Showing papers in "IEEE Sensors Journal in 2020"

Knowledge Transfer for Rotary Machine Fault Diagnosis

Abstract: This paper intends to provide an overview on recent development of knowledge transfer for rotary machine fault diagnosis (RMFD) by using different transfer learning techniques. After brief introduction of parameter-based, instance-based, feature-based and relevance-based knowledge transfer, the applications of knowledge transfer in RMFD are summarized from four categories: transfer between multiple working conditions, transfer between multiple locations, transfer between multiple machines, and transfer between multiple fault types. Case studies on four datasets including gears, bearing, and motor faults verified effectiveness of knowledge transfer on improving diagnostic accuracy. Meanwhile, research trends on transfer learning in the field of RMFD are discussed.

Fusion of 3D LIDAR and Camera Data for Object Detection in Autonomous Vehicle Applications

Abstract: It is vital that autonomous vehicles acquire accurate and real-time information about objects in their vicinity, which fully guarantees the safety of the passengers and vehicle in various environments. Three-dimensional light detection and ranging (3D LIDAR) sensors can directly obtain the position and geometric structure of an object within its detection range, whereas the use of vision cameras is most suitable for object recognition. Accordingly, in this paper, we present a novel object detection and identification method that fuses the complementary information obtained by two types of sensors. First, we utilise 3D LIDAR data to generate accurate object-region proposals. Then, these candidates are mapped onto the image space from which regions of interest (ROI) of the proposals are selected and input to a convolutional neural network (CNN) for further object recognition. To precisely identify the sizes of all the objects, we combine the features of the last three layers of the CNN to extract multi-scale features from the ROIs. The evaluation results obtained on the KITTI dataset demonstrate that: (1) unlike sliding windows that produce thousands of candidate object-region proposals, 3D LIDAR provides an average of 86 real candidates per frame and the minimal recall rate is better than 95%, which greatly decreases the extraction time; (2) The average processing time for each frame of the proposed method is only 66.79 ms, which meets the real-time demand of autonomous vehicles; (3) The average identification accuracies of our method for cars and pedestrians at a moderate level of difficulty are 89.04% and 78.18%, respectively, which is better than those of most previous methods.

Consumer Grade EEG Measuring Sensors as Research Tools: A Review

Abstract: Since the launch of the first consumer grade EEG measuring sensors ‘NeuroSky Mindset’ in 2007, the market has witnessed an introduction of at least one new product every year by competing manufacturers, which include NeuroSky, Emotiv, interaXon and OpenBCI. There are numerous variations in the make and versions, but these products clearly share the key selling points of affordability, portability, and ease of use. These features are patently well placed provided one of the main objectives for their development is to attract a new target group of commercial users. Nevertheless, with several decades of traditional EEG usage in clinical and experimental settings, the shift toward commercial and engineering sides has not been achieved without skepticism. With this in mind, researchers in related fields have been tirelessly working to ensure that these putatively novel features were not introduced at the expense of efficiency and accuracy by conducting validation studies to compare the performance of data derived from consumer grade EEG devices with ones from standard research grade counterparts. In this review, we seek to provide the detail of the products supplied by the major players, summarize studies that evaluate consumer product’s performance against research grade devices, the key areas of applications that these consumer grade devices have been employed in over the past five years or so, and finally give our perspectives on the limitations and what these innovative tools could offer going forward in terms of research and commercial applications.

A New Deep Transfer Learning Method for Bearing Fault Diagnosis Under Different Working Conditions

Abstract: Fault diagnosis is very important for condition based maintenance. Recently, deep learning models are introduced to learn hierarchical representations from raw data instead of using hand-crafted features, which exhibit excellent performance. The success of current deep learning lies in: 1) the training (source domain) and testing (target domain) datasets are from the same feature distribution; 2) Enough labeled data with fault information exist. However, because the machine operates under a non-stationary working condition, the trained model built on the source domain can not be directly applied on the target domain. Moreover, since no sufficient labeled or even unlabeled data are available in target domain, collecting the labeled data and building the model from scratch is time-consuming and expensive. Motivated by transfer learning (TL), we present a new fault diagnosis method, which generalizes convolutional neural network (CNN) to TL scenario. Two layers with regard to task-specific features are adapted in a layer-wise way to regularize the parameters of CNN. What's more, the domain loss is calculated by a linear combination of multiple Gaussian kernels so that the ability of adaptation is enhanced compared to single kernel. Through these two means, the distribution discrepancy is reduced and the transferable features are learned. The proposed method is validated by transfer fault diagnosis experiments. Compared to CNN without domain adaptation and shallow transfer learning methods, the proposed method gets the best performance for fault classification.

mm-Pose: Real-Time Human Skeletal Posture Estimation Using mmWave Radars and CNNs

Abstract: In this paper, mm-Pose , a novel approach to detect and track human skeletons in real-time using an mmWave radar, is proposed. To the best of the authors’ knowledge, this is the first method to detect >15 distinct skeletal joints using mmWave radar reflection signals. The proposed method would find several applications in traffic monitoring systems, autonomous vehicles, patient monitoring systems and defense forces to detect and track human skeleton for effective and preventive decision making in real-time. The use of radar makes the system operationally robust to scene lighting and adverse weather conditions. The reflected radar point cloud in range, azimuth and elevation are first resolved and projected in Range-Azimuth and Range-Elevation planes. A novel low-size high-resolution radar-to-image representation is also presented, that overcomes the sparsity in traditional point cloud data and offers significant reduction in the subsequent machine learning architecture. The RGB channels were assigned with the normalized values of range, elevation/azimuth and the power level of the reflection signals for each of the points. A forked CNN architecture was used to predict the real-world position of the skeletal joints in 3-D space, using the radar-to-image representation. The proposed method was tested for a single human scenario for four primary motions, (i) Walking, (ii) Swinging left arm, (iii) Swinging right arm, and (iv) Swinging both arms to validate accurate predictions for motion in range, azimuth and elevation. The detailed methodology, implementation, challenges, and validation results are presented.

Bi-LSTM Network for Multimodal Continuous Human Activity Recognition and Fall Detection

Abstract: This paper presents a framework based on multi-layer bi-LSTM network (bidirectional Long Short-Term Memory) for multimodal sensor fusion to sense and classify daily activities’ patterns and high-risk events such as falls. The data collected in this work are continuous activity streams from FMCW radar and three wearable inertial sensors on the wrist, waist, and ankle. Each activity has a variable duration in the data stream so that the transitions between activities can happen at random times within the stream, without resorting to conventional fixed-duration snapshots. The proposed bi-LSTM implements soft feature fusion between wearable sensors and radar data, as well as two robust hard-fusion methods using the confusion matrices of both sensors. A novel hybrid fusion scheme is then proposed to combine soft and hard fusion to push the classification performances to approximately 96% accuracy in identifying continuous activities and fall events. These fusion schemes implemented with the proposed bi-LSTM network are compared with conventional sliding window approach, and all are validated with realistic “leaving one participant out” (L1PO) method (i.e. testing subjects unknown to the classifier). The developed hybrid-fusion approach is capable of stabilizing the classification performance among different participants in terms of reducing accuracy variance of up to 18.1% and increasing minimum, worst-case accuracy up to 16.2%.

Image Dehazing by an Artificial Image Fusion Method Based on Adaptive Structure Decomposition

Abstract: Haze can seriously affect the visible and visual quality of outdoor images. As a challenge in practice, image dehazing techniques are always used to remove haze from the captured images. Existing image dehazing algorithms focus on enhancing both global image contrast and saturation, but ignore the local enhancement. So the dehazed images do not often have good performance in the visual quality of local details. This paper proposes a new single-image dehazing solution based on the adaptive structure decomposition integrated multi-exposure image fusion (PADMEF). A set of underexposed image sequences are extracted from a single blurred image first by a series of gamma correction and the spatial linear adjustment of saturation. Then different exposure-level images are fused into a haze-free image by applying a multi-exposure image fusion (MEF) scheme based adaptive structure decomposition to each image patch. The proposed image dehazing scheme can effectively eliminate the visual degradation caused by haze without the physical model inversion of haze formation. Both apriori estimation of scene depth and the expensive refinement process of depth mapping can be avoided. The entropy of image texture named as texture energy is used to measure the image energy and obtain the information size contained in an image. Meanwhile, a texture energy based method is presented to adaptively select the corresponding patch size for the decomposition of image structure. In addition, this paper verifies that the dehazed images obtained by the patch based MEF algorithm always meet the requirements of intensity decrease. The comparative experiment results are evaluated in both qualitative and quantitative aspects, which confirm the effectiveness of the proposed solution in haze removal.

Continuous Human Activity Classification From FMCW Radar With Bi-LSTM Networks

Abstract: Recognition of human movements with radar for ambient activity monitoring is a developed area of research that yet presents outstanding challenges to address In real environments, activities and movements are performed with seamless motion, with continuous transitions between activities of different duration and a large range of dynamic motions, compared with discrete activities of fixed-time lengths which are typically analysed in the literature This paper proposes a novel approach based on recurrent LSTM and Bi-LSTM network architectures for continuous activity monitoring and classification This approach uses radar data in the form of a continuous temporal sequence of micro-Doppler or range-time information, differently from from other conventional approaches based on convolutional networks that interpret the radar data as images Experimental radar data involving 15 participants and different sequences of 6 actions are used to validate the proposed approach It is demonstrated that using the Doppler-domain data together with the Bi-LSTM network and an optimal learning rate can achieve over 90% mean accuracy, whereas range-domain data only achieved approximately 76% The details of the network architectures, insights in their behaviour as a function of key hyper-parameters such as the learning rate, and a discussion on their performance across are provided in the paper

Security of RPL Based 6LoWPAN Networks in the Internet of Things: A Review

Abstract: Internet of Things (IoT) is one of the fastest emerging networking paradigms enabling a large number of applications for the benefit of mankind. Advancements in embedded system technology and compressed IPv6 have enabled the support of IP stack in resource constrained heterogeneous smart devices. However, global connectivity and resource constrained characteristics of smart devices have exposed them to different insider and outsider attacks, which put users’ security and privacy at risk. Various risks associated with IoT slow down its growth and become an obstruction in the worldwide adoption of its applications. In RFC 6550, the IPv6 Routing Protocol for Low Power and Lossy Network (RPL) is specified by IETF’s ROLL working group for facilitating efficient routing in 6LoWPAN networks, while considering its limitations. Due to resource constrained nature of nodes in the IoT, RPL is vulnerable to many attacks that consume the node’s resources and degrade the network’s performance. In this paper, we present a study on various attacks and their existing defense solutions, particularly to RPL. Open research issues, challenges, and future directions specific to RPL security are also discussed. A taxonomy of RPL attacks, considering the essential attributes like resources, topology, and traffic, is shown for better understanding. In addition, a study of existing cross-layered and RPL specific network layer based defense solutions suggested in the literature is also carried out.

Sensing Data Supported Traffic Flow Prediction via Denoising Schemes and ANN: A Comparison

Abstract: Short-term traffic flow prediction plays a key role of Intelligent Transportation System (ITS), which supports traffic planning, traffic management and control, roadway safety evaluation, energy consumption estimation, etc. The widely deployed traffic sensors provide us numerous and continuous traffic flow data, which may contain outlier samples due to expected sensor failures. The primary objective of the study was to evaluate the use of various smoothing models for cleaning anomaly in traffic flow data, which were further processed to predict short term traffic flow evolution with artificial neural network. The wavelet filter, moving average model, and Butterworth filter were carefully tested to smooth the collected loop detector data. Then, the artificial neural network was introduced to predict traffic flow at different time spans, which were quantitatively analyzed with commonly-used evaluation metrics. The findings of the study provide us efficient and accurate denoising approaches for short term traffic flow prediction.

Classification of Hand Movements From EEG Using a Deep Attention-Based LSTM Network

Abstract: Classifying limb movements using brain activity is an important task in Brain-computer Interfaces (BCI) that has been successfully used in multiple application domains, ranging from human-computer interaction to medical and biomedical applications. This paper proposes a novel solution for classification of left/right hand movement by exploiting a Long Short-Term Memory (LSTM) network with attention mechanism to learn the electroencephalogram (EEG) time-series information. To this end, a wide range of time and frequency domain features are extracted from the EEG signals and used to train an LSTM network to perform the classification task. We conduct extensive experiments with the EEG Movement dataset and show that our proposed solution our method achieves improvements over several benchmarks and state-of-the-art methods in both intra-subject and cross-subject validation schemes. Moreover, we utilize the proposed framework to analyze the information as received by the sensors and monitor the activated regions of the brain by tracking EEG topography throughout the experiments.

The Layer-Wise Training Convolutional Neural Networks Using Local Loss for Sensor-Based Human Activity Recognition

Abstract: Recently, deep learning, which are able to extract automatically features from data, has achieved state-of-the-art performance across a variety of sensor based human activity recognition (HAR) tasks. However, the existing deep neural networks are usually trained with a global loss, and all hidden layer weights have to be always kept in memory before the forward and backward pass has completed. The backward locking phenomenon prevents the reuse of memory, which is a crucial limitation for wearable activity recognition. In the paper, we proposed a layer-wise convolutional neural networks (CNN) with local loss for the use of HAR task. To our knowledge, this paper is the first that uses local loss based CNN for HAR in ubiquitous and wearable computing arena. We performed experiments on five public HAR datasets including UCI HAR dataset, OPPOTUNITY dataset, UniMib-SHAR dataset, PAMAP dataset, and WISDM dataset. The results show that local loss works better than global loss for tested baseline architectures. At no extra cost, the local loss can approach the state-of-the-arts on a variety of HAR datasets, even though the number of parameters was smaller. We believe that the layer-wise CNN with local loss can be used to update the existing deep HAR methods.

A Novel Trilateration Algorithm for RSSI-Based Indoor Localization

Abstract: This paper proposed a novel trilateration algorithm for indoor localization based on received signal strength indication (RSSI). Firstly, all the raw measurement data are preprocessed by a Gaussian filter to reducing the influence of measurement noise. Secondly, the transmit power and the path loss exponent are estimated by a novel least-squares curve fitting (LSCF) method in the RSSI-based localization. Thirdly, a novel trilateration algorithm is proposed based on the extreme value theory, which constructs a nonlinear error function depending on distances and anchor nodes position. To minimize the function, a Taylor series approximation can be used for reduce the computational complexity. And, an iteration condition is designed to further improve the positioning accuracy. Afterward, Bayesian filtering is used to smoothing the localization error, and decrease the influence of the process noise. Both the simulation and experimental results demonstrate the effectiveness of the proposed methodology.

PP-Net: A Deep Learning Framework for PPG-Based Blood Pressure and Heart Rate Estimation

Abstract: This paper presents a deep learning model ’PP-Net’ which is the first of its kind, having the capability to estimate the physiological parameters: Diastolic blood pressure (DBP), Systolic blood pressure (SBP), and Heart rate (HR) simultaneously from the same network using a single channel PPG signal. The proposed model is designed by exploiting the deep learning framework of Long-term Recurrent Convolutional Network (LRCN), exhibiting inherent ability of feature extraction, thereby, eliminating the cost effective steps of feature selection and extraction, making less-complex for deployment on resource constrained platforms such as mobile platforms. The performance demonstration of the PP-Net is done on a larger and publically available MIMIC-II database. We achieved an average NMAE of 0.09 (DBP) and 0.04 (SBP) mmHg for BP, and 0.046 bpm for HR estimation on total population of 1557 critically ill subjects. The accurate estimation of HR and BP on a larger population compared to the existing methods, demonstrated the effectiveness of our proposed deep learning framework. The accurate evaluation on a huge population with CVD complications, validates the robustness of the proposed framework in pervasive healthcare monitoring especially cardiac and stroke rehabilitation monitoring.

Printed Temperature Sensor Based on PEDOT: PSS-Graphene Oxide Composite

Abstract: Temperature sensing is an important parameter needed to be measured by the eSkin during the physical interaction of robots with real-world objects. Yet, most of the work on sensors in eSkin has focused on pressure sensing. Here we present a skin conformable printed temperature sensor with poly(3,4-ethylenedioxythiophene): poly (styr-enesulfonate) (PEDOT:PSS)-graphene oxide (GO) as a temperature sensitive layer and silver (Ag) as contact electrodes. The demonstration of PEDOT:PSS/GO as a highly temperature sensitive layer is the distinct feature of the work. The response of presented sensor observed over ~25 °C (room temperature (RT)) to 100°C, by measuring the variation in resistance across the GO/PEDOT:PSS layer showed ~80% decrease in resistance. The sensitivity of the sensor was found to be 1.09% per °C. The sensor’s response was also observed under static and dynamic bending (for 1000 cycles) conditions. The stable and repeatable response of sensor, in both cases, signifies strong adhesion of the layers with negligible delamination or debonding. In comparison to the commercial thermistor, the printed GO/PEDOT:PSS sensor is faster (~73% superior) with response and recovery times of 18 s and 32 s respectively. Finally, the sensor was attached to a robotic hand to allow the robot to act by using temperature feedback.

Recent Advances and Tendency in Fiber Bragg Grating-Based Vibration Sensor: A Review

Abstract: Vibration sensing is critical to monitor and ultimately preserve the health state of engineering systems. These systems with a large structure are typically working in some harsh environments including strong magnetic fields. However, traditional electrical sensors are difficult to accurately measure the vibration under harsh environments. Besides these instinct advantages of normal fiber optic sensors (FOS) sensors such as compact size, passive sensing, resistance to electromagnetic interference, etc., fiber Bragg grating (FBG) sensors have a capability of distributed sensing based on wavelength demodulation and resistance to light intensity fluctuation and unwanted fiber bending losses. Such merits lead them to be a hot topic in FOS field and excellent candidates for vibration sensing. Three types of FBG-based vibration sensors have been classified based on the difference of vibration-strain coupling way to FBG in this survey, which are pasted FBG-based, axial property of FBG-based and transverse property of FBG-based, respectively. FBG-based vibration sensors' principles and designs have been introduced and discussed. Recent advances in the applications of FBG-based vibration sensors have been investigated. The limitations and prospects of the FBG-based vibration sensing technologies have been analyzed and discussed.

MEMS-Based IMU Drift Minimization: Sage Husa Adaptive Robust Kalman Filtering

Abstract: The Attitude Heading Reference System (AHRS) has been widely used to provide the position and orientation of a rigid body. A low cost MEMS based inertial sensor measurement unit (IMU) is a core device in AHRS. To improve the AHRS system performance, there is a need to develop (i) stochastic IMU error models and (ii) random noise minimization techniques. In this paper, we modify the Sage-Husa Adaptive Kalman Filter (SHAKF) to incorporate time-varying noise estimator and robustifier, termed as Modified Sage-Husa Adaptive Robust Kalman Filter (MSHARKF). In the proposed algorithm, a three segment approach is used to evaluate the adaptive scale factor followed by the learning statistics. The scale factor is iteratively updated in the MSHARKF equations. In addition, angle random walk (ARW) and bias instability (BI) errors are represented by state-space models. The proposed algorithm is applied to restrain the drift error and random noise in the MEMS IMUs signals. The performance of this algorithm is analyzed using Allan variance (AV) analysis for static signals whereas the Root Mean Square Error (RMSE) values are evaluated for dynamic signals. Experimental results demonstrate the effectiveness of MSHARKF in reducing the drift and random noise in static and dynamic conditions as compared with other existing algorithms. Finally, we present sufficient conditions for convergence proof of the MSHARKF algorithm.

WiAct: A Passive WiFi-Based Human Activity Recognition System

Abstract: Nowadays, human behavior recognition research plays a pivotal role in the field of human-computer interaction. However, comprehensive approaches mainly rely on video camera, ambient sensors or wearable devices, which either require arduous deployment or arouse privacy concerns. In this paper, we propose WiAct, a passive WiFi-based human activity recognition system, which explores the correlations between body movement and the amplitude information in Channel State Information (CSI) to classify different activities. The system designs a novel Adaptive Activity Cutting Algorithm (AACA) based on the difference in signal variance between the action and non-action parts, which adjusts the threshold adaptively to achieve the best trade-off between performance and robustness. The Doppler shift correlation value is used as classification features, which is extracted by using the correlation of the WiFi device’s antennas. Extreme Learning Machine (ELM) is utilized for activity data classification because of its strong generalization ability and fast learning speed. We implement the WiAct prototype using commercial WiFi equipment and evaluate its performance in real-world environments. In the evaluation, WiAct achieves an average accuracy of 94.2% for distinguishing ten actions. We compare different experimental conditions and classification methods, and the results demonstrate its robustness.

A New Hybrid Fault Detection Method for Wind Turbine Blades Using Recursive PCA and Wavelet-Based PDF

Abstract: This paper introduces a new condition monitoring approach for extracting fault signatures in wind turbine blades by utilizing the data from a real-time Supervisory Control and Data Acquisition (SCADA) system. A hybrid fault detection system based on a combination of Generalized Regression Neural Network Ensemble for Single Imputation (GRNN-ESI) algorithm, Principal Component Analysis (PCA), and wavelet-based Probability Density Function (PDF) approach is proposed in this work. The proposed fault detection strategy accurately detects incipient blade failures and leads to improved maintenance cost and availability of the system. Experimental test results based on data from a wind farm in southwestern Ontario, Canada, illustrate the effectiveness and high accuracy of the proposed monitoring approach.

Real-Time Radar-Based Gesture Detection and Recognition Built in an Edge-Computing Platform

Abstract: In this paper, a real-time signal processing framework based on a 60 GHz frequency-modulated continuous wave (FMCW) radar system to recognize gestures is proposed. In order to improve the robustness of the radar-based gesture recognition system, the proposed framework extracts a comprehensive hand profile, including range, Doppler, azimuth and elevation, over multiple measurement-cycles and encodes them into a feature cube. Rather than feeding the range-Doppler spectrum sequence into a deep convolutional neural network (CNN) connected with recurrent neural networks, the proposed framework takes the aforementioned feature cube as input of a shallow CNN for gesture recognition to reduce the computational complexity. In addition, we develop a hand activity detection (HAD) algorithm to automatize the detection of gestures in real-time case. The proposed HAD can capture the time-stamp at which a gesture finishes and feeds the hand profile of all the relevant measurement-cycles before this time-stamp into the CNN with low latency. Since the proposed framework is able to detect and classify gestures at limited computational cost, it could be deployed in an edge-computing platform for real-time applications, whose performance is notedly inferior to a state-of-the-art personal computer. The experimental results show that the proposed framework has the capability of classifying 12 gestures in real-time with a high ${F}_{\sf 1}$ -score.

Sensor Technologies for Fall Detection Systems: A Review

Abstract: The risk of falls in older adults restrict their social life and independent living. The assisted living devices help older adults to live independently in their home, giving a psychological boost, and releasing the burden on the caregiver and the healthcare providers. A robust and accurate fall detection system is essential to provide immediate help and to reduce the severe post-fall consequences, and the associated medical care cost significantly. This review aims to provide a comprehensive technical insight into the existing fall detection system, to classify various approaches and the challenges encountered during implementation. The fall detectors are broadly classified into three categories, namely wearable, ambiance-based, and hybrid sensing detectors, which are further explored by the sensor technology. This review provides a comprehensive overview of each competing sensor technology ranging from an accelerometer, pressure sensor, and radar to camera-based and their infusion into a complete fall detection system. It outlines the strength and limitations of different sensor fall detection systems in terms of feature extraction, classification, performance, and experimental dataset. The user adaptability, installation complexity, and power requirement of the systems are the main areas, which are not addressed adequately in the literature. In the end, the review provides a basic framework in deciding the technology for a specific scenario or location according to the prerequisites for the deployment.

Potential Chipless RFID Sensors for Food Packaging Applications: A Review

Abstract: Recently, there is an increasing demand for food safety and quality monitoring in the agrifood sector, needing active and smart packaging. Radio Frequency Identification (RFID) has the potential to become one of the most promising technology for the food industry since it can satisfy all the requirements of a sensor and identification in food safety monitoring, packages tracking, inventory control, early warning, and easy check-out. The purpose of this paper is to introduce the recent advancement of the chipless RFID humidity, temperature, gas, pH, and bio-sensors, as these types of sensors are required for evaluating the quality of the food in the food industry. Additionally, the paper explains the requirements of the food industry for quality monitoring of the food products, covers the existing challenges of chipless RFID sensor technology, and highlights the future direction of the research.

Dual-Core Photonic Crystal Fiber-Based Plasmonic RI Sensor in the Visible to Near-IR Operating Band

Abstract: In this paper, a dual-core photonic crystal fiber (DC-PCF) based surface plasmon resonance (SPR) bio-compatible sensor is proposed for various bio-organic molecules and biochemical analytes refractive index (RI) detection in the visible to near-infrared region (0.5 to $2~\mu \text{m}$ ). Two hexagonal ring lattice with all circular air-holes are used to simplify the sensor structure. To make the practical applications feasible, plasmonic material and analyte sensing layer both are employed at the outer surface of the fiber. Noble plasmonic material gold (Au) having a thickness of 30 nm is used to excite the surface plasmons. A thin layer of titanium oxide (TiO2) having a thickness of 5 nm is also considered as an adhesive layer between the Au and silica glass. The sensor response is investigated using the mode solver based finite element method (FEM). Numerical results indicate that the proposed sensor shows a maximum amplitude sensitivity (AS) of 6829 RIU $^{-1}$ , amplitude resolution (AR) of $5\times 10 ^{-6}$ RIU, maximum wavelength sensitivity (WS) of 28,000 nm/RIU, and wavelength resolution (WR) of $3.57\times 10 ^{-6}$ RIU, using the amplitude and wavelength interrogation methods, respectively. Moreover, a maximum figure of merit (FOM) of 2800 RIU $^{-1}$ is obtained, which is the highest among the reported PCF-SPR sensor. Owing to the promising sensitivity and simple structure, the proposed sensor can be potentially applicable for the detection of biochemical solutions and biological samples.

Accurate Empirical Path-Loss Model Based on Particle Swarm Optimization for Wireless Sensor Networks in Smart Agriculture

Abstract: Wireless sensor networks (WSNs) have received significant attention in the last few years in the agriculture field. Among the major challenges for sensor nodes’ deployment in agriculture is the path loss in the presence of dense grass or the height of trees. This results in degradation of communication link quality due to absorption, scattering, and attenuation through the crop’s foliage or trees. In this study, two new path-loss models were formulated based on the MATLAB curve-fitting tool for ZigBee WSN in a farm field. The path loss between the router node (mounted on a drone) and the coordinator node was modeled and derived based on the received signal strength indicator (RSSI) measurements with the particle swarm optimization (PSO) algorithm in the farm field. Two path-loss models were formulated based on exponential (EXP) and polynomial (POLY) functions. Both functions were combined with PSO, namely, the hybrid EXP-PSO and POLY-PSO algorithms, to find the optimal coefficients of functions that would result in accurate path-loss models. The results show that the hybrid EXP-PSO and POLY-PSO models noticeably improved the coefficient of determination (R2) of the regression line, with the mean absolute error (MAE) found to be 1.6 and 2.7 dBm for EXP-PSO and POLY-PSO algorithms. The achieved R2 in this study outperformed the previous state-of-the-art models. An accurate path-loss model is essential for smart agriculture application to determine the behavior of the propagated signals and to deploy the nodes in the WSN in a position that ensures data communication without unnecessary packets’ loss between nodes.

Printed Carbon Nanotubes-Based Flexible Resistive Humidity Sensor

Abstract: A resistive flexible humidity sensor based on multi-walled carbon nanotubes (MWCNTs) was designed and fabricated. Screen and gravure printing processes were used for monolithically fabricating the humidity sensor containing interdigitated electrodes (IDE), a sensing layer and a meandering conductive heater. An average thickness and surface roughness of $0.99~\mu \text{m}$ and $0.23~\mu \text{m}$ , respectively, was registered for the printed MWCNTs sensing layer. The capability of the printed sensor, with heater, was investigated by subjecting it to relative humidity (RH) ranging from 10% to 90%. The response demonstrated an overall resistance change of 55% when the sensor was subjected to 90% RH, when compared to 10% RH. A maximum hysteresis of 5.1%, at 70% RH, was calculated for the resistive response of the sensor. The printed sensors can be bend with radius of curvature of 1.5 inch with literally no effect.

Time-Frequency Domain Deep Convolutional Neural Network for the Classification of Focal and Non-Focal EEG Signals

Abstract: The neurological disease such as the epilepsy is diagnosed using the analysis of electroencephalogram (EEG) recordings. The areas of the brain associated with the consequence of epilepsy are termed as epileptogenic regions. The focal EEG signals are generated from epileptogenic areas, and the nonfocal signals are obtained from other regions of the brain. Thus, the classification of the focal and non-focal EEG signals are necessary for locating the epileptogenic areas during surgery for epilepsy. In this paper, we propose a novel method for the automated classification of focal and non-focal EEG signals. The method is based on the use of the synchrosqueezing transform (SST) and deep convolutional neural network (CNN) for the classification. The time-frequency matrices of EEG signal are evaluated using both Fourier SST (FSST) and wavelet SST (WSST). The two-dimensional (2D) deep CNN is used for the classification using the time-frequency matrix of EEG signals. The experimental results reveal that the proposed method attains the accuracy, sensitivity, and specificity values of more than 99% for the classification of focal and non-focal EEG signals. The method is compared with existing approaches for the discrimination of focal and non-focal categories of EEG signals.

Autocorrelation Aided Random Forest Classifier-Based Bearing Fault Detection Framework

Abstract: Rolling bearing defects in induction motors are usually diagnosed using vibration signal analysis. For accurate detection of rolling bearing defects, appropriate feature extraction from vibration signals is necessary, failure of which may lead to incorrect interpretation. Considering the above fact, this article presents an autocorrelation aided feature extraction method for diagnosis of rolling bearing defects. To this end, the vibration signals of healthy as well as different faulty bearings were recorded using accelerometers and autocorrelation of the respective vibration signals were done to examine their self-similarity in time scale. Following this, several statistical, hjorth as well as non-linear features were extracted from the respective vibration correlograms and were subjected to feature reduction using recursive feature elimination technique. The dimensionally reduced top ranked feature vectors were subsequently fed to a random forest classifier for classification of vibration signals. A large number of experiments were carried out for (i) three different fault diameters at (ii) four different shaft speeds and also at (iii) two different sampling frequencies. Besides, for each condition, six binary class and one multiclass classification problem is also addressed in this paper, resulting in a total 112 different classification tasks. It was observed that the proposed method has yielded very high accuracy in identifying different bearing defects which can be practically implemented for automated bearing fault detection of induction motors.

A Machine Learning Approach to Road Surface Anomaly Assessment Using Smartphone Sensors

Abstract: Road surface quality is essential for improving driving experience and reducing traffic accidents. Traditional road condition monitoring systems are limited in their temporal (speed) and spatial (coverage) responses needed for maintaining overall road quality. Several alternative systems have been proposed that utilize sensors mounted on vehicles. In particular, with the ubiquitous use of smartphones for navigation, smartphone-based road condition assessment has emerged as a promising new approach. In this paper, we propose to analyze different multiclass supervised machine learning techniques to effectively classify road surface conditions using accelerometer, gyroscope and GPS data collected from smartphones. Our work focuses on classification of three main class labels- smooth road, potholes, and deep transverse cracks. We hypothesize that using features from all three axes of the sensors provides more accurate results as compared to using features from only one axis. We also investigate the performance of deep neural networks to classify road conditions with and without explicit manual feature extraction. Our results indicate that models trained with features from all axes of the smartphone sensors outperform models that use only one axis. We also observe that the use of neural networks provides a significantly improved data classification. The machine learning approach discussed here can be implemented on a larger scale to monitor roads for defects that present a safety risk to commuters as well as to provide maintenance information to relevant authorities.

UWB-Based Localization System Aided With Inertial Sensor for Underground Coal Mine Applications

Abstract: Robotic mining equipment plays an increasingly important role in the coal mining industry. Due to the complexity of the confined underground environment, available localization methods are limited, and restrict the development of coal mine robots (CMRs). Ultra-wideband (UWB) is a promising positioning sensor with high ranging accuracy. However, current applications about UWB positioning in coal mine focus mainly on position information, but rarely on orientation information. Positioning accuracy is often plagued by the loss of transmitted signals and multipath effects. In this paper, a pseudo-GPS positioning system in underground coal mine, composed by noisy UWB range measurements, is proposed to provide localization service for CMRs. An Error-State Kalman Filter (ESKF) is used for fusing measurements from the inertial measurement unit (IMU) and the established UWB positioning system. Then the complete six degree of freedom (6-DOF) state estimation can be realized. Meanwhile the biases of the IMU and the translation parameters of IMU w.r.t. UWB mobile node are also estimated online to adapt to long-term operation in harsh underground environments. In addition, an UWB anchor optimal deployment strategy is discussed to deploy UWB nodes appropriately in the laneway, and maintain realistic positioning accuracy for CMR in the meantime. A large number of field tests in different environments including the actual underground coal mine were conducted. The experimental results showed that our method could obtain the pose estimation performance close to the state-of-the-art lidar odometry approach that has been currently utilized in underground coal mine, providing robust and precise localization estimation for CMR applications.

Impedance and Noise of Passive and Active Dry EEG Electrodes: A Review

Abstract: Dry electrodes are a promising solution for prolonged EEG signal acquisition, whereas wet electrodes may lose their signal quality in the same situation and require skin preparation for set-up. Here, we review the impedance and noise of passive and active dry EEG electrodes. In addition, we compare noise and input impedance of the EEG amplifiers. As there are multiple definitions of impedance in each EEG system, they are all first defined. Electrodes must be compatible with amplifiers to accurately record EEG signals. This implies that their impedance plays a significant role in amplifier compatibility and affects total input-referred noise. Therefore, we review the impedance and noise of state-of-the-art amplifiers and electrodes. Furthermore, we compare the various structures and materials used and their final impedance to that of wet electrodes. Finally, we compare state-of-the-art electrodes and amplifiers to the standards of the IFCN and IEC80601-2-26. We investigate bottlenecks and propose a guideline for future work on passive and active dry electrodes, as well as EEG amplifiers.