Showing papers in "IEEE Sensors Journal in 2018"

Latern: Dynamic Continuous Hand Gesture Recognition Using FMCW Radar Sensor

Abstract: Recently, hand gesture recognition systems have become increasingly interesting to researchers in the field of human–computer interfaces. Real-world systems for human dynamic hand gesture recognition is challenging as: 1) the system must be robust to various conditions; 2) there is a rich diversity in how people perform hand gestures, making hand gesture recognition difficult; and 3) the system must detect and recognize hand gestures continuously using unsegmented input streams in order to avoid noticeable lag between performing a gesture and its classification. In this paper, to address these challenges, we present Latern, a novel system for dynamic continuous hand gesture recognition based on a frequency-modulated continuous wave radar sensor. The radar system does not depend on lighting, noise, or atmospheric conditions. We employ a recurrent 3-D convolutional neural network to perform the classification of dynamic hand gestures. To enhance the processing performance, a connectionist temporal classification algorithm is used to train the network to predict class labels from inprogress gestures in unsegmented input streams. The experimental results show that Latern is able to achieve high recognition rates of 96%, which is higher than state-of-the-art hand gesture recognition systems. In addition, the conclusion in this paper can be used for a real-time hand gesture recognition system design.

Deep-Learning-Based Earth Fault Detection Using Continuous Wavelet Transform and Convolutional Neural Network in Resonant Grounding Distribution Systems

Abstract: Feature extraction for fault signals is critical and difficult in all kinds of fault detection schemes. A novel simple and effective method of faulty feeder detection in resonant grounding distribution systems based on the continuous wavelet transform (CWT) and convolutional neural network (CNN) is presented in this paper. The time-frequency gray scale images are acquired by applying the CWT to the collected transient zero-sequence current signals of the faulty feeder and sound feeders. The features of the gray scale image will be extracted adaptively by the CNN, which is trained by a large number of gray scale images under various kinds of fault conditions and factors. The features extraction and the faulty feeder detection can be implemented by the trained CNN simultaneously. As a comparison, two faulty feeder detection methods based on artificial feature extraction and traditional machine learning are introduced. A practical resonant grounding distribution system is simulated in power systems computer aided design/electromagnetic transients including DC, the effectiveness and performance of the proposed faulty feeder detection method is compared and verified under different fault circumstances.

A Novel Approach for Spectroscopic Chemical Identification Using Photonic Crystal Fiber in the Terahertz Regime

Abstract: A novel highly sensitive porous core-photonic crystal fiber (PC-PCF) has been designed and analyzed for detection of chemical analytes in the terahertz frequency range. The PC-PCF is designed using rectangular structured air holes in the core with a kagome structured cladding. The full vectorial finite-element method is used to tune the geometrical parameters and to characterize the fiber. Our results demonstrate a high relative chemical sensitivity with significantly lower confinement loss for different analytes. Moreover, the PCF shows near zero dispersion variation, high modal effective area, high birefringence, and high numerical aperture. The practical realization of the fiber is feasible with present fabrication techniques. Our optimized PCF has commercial applications in chemical sensing as well as applications in terahertz systems that require guided polarization preserving transmission.

Differential Sensors Using Microstrip Lines Loaded With Two Split-Ring Resonators

Abstract: This paper presents a microwave sensor using a pair of split-ring resonators (SRRs). The sensor is designed by loading a microstrip transmission line by two identical SRRs on its sides. Differential permittivity sensing is performed by loading the SRRs with dielectric samples. One transmission notch is observed for the identical loads, whereas the non-identical samples produce two split notches. The sensor’s operating principle is described through a circuit model analysis. A prototype of the designed sensor is fabricated and experimentally validated for verifying the differential sensing concept. The developed device can be used to compare or differentially characterize solid dielectric samples improving the robustness to environmental factors producing cross-sensitivity or miscalibration.

Spiral Photonic Crystal Fiber-Based Dual-Polarized Surface Plasmon Resonance Biosensor

Abstract: We numerically demonstrate a surface plasmon resonance biosensor-based on dual-polarized spiral photonic crystal fiber (PCF). Chemically stable gold material is used as the active plasmonic material, which is placed on the outer layer of the PCF to facilitate practical fabrication. Finite-element method-based numerical investigations show that the proposed biosensor shows maximum wavelength sensitivity of 4600 and 4300 nm/RIU in ${x}$ - and ${y}$ -polarized modes at an analyte refractive index of 1.37. Moreover, for analyte refractive index ranging from 1.33 to 1.38, maximum amplitude sensitivities of 371.5 RIU−1 and 420.4 RIU−1 are obtained in ${x}$ - and ${y}$ -polarized modes, respectively. In addition, the effects of changing pitch, different air hole diameter of the PCF and thickness of the gold layer on the sensing performance are also investigated. Owing to high sensitivity, improved sensing resolution and appropriate linearity characteristics, the proposed dual-polarized spiral PCF can be implemented for the detection of biological analytes, organic chemicals, biomolecules, and other unknown analytes.

Fault Detection in Wireless Sensor Networks Through SVM Classifier

Abstract: Wireless sensor networks (WSNs) are prone to many failures such as hardware failures, software failures, and communication failures. The fault detection in WSNs is a challenging problem due to sensor resources limitation and the variety of deployment field. Furthermore, the detection has to be precise to avoid negative alerts, and rapid to limit loss. The use of machine learning seems to be one of the most convenient solutions for detecting failure in WSNs. In this paper, support vector machines (SVMs) classification method is used for this purpose. Based on statistical learning theory, SVM is used in our context to define a decision function. As a light process in term of required resources, this decision function can be easily executed at cluster heads to detect anomalous sensor. The effectiveness of SVM for fault detection in WSNs is shown through an experimental study, comparing it to latest for the same application.

Tactile Sensors for Friction Estimation and Incipient Slip Detection—Toward Dexterous Robotic Manipulation: A Review

Abstract: Humans can handle and manipulate objects with ease; however, human dexterity has yet to be matched by artificial systems. Receptors in our fingers and hands provide essential tactile information to the motor control system during dexterous manipulation such that the grip force is scaled to the tangential forces according to the coefficient of friction. Likewise, tactile sensing will become essential for robotic and prosthetic gripping performance as applications move toward unstructured environments. However, most existing research ignores the need to sense the frictional properties of the sensor–object interface, which (along with contact forces and torques) is essential for finding the minimum grip force required to securely grasp an object. Here, we review this problem by surveying the field of tactile sensing from the perspective that sensors should: 1) detect gross slip (to adjust the grip force); 2) detect incipient slip (dependent on the frictional properties of the sensor–object interface and the geometries and mechanics of the sensor and the object) as an indication of grip security; or 3) measure friction on contact with an object and/or following a gross or incipient slip event while manipulating an object. Recommendations are made to help focus future sensor design efforts toward a generalizable and practical solution to sense, and hence control grip security. Specifically, we propose that the sensor mechanics should encourage incipient slip, by allowing parts of the sensor to slip while other parts remain stuck, and that instrumentation should measure displacement and deformation to complement conventional force, pressure, and vibration tactile sensing.

Photonic Crystal Fiber-Based Refractive Index Sensor for Early Detection of Cancer

Abstract: This paper proposes a novel cancer sensor based on dual core photonic crystal fiber for the detection of cancer cells in cervical, breast, and basal parts. The samples are taken in fluid form and infiltrated into the farmed cavity using selective infiltration method. Each fluid form has its own refractive index values which give the various responses in the transmission and loss spectrum. The spectral shift is obtained by inducing the coupling mechanism between silica core and cancer cell core for its launching input optical field which is investigated by finite element method. The proposed structure is also optimized with its structural dimensional property for enhancing the sensitivity. The sensing performances for the cervical cancer cell are obtained as high as 7916 nm/RIU for $x$ -polarization and 10625 nm/RIU for $y$ -polarization with the detection limit of 0.024. The sensitivity to breast cancer cells for $x$ - and $y$ -polarization is 5714.28 and 7857.14 nm/RIU, respectively, with detection limit of 0.014. Similarly, the sensitivity to basal cells can also reach 4500 nm/RIU for $x$ -polarization and 6000 nm/RIU for $y$ -polarization. To the best of our knowledge, such sensitivities are the highest reported thus so far.

Human Activity Classification in Smartphones Using Accelerometer and Gyroscope Sensors

Abstract: Activity classification in smartphones helps us to monitor and analyze the physical activities of the user in daily life and has potential applications in healthcare systems. This paper proposes a descriptor-based approach for activity classification using built-in sensors of smartphones. Accelerometer and gyroscope sensor signals are acquired to identify the activities performed by the user. In addition, time and frequency domain signals are derived using the collected signals. In the proposed approach, two descriptors, namely, histogram of gradient and centroid signature-based Fourier descriptor, are employed to extract feature sets from these signals. Feature and score level fusion are explored for information fusion. For classification, we have studied the performance of multiclass support vector machine and $k$ -nearest neighbor classifiers. The proposed approach is evaluated on two publicly available data sets, namely, UCI HAR data set and physical activity sensor data. Our experimental results show that the feature level fusion provides better performance than the score level fusion. In addition, our approach provides considerable improvement in classifying different activities as compared with the existing works. The average activity classification accuracy achieved using the proposed method is 97.12% as against the existing work, which provided 96.33% on UCI HAR data set. On the second data set, the proposed approach attained 96.83% classification accuracy, whereas the existing work achieved 90.2%.

Smart Wearable Hand Device for Sign Language Interpretation System With Sensors Fusion

Abstract: Gesturing is an instinctive way of communicating to present a specific meaning or intent. Therefore, research into sign language interpretation using gestures has been explored progressively during recent decades to serve as an auxiliary tool for deaf and mute people to blend into society without barriers. In this paper, a smart sign language interpretation system using a wearable hand device is proposed to meet this purpose. This wearable system utilizes five flex-sensors, two pressure sensors, and a three-axis inertial motion sensor to distinguish the characters in the American sign language alphabet. The entire system mainly consists of three modules: 1) a wearable device with a sensor module; 2) a processing module; and 3) a display unit mobile application module. Sensor data are collected and analyzed using a built-in embedded support vector machine classifier. Subsequently, the recognized alphabet is further transmitted to a mobile device through Bluetooth low energy wireless communication. An Android-based mobile application was developed with a text-to-speech function that converts the received textinto audible voice output. Experiment results indicate that a true sign language recognition accuracy rate of 65.7% can be achieved on average in the first version without pressure sensors. A second version of the proposed wearable system with the fusion of pressure sensors on the middle finger increased the recognition accuracy rate dramatically to 98.2%. The proposed wearable system outperforms the existing method, for instance, although background lights, and other factors are crucial to a vision-based processing method, they are not for the proposed system.

EAPC: Energy-Aware Path Construction for Data Collection Using Mobile Sink in Wireless Sensor Networks

Abstract: Data collection is one of the paramount concerns in wireless sensor networks. Many data collection algorithms have been proposed for collecting data in particular monitoring regions. However, the efficiency of the paths for such data collection can be improved. This paper proposes an energy-aware path construction (EAPC) algorithm, which selects an appropriate set of data collection points, constructs a data collection path, and collects data from the points burdened with data. EAPC is intended to prolong the network lifetime, it accounts for the path cost from its current data collection point to the next point and the forwarding load of each sensor node. Performance evaluation reveals that the proposed EAPC has more efficient performance than existing data collection mechanisms in terms of network lifetime, energy consumption, fairness index, and efficiency index.

High Sensitivity and Tunable Nanoscale Sensor Based on Plasmon-Induced Transparency in Plasmonic Metasurface

Abstract: A nanoscale sensor which comprises dielectric-metal-dielectric waveguide and plasmonic metasurface resonators is proposed to realize plasmon-induced transparency responses. The properties of the device are numerically and analytically investigated with different physical parameters. The effect of the incident polarization, geometrical parameters, plasmonic metasurface materials, prism dielectric constant, and the shape of metasurface in near infrared region are then studied to enhance the performance parameters of the structure including the sensitivity, figure-of-merit, and tunability. Our calculations show that the proposed devices are able to operate as a high-sensitivity and tunable sensor with maximum figure-of-merit of 480, and sensitivity of 497.8 nm/refractive index unit for slight change of $\Delta n = 0.24$ , in the refractive index, which stems from its ultra-narrow transparency window and long coupling length between resonators. Furthermore, the structure can be utilized for harnessing light propagation. It is shown that by using silver plasmonic metasurface and appropriate dielectric, a slow up factor as high as 800 is achievable in the proposed structure. We believe that the proposed sensor can be used as a promising platform for future sensing applications such as nanostructure biosensors.

Microwave Sensor for Liquid Dielectric Characterization Based on Metamaterial Complementary Split Ring Resonator

Abstract: A metamaterial-based microwave sensor with complementary split ring resonator (CSRR) is implemented for dielectric characterization of liquids. The novelty in the proposed contactless sensor is the use of liquid samples placed normal to the sensor surface. Placed inside of glass capillary tubes, it is quickly possible to analyze the dielectric properties of liquids simply by replacing the capillary tubes with new samples. The liquid samples inside the glass capillary tubes modify the resonant frequency and $Q$ -factor of the CSRR sensor. Thereby, a relation between the sensor resonant frequency, $Q$ -factor, and complex permittivity of the liquid samples can be estimated. A measurement setup was used to test the proposed sensor that exhibited successfully detection of 10% steps in binary mixtures of ethanol and water. The proposed sensor is compact, low cost, contactless, reusable, easily fabricated, and easy to work.

An IoT Cloud System for Traffic Monitoring and Vehicular Accidents Prevention Based on Mobile Sensor Data Processing

Abstract: The sudden traffic slowdown especially in fast scrolling roads and highways characterized by a scarce visibility is one of the major causes of accidents among motorized vehicles. It can be caused by other accidents, work-in-progress on roads, excessive motorized vehicles especially at peak times and so on. Typically, fixed traffic sensors installed on roads that interact with drivers’ mobile App through the 4G network can mitigate such a problem, but unfortunately not all roads and highways are equipped with such devices. In this paper, we discuss a possible alternative solution for addressing such an issue considering mobile traffic sensors directly installed in private and/or public transportation and volunteer vehicles. In this scenario a fast real-time processing of big traffic data is fundamental to prevent accidents. In particular, we discuss an IoT Cloud system for traffic monitoring and alert notification based on OpenGTS and MongoDB. Our IoT Cloud system, besides for private drivers, it is very useful for drivers of critical rescue vehicles such as ambulances. Experiments prove that our system provides acceptable response times that allows drivers to receive alert messages in useful time so as to avoid the risk of possible accidents.

Phonocardiographic Sensing Using Deep Learning for Abnormal Heartbeat Detection

Abstract: Deep learning-based cardiac auscultation is of significant interest to the healthcare community as it can help reducing the burden of manual auscultation with automated detection of abnormal heartbeats. However, the problem of automatic cardiac auscultation is complicated due to the requirement of reliable and highly accurate systems, which are robust to the background noise in the heartbeat sound. In this paper, we propose a Recurrent Neural Networks (RNNs)-based automated cardiac auscultation solution. Our choice of RNNs is motivated by their great success of modeling sequential or temporal data even in the presence of noise. We explore the use of various RNN models, and demonstrate that these models significantly outperform the best reported results in the literature. We also present the run-time complexity of various RNNs, which provides insight about their complexity versus performance trade-offs.

Sensor Applications and Physiological Features in Drivers’ Drowsiness Detection: A Review

Abstract: Drowsiness in drivers has become a serious cause of concern due to the occurrences of a large number of fatalities on the road each year. Lives of pedestrians and passengers are put to risk as drivers tend to fall asleep at the steering wheel. In the recent past, many researchers have paid attention to the problem of drowsiness detection since safe roads and safe driving are of paramount concern to all societies. This paper has led to the development of several novel and effective methods in detecting drivers’ drowsiness. These include: 1) Vehicle based methods; 2) Behavioral methods; and 3) Physiological methods. Since wake-sleep is an intermediate state between two physiologically dissimilar states, physiological signals can define this transition more accurately when compared with approaches that fall in other categories. This paper focuses on the role of physiological signals in detecting driver’s drowsiness level. The proposed methods measure the physiological signals by means of various sensors, which monitor the driver’s physiological parameters on a continual basis. Multiple sensors can be embedded on the driver or in the vicinity of the driver to capture vital signs indicating the onset of drowsiness. The aim here is to provide an insightful review of all such key approaches that fall in this category. This paper conducts a detailed study in which key physiological parameters that relate to drowsiness are identified, described, and analyzed. Furthermore, the overall advantages and limitations of these physiological based schemes are also highlighted.

Terahertz Sensing in a Hollow Core Photonic Crystal Fiber

Abstract: A terahertz sensor based on a hollow core photonic crystal fiber has been proposed in this paper for chemical analyte detection in the terahertz frequency range. The Zeonex-based asymmetrical hollow core is filled with an analyte and surrounded by a number of asymmetrical rectangular air holes bounded by a perfectly matched layer with absorbing boundary conditions. The performance of the proposed sensor is numerically investigated by using finite element method-based COMSOL software. It is found that a hollow core provides a high relative sensitivity as well as low transmission loss. Moreover, simplicity in design facilitates manufacturability, making it practical for a number of different biological and industrial applications.

A Movement Decomposition and Machine Learning-Based Fall Detection System Using Wrist Wearable Device

Abstract: Falls in the elderly is a world health problem. Although many fall detection solutions were presented in literature, few of them are wrist-wearable devices, mainly due to typical processing and classification challenges to achieve accuracy greater than 95%. Considering the wrist as a more comfortable, discrete and acceptable place for an elderly wearable device, this paper presents the development and evaluation of a wrist-worn fall detection solution. Different sensors (accelerometer, gyroscope, and magnetometer), signals (acceleration, velocity, and displacement), and direction components (vertical and non-vertical) were combined and a comprehensive set of threshold-based and machine learning methods were applied in order to define the best approach for fall detection. Data was acquired for fall and non-fall movements from 22 volunteers. For threshold-based methods, a maximum accuracy of 91.1% was achieved with 95.8% and 86.5% of sensitivity and specificity, respectively, using Madgwick’s decomposition. With the same movement decomposition and machine learning methods in the classification stage, an impressive accuracy of 99.0% was achieved, with 100% of sensitivity and 97.9% of specificity in our data set. Prolonged tests with a volunteer wearing the fall detector also demonstrate the advantages of machine learning methods in terms of practical applications.

Optimal Foot Location for Placing Wearable IMU Sensors and Automatic Feature Extraction for Gait Analysis

Abstract: Our aim is to maximize the interpretable information for gait analysis. To achieve this, it is important to find the optimal sensor placement and the parameters that influence the extraction of automatic gait features. We investigated the effect of different anatomical foot locations on inertial measurement unit (IMU) sensor output. We designed and developed an android app to collect real-time synchronous sensor output. We selected a set of five anatomical foot locations covering most of the foot regions to place wearable IMU sensors for data collection. Each participant performed a trial in a straight corridor comprising 25 strides of normal walking, a turn-around, and another 25 strides. We proposed an automatic gait features extraction method to analyze the data for stride number, distance, speed, length and period of stride, stance, and swing phases during walking. The highest accuracy for detecting stride number was in location 1 (first cuneiform) followed by location 5 (Achilles Tendon) and 4 (Talus). Location 1 was the closest to correlate estimate to the measured distance travelled. The accuracy of detecting number of strides on average is 95.47% from accelerometer data and 93.60% from gyroscope data and closest to the 60:40% split for average stance and swing for 15 subjects. To validate our results, we conducted trials using the Qualisys motion capture instrument and from our sensors concurrently. The average accuracy of the result is 97.77% with 95% confidence interval 0.767 for estimated and 99.01% with 95% confidence interval 0.266 for period.

Fault Feature Selection and Diagnosis of Rolling Bearings Based on EEMD and Optimized Elman_AdaBoost Algorithm

Abstract: To realize accurate fault diagnosis of rolling bearing under random noise, a novel fault diagnosis method based on ensemble empirical mode decomposition (EEMD) and optimized Elman_AdaBoost is proposed in this paper. First, the EEMD method is used to decompose the original vibration signal into several intrinsic mode functions (IMFs). Then, the correlation coefficient and kurtosis of IMFs are used to remove any excess ingredients, and the signal is reconstructed by using the rest of IMFs. In addition, based on the analysis of the time–frequency characteristics of the reconstructed signal, the root-mean-square value and the power spectrum center are extracted as a 2-D feature vector from each reconstructed signal and regarded as the input characteristic vector of the Elman neural network. Combined with the AdaBoost classification algorithm, the Elman_AdaBoost algorithm is improved and proposed to establish the strong classifier. Finally, based on the vibration data recorded in the bearing center of the Case Western Reserve University, three kinds of bearings with different faults and normal bearings are selected as the sample data and the training samples and test samples are constructed. Experimental results show that the proposed EEMD method and optimized Elman_AdaBoost model can effectively diagnose the rolling bearing under random noise, and has the advantages of fast speed, small error, and stable performance. Comparing with the traditional Elman_AdaBoost and single Elman neural network algorithm, this paper with fewer characteristics can get better accuracy and real-time processing performance, and presents a simple and practical resolution for the fault diagnosis of rolling bearings.

Wireless Sensor’s Civil Applications, Prototypes, and Future Integration Possibilities: A Review

Abstract: Advances in wireless communication are forging new possibilities for sensors. New sensors are equipping major systems around us with unparalleled intelligence as in the case of smart grids, smart homes, and driverless vehicles. Considering the current developments in the field of sensor networks, one feels that it has reached an interesting stage, where the role of the sensors becoming crucial in numerous applications. This all speaks volumes of the fact that sensors are going to be at the front and center of most of future technologies, needless to say the Internet of Things. Considering their vital role from futuristic perspective this survey reports variety of sensors along with their characteristics and applications, which impact human life and well-being. In addition, this survey considers recent prototypes, leading sensor manufacturers as well as major projects that have made use of sensors since the last decade. Moreover, significance of this effort is that integration possibilities of sensors with other networks and major technologies are discussed, while possible challenges and key benefits are highlighted. This research effort focuses the latest developments in the area of sensors and sensor networks as research gears up to meet the challenges of the emerging technologies and their applications particularly those that emphasize smart sensors.

Particle Swarm Optimization-Based Unequal and Fault Tolerant Clustering Protocol for Wireless Sensor Networks

Abstract: Clustering is one of most efficient energy saving techniques for maximizing network lifetime in wireless sensor networks. In the multi-hop approach, cluster heads (CHs) close to the base station deplete their energy very quickly due to high inter-cluster relay traffic load, causing the hot spot problem. Thus, a clustering protocol is required to be energy efficient as well as fault tolerant. This paper presents a particle swarm optimization (PSO)-based unequal and fault tolerant clustering protocol referred as PSO-UFC. The proposed protocol addresses imbalanced clustering and fault tolerance issues in the existing energy-balanced unequal clustering (EBUC) protocol for the long-run operation of the network. To solve the imbalanced clustering problem, the PSO-UFC protocol utilizes unequal clustering mechanism to balance intra-cluster and inter-cluster energy consumption between the Master CHs (MCHs). Also, in PSO-UFC protocol the network connectivity is restored by electing an extra CH called Surrogate CH due to sudden failure of MCH. The obtained simulation results demonstrate that the PSO-UFC protocol prolongs the network lifetime against EBUC, PSO-C, and LEACH-C protocols.

Design of a Wireless Sensor Network-Based IoT Platform for Wide Area and Heterogeneous Applications

Abstract: Internet of Things (IoT) is not only a promising research topic but also a blooming industrial trend. Although the basic idea is to bring things or objects into the Internet, there are various approaches, because an IoT system is highly application oriented. This paper presents a wireless sensor network(WSN)-based IoT platform for wide area and heterogeneous sensing applications. The platform, consisting of one or multiple WSNs, gateways, a Web server, and a database, provides a reliable connection between sensors at fields and the database on the Internet. The WSN is built based on the IEEE 802.15.4e time slotted channel hopping protocol, because it has the benefits such as multi-hop transmission, collision-free transmission, and high energy efficiency. In addition to the design of a customized hardware for range extension, a new synchronization scheme and a burst transmission feature are also presented to boost the network capacity and reduce the energy waste. As a result, the proposed platform can fulfill the high throughput requirement for high-rate applications and the requirement of long battery life for low-rate applications at the same time. We have developed a testbed in our campus to validate the proposed system.

A Three-Dimensional Fiber Bragg Grating Force Sensor for Robot

Abstract: This paper presents a three-dimensional (3-D) force sensor based on fiber Bragg grating (FBG) for robot plantar force measuring. A classical Maltese-cross beam with multiplexed FBGs has been designed for 3-D force sensing. Strain distribution characteristics and dynamic performance of the Maltese-cross elastomer have been investigated by using finite element analysis. Through ingenious design, 3-D forces of Fx, Fy, and Fz have been measured by only five sensitive elements of FBG, meanwhile, decoupling and temperature compensation have also been realized, which greatly reduces the number of sensitive element compared with the traditional resistance strain gauge based multi-axis force sensor. Comprehensive performance test has been carried out, and the experimental results demonstrate that the sensor possesses good linearity, weak coupling, and creep resistance.

Liquid Level Measurement Based on FBG-Embedded Diaphragms With Temperature Compensation

Abstract: This paper proposes a fiber optic liquid level sensor system based on a silica fiber Bragg grating embedded into an epoxy resin diaphragm coupled to a temperature reference sensor, used to compensate the temperature cross-sensitivity for improving the liquid level measurement accuracy. The proposed system was tested in an industrial water tank with heating and recirculation. The results demonstrated a temperature cross-sensitivity reduction, enhancing the liquid level measurement thermal stability by a factor of nine, when compared with some single head sensor configurations reported in literature. Our system presents high linearity ( $R>0.999$ ), superior sensitivity (2.8 pm/mm), and much lower temperature related error (1.04 mm/°C), when compared with the other diaphragm-based sensors recently reported in the literature.

Performance Enhancement of SPR Biosensor Based on Phosphorene and Transition Metal Dichalcogenides for Sensing DNA Hybridization

Abstract: Two-dimensional (2-D) materials have attracted a lot of attention for using in surface plasmon resonance (SPR) biosensors. High sensitivity and figure of merit are two desirable parameters in sensor performance analysis. Using heterostructures of phosphorene, graphene, and transition metal dichalcogenides is proposed to enhance the sensitivity of biosensor. Although the proposed biosensor can be used for detection of different analytes with a wide refractive index range, DNA hybridization is the sensing target in this paper. Sensing parameters of different biosensor configurations based on a few-layer black phosphorus (BP) and other 2-D materials are compared with the conventional Au-based SPR sensor and the highest sensitivity of 187°/RIU for the structure consists of 10-layer BP and monolayer WS2 is obtained. The suggested sensor structure provides the sensitivity more than two times of the conventional biosensor. This high sensor performance may have potential applications in medical diagnosis and biochemical detection.

Lane Detection and Classification for Forward Collision Warning System Based on Stereo Vision

Abstract: This paper presents a lightweight stereo vision-based driving lane detection and classification system to achieve the ego-car’s lateral positioning and forward collision warning to aid advanced driver assistance systems (ADAS). For lane detection, we design a self-adaptive traffic lanes model in Hough Space with a maximum likelihood angle and dynamic pole detection region of interests (ROIs), which is robust to road bumpiness, lane structure changing while the ego-car’s driving and interferential markings on the ground. What’s more, this model can be improved with geographic information system or electronic map to achieve more accurate results. Besides, the 3-D information acquired by stereo matching is used to generate an obstacle mask to reduce irrelevant objects’ interfere and detect forward collision distance. For lane classification, a convolutional neural network is trained by using manually labeled ROI from KITTI data set to classify the left/right-side line of host lane so that we can provide significant information for lane changing strategy making in ADAS. Quantitative experimental evaluation shows good true positive rate on lane detection and classification with a real-time (15Hz) working speed. Experimental results also demonstrate a certain level of system robustness on variation of the environment.

Multi-Band RF Planar Sensor Using Complementary Split Ring Resonator for Testing of Dielectric Materials

Abstract: In this paper, an attractive multi-band RF planar sensor, suitable for non-destructive testing of dispersive materials, is proposed. The proposed sensor is based on a number of complementary split ring resonator (CSRR) unit cells etched in the ground plane of the microstrip line. Each CSRR unit cell can be represented by a narrow band reject filter with its center frequency corresponding to the resonant frequency of the respective CSRR cell. The proposed technique is used to design the two, three and four band microwave sensors operating at 1.5 GHz, 2.45 GHz, 3.8 GHz, and 5.8 GHz. The distance between the two adjoining CSRRs is minimized for each case without appreciably increasing the inter-cell coupling effect. The transcendental equations required for determining the complex permittivity of the material under test in terms of the resonant frequency are derived from the numerical data obtained using the electromagnetic solver, the CST studio. These numerical equations are then used to obtain the dielectric properties of various test samples, which are measured using the vector network analyzer. The detailed air gap analysis is also performed for checking the accuracy of the designed planar sensor under the real situation. The proposed sensors are fabricated on 0.8 mm thick FR4 substrates using the standard photolithography technique. A number of standard samples are tested using the fabricated sensors in multiple frequency bands, and a good agreement between the obtained results and the data available in literature shows the applicability of the proposed scheme.

Twin Core Photonic Crystal Fiber Plasmonic Refractive Index Sensor

Abstract: This paper presents a highly sensitive plasmonic refractive index (RI) sensor on a dual-core photonic crystal fiber (PCF). The performance of the RI sensor is weighed in terms of amplitude sensitivity, wavelength sensitivity, wavelength resolution, and linearity of the resonance wavelength. Numerical result shows that the maximum amplitude sensitivity of 725.8918 RIU $^{-1}$ and 1085 RIU $^{-1}$ are achieved for $x$ - and $y$ -polarization, respectively, by using amplitude interrogation method. In addition, the wavelength interrogation method gives the maximum wavelength sensitivity of around 9000 nm/RIU for both $x$ - and $y$ -polarized modes. Moreover, the maximum wavelength resolution of $1.11\times 10^{-5}$ RIU and the value of the co-efficient of determination ( $\text{R}^{2}$ ) of 0.9784 are reported. The proposed high-performance plasmonic RI sensor can be a favorable candidate for the detection of biological and biochemical analytes.

Fault Diagnosis Based on Weighted Extreme Learning Machine With Wavelet Packet Decomposition and KPCA

Abstract: Fault diagnosis has received considerable attention because its implementation can effectively prevent costly and even catastrophic downtime. However, quickly identifying faults and accurately obtaining diagnosis results from a feature set of rotating machinery are still a problem. To this end, this paper proposes an effective method based on a weighted extreme learning machine (WELM) with wavelet packet decomposition (WPD) and kernel principal component analysis (KPCA). The feature set affecting classification accuracy can be obtained using WPD and KPCA. By taking feature reliability into consideration, a new type of improvement to the extreme learning machine (ELM), i.e., WELM, is proposed by associating the hidden layer and input layer with a weight matrix. The WELM model can help in guaranteeing a quick and an accurate identification of fault status. To verify the superiority of the fault identification speed and accuracy of the proposed method, results from other methods, namely, using the sensitive features based on WPD and KPCA with ELM, a back-propagation neural network, and a support vector machine, were compared. The experimental results indicate that the proposed method can effectively improve the accuracy and quickly diagnose the fault. The average accuracy of fault classification could reach 95.45%, and the computation time of WELM was only 0.0156 s.