-2015; (<b>b</b>) Success plots over sequences with occlusion attribute.</p> Full article ">Figure 9
<p>Scale estimation performance using different features. (<b>a</b>) Success plots on OTB-2015; (<b>b</b>) Success plots over sequences with the attribute of scale variation.</p> Full article ">Figure 10
<p>Failure cases on the following sequences: (<b>a</b>) Biker; (<b>b</b>) Matrix; (<b>c</b>) Liquor; (<b>d</b>) Walking2.</p> Full article ">
31 pages, 14153 KiB  
Article
The Spectrum Analysis Solution (SAS) System: Theoretical Analysis, Hardware Design and Implementation
by Ram M. Narayanan, Richard K. Pooler, Anthony F. Martone, Kyle A. Gallagher and Kelly D. Sherbondy
Sensors 2018, 18(2), 652; https://doi.org/10.3390/s18020652 - 22 Feb 2018
Cited by 3 | Viewed by 4490
Abstract
This paper describes a multichannel super-heterodyne signal analyzer, called the Spectrum Analysis Solution (SAS), which performs multi-purpose spectrum sensing to support spectrally adaptive and cognitive radar applications. The SAS operates from ultrahigh frequency (UHF) to the S-band and features a wideband channel with [...] Read more.
This paper describes a multichannel super-heterodyne signal analyzer, called the Spectrum Analysis Solution (SAS), which performs multi-purpose spectrum sensing to support spectrally adaptive and cognitive radar applications. The SAS operates from ultrahigh frequency (UHF) to the S-band and features a wideband channel with eight narrowband channels. The wideband channel acts as a monitoring channel that can be used to tune the instantaneous band of the narrowband channels to areas of interest in the spectrum. The data collected from the SAS has been utilized to develop spectrum sensing algorithms for the budding field of spectrum sharing (SS) radar. Bandwidth (BW), average total power, percent occupancy (PO), signal-to-interference-plus-noise ratio (SINR), and power spectral entropy (PSE) have been examined as metrics for the characterization of the spectrum. These metrics are utilized to determine a contiguous optimal sub-band (OSB) for a SS radar transmission in a given spectrum for different modalities. Three OSB algorithms are presented and evaluated: the spectrum sensing multi objective (SS-MO), the spectrum sensing with brute force PSE (SS-BFE), and the spectrum sensing multi-objective with brute force PSE (SS-MO-BFE). Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Block diagram of a typical spectrum sensing (SS) radar.</p> Full article ">Figure 2
<p>Effect of Percent Occupancy (PO) threshold setting: (<b>a</b>). If the threshold is set too low, the likelihood of false detection will increase; (<b>b</b>) If the threshold is set too high, the likelihood of missed detection will increase.</p> Full article ">Figure 3
<p>A four-level representation of the binomial summation technique used in spectrum sensing multi objective (SS-MO) and spectrum sensing multi-objective with brute force PSE (SS-MO-BFE) for the brute force calculation of noise plus interference power in the spectrum. Ultimately, the number of levels would correspond to the number of samples in the spectrum.</p> Full article ">Figure 4
<p>(<b>a</b>) Assumed spectrum; (<b>b</b>) Corresponding SS-MO linear weighting function, wherein the green star indicates the optimal sub-band OSB value. The black box in <a href="#sensors-18-00652-f004" class="html-fig">Figure 4</a>a depicts the SS-MO solution.</p> Full article ">Figure 5
<p>(<b>a</b>) Assumed signal spectrum; (<b>b</b>) Corresponding grey-scale normalized brute force power spectral entropy (PSE) image.</p> Full article ">Figure 6
<p>(<b>a</b>) Assumed noise spectrum; (<b>b</b>) Corresponding grey-scale normalized brute force PSE image.</p> Full article ">Figure 7
<p>(<b>a</b>) Assumed signal spectrum; (<b>b</b>) Corresponding grey-scale normalized brute force PSE image showing corners detected by the Shi-Tomasi approach. The black box in <a href="#sensors-18-00652-f007" class="html-fig">Figure 7</a>a depicts the spectrum sensing with brute force PSE (SS-BFE) solution.</p> Full article ">Figure 8
<p>(<b>a</b>) Assumed spectrum; (<b>b</b>) Corresponding SS-MO-BFE linear weighting function, wherein the green star indicates the optimal sub-band (OSB) value. The black box in <a href="#sensors-18-00652-f008" class="html-fig">Figure 8</a>a depicts the SS-MO-BFE solution.</p> Full article ">Figure 9
<p>Block diagram of the Spectrum Analysis Solution (SAS) hardware. The block diagram is broken down into four subsections of the SAS which will be explained individually. The band pass filter blocks represent a series of filters.</p> Full article ">Figure 10
<p>Detailed block diagram of Block 1 in <a href="#sensors-18-00652-f009" class="html-fig">Figure 9</a>.</p> Full article ">Figure 11
<p>Gain of Block 1 (minus antenna) as a function of frequency.</p> Full article ">Figure 12
<p>Detailed block diagram of Block 2 in <a href="#sensors-18-00652-f009" class="html-fig">Figure 9</a>.</p> Full article ">Figure 13
<p>Gain of Block 2 as a function of frequency.</p> Full article ">Figure 14
<p>Detailed block diagram of Block 3 in <a href="#sensors-18-00652-f009" class="html-fig">Figure 9</a>.</p> Full article ">Figure 15
<p>Gain of Block 3 as a function of frequency.</p> Full article ">Figure 16
<p>Detailed block diagram of Block 4 in <a href="#sensors-18-00652-f009" class="html-fig">Figure 9</a>.</p> Full article ">Figure 17
<p>Gain of Block 4 prior to the 7120 downconverters as a function of frequency. (<b>a</b>) Channels 1 and 2; (<b>b</b>) Channels 3 and 4.</p> Full article ">Figure 18
<p>The SAS LabVIEW GUI front panel.</p> Full article ">Figure 19
<p>Gain versus frequency of the Schwarzbeck BBHA 9120 E double ridged broadband horn antenna.</p> Full article ">Figure 20
<p>(<b>a</b>) SAS system. (<b>b</b>) Oscilloscope and real-time spectrum analyzer (RSA).</p> Full article ">Figure 21
<p>Scatter plot of the total average power versus percent occupancy of the spectra from the ambient data collection.</p> Full article ">Figure 22
<p>Range of total average power and PO calculated for each center frequencies (CF) of the ambient data collection from Equations (31) and (32).</p> Full article ">Figure 23
<p>Scatter plot of the total average power versus percent occupancy of the pseudo-randomly generated spectra data collected by the SAS.</p> Full article ">Figure 24
<p>Plotted in blue is an example spectra from the closed loop data. Any frequency bin with an amplitude above −127 dBm on the red line corresponds to an emitter.</p> Full article ">Figure 25
<p>Average percent interference (PI) values for each SS technique for each pseudo-random waveform.</p> Full article ">Figure 26
<p>Average bandwidth (BW) values for each SS technique for each pseudo-random waveform.</p> Full article ">Figure 27
<p>Average signal-to-interference-plus-noise ratio (SINR) values for each SS technique for each pseudo-random waveform.</p> Full article ">
17 pages, 1905 KiB  
Article
A Cross-Layer, Anomaly-Based IDS for WSN and MANET
by Amar Amouri, Salvatore D. Morgera, Mohamed A. Bencherif and Raju Manthena
Sensors 2018, 18(2), 651; https://doi.org/10.3390/s18020651 - 22 Feb 2018
Cited by 30 | Viewed by 4327
Abstract
Intrusion detection system (IDS) design for mobile adhoc networks (MANET) is a crucial component for maintaining the integrity of the network. The need for rapid deployment of IDS capability with minimal data availability for training and testing is an important requirement of such [...] Read more.
Intrusion detection system (IDS) design for mobile adhoc networks (MANET) is a crucial component for maintaining the integrity of the network. The need for rapid deployment of IDS capability with minimal data availability for training and testing is an important requirement of such systems, especially for MANETs deployed in highly dynamic scenarios, such as battlefields. This work proposes a two-level detection scheme for detecting malicious nodes in MANETs. The first level deploys dedicated sniffers working in promiscuous mode. Each sniffer utilizes a decision-tree-based classifier that generates quantities which we refer to as correctly classified instances (CCIs) every reporting time. In the second level, the CCIs are sent to an algorithmically run supernode that calculates quantities, which we refer to as the accumulated measure of fluctuation (AMoF) of the received CCIs for each node under test (NUT). A key concept that is used in this work is that the variability of the smaller size population which represents the number of malicious nodes in the network is greater than the variance of the larger size population which represents the number of normal nodes in the network. A linear regression process is then performed in parallel with the calculation of the AMoF for fitting purposes and to set a proper threshold based on the slope of the fitted lines. As a result, the malicious nodes are efficiently and effectively separated from the normal nodes. The proposed scheme is tested for various node velocities and power levels and shows promising detection performance even at low-power levels. The results presented also apply to wireless sensor networks (WSN) and represent a novel IDS scheme for such networks. Full article
(This article belongs to the Section Sensor Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Simplified architecture of the proposed intrusion detection system (IDS).</p> Full article ">Figure 2
<p>Data fitting for the CCIs of sensor 5: (<b>a</b>) Data fitting for three distributions, extreme value (EV), Gamma, and Nakagami in scenario NS1P3 for node under test (NUT) 13; (<b>b</b>) Data fitting for three distributions, EV, Gamma, and Nakagami in scenario NS1P3 for NUT 19.</p> Full article ">Figure 3
<p>Data fitting for the CCIs of sensor 5: (<b>a</b>) Data fitting for three distributions, EV, Gamma, and Nakagami in scenario NS1P3 for NUT 13; (<b>b</b>) Data fitting for three distributions, EV, Gamma, and Nakagami in scenario NS15P7 for NUT 19.</p> Full article ">Figure 4
<p>Data fitting for the CCIs of sensor 5: (<b>a</b>) Data fitting for three distributions, EV, Gamma, and Nakagami in scenario NS5P7 for NUT 13; (<b>b</b>) Data fitting for three distributions, EV, Gamma, and Nakagami in scenario NS5P7 for NUT 19.</p> Full article ">Figure 5
<p>A two stage cross layer IDS.</p> Full article ">Figure 6
<p>The AMoF for different nodes and the fitted slope for those nodes for scenario NS1P7 (<span class="html-italic">Tr</span> = 50 s, <span class="html-italic">Ts</span> = 5 s): (<b>a</b>) The AMoF for different TNs; (<b>b</b>) The fitted slope, and its confidence for different NUTs.</p> Full article ">Figure 7
<p>The AMoF for different nodes and the fitted slope for those nodes for scenario NS1P7 (<span class="html-italic">Tr</span> = 100 s, <span class="html-italic">Ts</span> = 10 s): (<b>a</b>) The AMoF for different NUTs; (<b>b</b>) The fitted slope, and its confidence for different NUTs.</p> Full article ">Figure 8
<p>The AMoF for different nodes and the fitted slope for those nodes for scenario NS15P3 (<span class="html-italic">Tr</span> = 100 s, <span class="html-italic">Ts</span> = 5 s): (<b>a</b>) The AMoF for different NUTs; (<b>b</b>) The fitted slope, and its confidence for different NUTs.</p> Full article ">
12 pages, 3633 KiB  
Article
Optical Method for Estimating the Chlorophyll Contents in Plant Leaves
by Madaín Pérez-Patricio, Jorge Luis Camas-Anzueto, Avisaí Sanchez-Alegría, Abiel Aguilar-González, Federico Gutiérrez-Miceli, Elías Escobar-Gómez, Yvon Voisin, Carlos Rios-Rojas and Ruben Grajales-Coutiño
Sensors 2018, 18(2), 650; https://doi.org/10.3390/s18020650 - 22 Feb 2018
Cited by 81 | Viewed by 10849
Abstract
This work introduces a new vision-based approach for estimating chlorophyll contents in a plant leaf using reflectance and transmittance as base parameters. Images of the top and underside of the leaf are captured. To estimate the base parameters (reflectance/transmittance), a novel optical arrangement [...] Read more.
This work introduces a new vision-based approach for estimating chlorophyll contents in a plant leaf using reflectance and transmittance as base parameters. Images of the top and underside of the leaf are captured. To estimate the base parameters (reflectance/transmittance), a novel optical arrangement is proposed. The chlorophyll content is then estimated by using linear regression where the inputs are the reflectance and transmittance of the leaf. Performance of the proposed method for chlorophyll content estimation was compared with a spectrophotometer and a Soil Plant Analysis Development (SPAD) meter. Chlorophyll content estimation was realized for Lactuca sativa L., Azadirachta indica, Canavalia ensiforme, and Lycopersicon esculentum. Experimental results showed that—in terms of accuracy and processing speed—the proposed algorithm outperformed many of the previous vision-based approach methods that have used SPAD as a reference device. On the other hand, the accuracy reached is 91% for crops such as Azadirachta indica, where the chlorophyll value was obtained using the spectrophotometer. Additionally, it was possible to achieve an estimation of the chlorophyll content in the leaf every 200 ms with a low-cost camera and a simple optical arrangement. This non-destructive method increased accuracy in the chlorophyll content estimation by using an optical arrangement that yielded both the reflectance and transmittance information, while the required hardware is cheap. Full article
(This article belongs to the Special Issue Optical Biochemical Sensor Systems and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Optical system for the image acquisition: (<b>a</b>) Schematic view, (<b>b</b>) optical system test.</p> Full article ">Figure 2
<p>Image of the adaxial and abaxial leaf side in Bayer format.</p> Full article ">Figure 3
<p>Image in RGB format.</p> Full article ">Figure 4
<p>Leaf and background, separation process.</p> Full article ">Figure 5
<p>Reflectance according to the chlorophyll content in Soil Plant Analysis Development (SPAD) values.</p> Full article ">Figure 6
<p>Transmittance according to the chlorophyll content in SPAD values.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-362
Previous Issue
Next Issue
Back to TopTop