Sensors

The Varroa destructor mite is one of the most dangerous Honey Bee (Apis mellifera) parasites worldwide and the bee colonies have to be regularly monitored in order to control its spread. In this paper we present an object detector based method for health state monitoring of bee colonies. This method has the potential for online measurement and processing. In our experiment, we compare the YOLO and SSD object detectors along with the Deep SVDD anomaly detector. Based on the custom dataset with 600 ground-truth images of healthy and infected bees in various scenes, the detectors reached the highest F1 score up to 0.874 in the infected bee detection and up to 0.714 in the detection of the Varroa destructor mite itself. The results demonstrate the potential of this approach, which will be later used in the real-time computer vision based honey bee inspection system. To the best of our knowledge, this study is the first one using object detectors for the Varroa destructor mite detection on a honey bee. We expect that performance of those object detectors will enable us to inspect the health status of the honey bee colonies in real time. Full article

A new slot-based antenna design capable of producing horizontal polarization for unmanned aerial vehicle (UAV) ground control station (GCS) applications is outlined in this paper. The proposed antenna consists of oversize coaxial cylinders, slots, and slot-feed assembly. Each of the four vertical slots, arranged periodically around the antenna’s outer cylinder, emits a horizontally polarized broad beam of radiation, in phase, to produce an omnidirectional pattern. The antenna possesses a low-ripple ±0.5 dB in azimuth gain (yaw) due to its symmetric axis shape and an enclosed feed within itself, which does not radiate and interfere with the main azimuth pattern. This is crucial for a UAV GCS to symmetrically extend its coverage range in all directions against yaw planes. Simulation and measurement results reveal that the antenna maintains stable gain in the omnidirectional pattern (+0.5 dB) over the entire operational frequency band (2.55 GHz to 2.80 GHz), where S11 is lower than −10 dB. A further advantage of this configuration is its enhanced polarization purity of −40 dB over the full frequency band. The direct-current (DC) grounding approach used in this antenna is beneficial due to its electrostatic discharge (ESD) and lightning protection. Furthermore, its aerodynamic, self-supporting, and surface-mount structural shape makes this antenna a good and worthy choice for a UAV GCS. Full article

This paper presents the design and implementation of a supervisory control and data acquisition (SCADA) system for automatic fault detection. The proposed system offers advantages in three areas: the prognostic capacity for preventive and predictive maintenance, improvement in the quality of the machined product and a reduction in breakdown times. The complementary technologies, the Industrial Internet of Things (IIoT) and various machine learning (ML) techniques, are employed with SCADA systems to obtain the objectives. The analysis of different data sources and the replacement of specific digital sensors with analog sensors improve the prognostic capacity for the detection of faults with an undetermined origin. Also presented is an anomaly detection algorithm to foresee failures and to recognize their occurrence even when they do not register as alarms or events. The improvement in machine availability after the implementation of the novel system guarantees the accomplishment of the proposed objectives. Full article

Recently, vast investments have been made worldwide in developing Cyber-Physical Systems (CPS) as solutions to key socio-economic challenges. The Internet-of-Things (IoT) has also enjoyed widespread adoption, mostly for its ability to add “sensing” and “actuation” capabilities to existing CPS infrastructures. However, attention must be paid to the impact of IoT protocols on the dependability of CPS infrastructures. We address the issues of CPS dependability by using an epidemic model of the underlying dynamics within the CPS’ IoT subsystem (CPS-IoT) and an interference-aware routing reconfiguration. These help to efficiently monitor CPS infrastructure—avoiding routing oscillation, while improving its safety. The contributions of this paper are threefold. Firstly, a CPS orchestration model is proposed that relies upon: (i) Inbound surveillance and outbound actuation to improve dependability and (ii) a novel information diffusion model that uses epidemic states and diffusion sets to produce diffusion patterns across the CPS-IoT. Secondly, the proposed CPS orchestration model is numerically analysed to show its dependability for both sensitive and non-sensitive applications. Finally, a novel interference-aware clustering protocol called “INMP”, which enables network reconfiguration through migration of nodes across clusters, is proposed. It is then bench-marked against prominent IoT protocols to assess its impact on the dependability of the CPS. Full article

In recent years, various studies have begun to use deep learning models to conduct research in the field of human activity recognition (HAR). However, there has been a severe lag in the absolute development of such models since training deep learning models require a lot of labeled data. In fields such as HAR, it is difficult to collect data and there are high costs and efforts involved in manual labeling. The existing methods rely heavily on manual data collection and proper labeling of the data, which is done by human administrators. This often results in the data gathering process often being slow and prone to human-biased labeling. To address these problems, we proposed a new solution for the existing data gathering methods by reducing the labeling tasks conducted on new data based by using the data learned through the semi-supervised active transfer learning method. This method achieved 95.9% performance while also reducing labeling compared to the random sampling or active transfer learning methods. Full article

Due to the increase in aging structures and the decrease in construction workforce, there is an increasing interest in automating structural damage monitoring. Surface damage on concrete structures, such as cracks, delamination, and rebar exposure, is one of the important parameters that can be used to estimate the condition of the structure. In this paper, deep learning-based detection and quantification of structural damage using structured lights and a depth camera is proposed. The proposed monitoring system is composed of four lasers and a depth camera. The lasers are projected on the surface of the structures, and the camera captures images of the structures while measuring distance. By calculating an image homography, the captured images are calibrated when the structure and sensing system are not in parallel. The Faster RCNN (Region-based Convolutional Neural Network) with Inception Resnet v2 architecture is used to detect three types of surface damage: (i) cracks; (ii) delamination; and (iii) rebar exposure. The detected damage is quantified by calculating the positions of the projected laser beams with the measured distance. The experimental results show that structural damage was detected with an F1 score of 0.83 and a median value of the quantified relative error of less than 5%. Full article

Photodetectors based on transition metal dichalcogenides (TMDs) have been widely reported in the literature and molybdenum disulfide (MoS₂) has been the most extensively explored for photodetection applications. The properties of MoS₂, such as direct band gap transition in low dimensional structures, strong light–matter interaction and good carrier mobility, combined with the possibility of fabricating thin MoS₂ films, have attracted interest for this material in the field of optoelectronics. In this work, MoS2-based photodetectors are reviewed in terms of their main performance metrics, namely responsivity, detectivity, response time and dark current. Although neat MoS2-based detectors already show remarkable characteristics in the visible spectral range, MoS₂ can be advantageously coupled with other materials to further improve the detector performance Nanoparticles (NPs) and quantum dots (QDs) have been exploited in combination with MoS₂ to boost the response of the devices in the near ultraviolet (NUV) and infrared (IR) spectral range. Moreover, heterostructures with different materials (e.g., other TMDs, Graphene) can speed up the response of the photodetectors through the creation of built-in electric fields and the faster transport of charge carriers. Finally, in order to enhance the stability of the devices, perovskites have been exploited both as passivation layers and as electron reservoirs. Full article

This perspective presents an overview of approaches to the preparation of molecular recognition agents for chemical sensing. These approaches include chemical synthesis, using catalysts from biological systems, partitioning, aptamers, antibodies and molecularly imprinted polymers. The latter three approaches are general in that they can be applied with a large number of analytes, both proteins and smaller molecules like drugs and hormones. Aptamers and antibodies bind analytes rapidly while molecularly imprinted polymers bind much more slowly. Most molecularly imprinted polymers, formed by polymerizing in the presence of a template, contain a high level of covalent crosslinker that causes the polymer to form a separate phase. This results in a material that is rigid with low affinity for analyte and slow binding kinetics. Our approach to templating is to use predominantly or exclusively noncovalent crosslinks. This results in soluble templated polymers that bind analyte rapidly with high affinity. The biggest challenge of this approach is that the chains are tangled when the templated polymer is dissolved in water, blocking access to binding sites. Full article

Maritime traffic and fishing activities have accelerated considerably over the last decade, with a consequent impact on the environment and marine resources. Meanwhile, a growing number of ship-reporting technologies and remote-sensing systems are generating an overwhelming amount of spatio-temporal and geographically distributed data related to large-scale vessels and their movements. Individual technologies have distinct limitations but, when combined, can provide a better view of what is happening at sea, lead to effectively monitor fishing activities, and help tackle the investigations of suspicious behaviors in close proximity of managed areas. The paper integrates non-cooperative Synthetic Aperture Radar (SAR) Sentinel-1 images and cooperative Automatic Identification System (AIS) data, by proposing two types of associations: (i) point-to-point and (ii) point-to-line. They allow the fusion of ship positions and highlight “suspicious” AIS data gaps in close proximity of managed areas that can be further investigated only once the vessel—and the gear it adopts—is known. This is addressed by a machine-learning approach based on the Fast Fourier Transform that classifies single sea trips. The approach is tested on a case study in the central Adriatic Sea, automatically reporting AIS-SAR associations and seeking ships that are not broadcasting their positions (intentionally or not). Results allow the discrimination of collaborative and non-collaborative ships, playing a key role in detecting potential suspect behaviors especially in close proximity of managed areas. Full article

In this article, we address the problem of prolonging the battery life of Internet of Things (IoT) nodes by introducing a smart energy harvesting framework for IoT networks supported by femtocell access points (FAPs) based on the principles of Contract Theory and Reinforcement Learning. Initially, the IoT nodes’ social and physical characteristics are identified and captured through the concept of IoT node types. Then, Contract Theory is adopted to capture the interactions among the FAPs, who provide personalized rewards, i.e., charging power, to the IoT nodes to incentivize them to invest their effort, i.e., transmission power, to report their data to the FAPs. The IoT nodes’ and FAPs’ contract-theoretic utility functions are formulated, following the network economic concept of the involved entities’ personalized profit. A contract-theoretic optimization problem is introduced to determine the optimal personalized contracts among each IoT node connected to a FAP, i.e., a pair of transmission and charging power, aiming to jointly guarantee the optimal satisfaction of all the involved entities in the examined IoT system. An artificial intelligent framework based on reinforcement learning is introduced to support the IoT nodes’ autonomous association to the most beneficial FAP in terms of long-term gained rewards. Finally, a detailed simulation and comparative results are presented to show the pure operation performance of the proposed framework, as well as its drawbacks and benefits, compared to other approaches. Our findings show that the personalized contracts offered to the IoT nodes outperform by a factor of four compared to an agnostic type approach in terms of the achieved IoT system’s social welfare. Full article

In a conventional Unmanned aerial vehicles (UAV) navigational system Global Navigation Satellite System (GNSS) sensor is often a main source of data for trajectory generation. Even video tracking based systems need some GNSS data for proper work. The goal of this study is to develop an optics-based system to estimate the ground speed of the UAV in the case of the GNSS failure, jamming, or unavailability. The proposed approach uses a camera mounted on the fuselage belly of the UAV. We can obtain the ground speed of the airplane by using the digital cropping, the stabilization of the real time image, and template matching algorithms. By combining the ground speed vector components with measurements of airspeed and altitude, the wind velocity and drift are computed. The obtained data were used to improve efficiency of the video-tracking based on a navigational system. An algorithm allows this computation to be performed in real time on board of a UAV. The algorithm was tested in Software-in-the-loop and implemented on the UAV hardware. Its effectiveness has been demonstrated through the experimental test results. The presented work could be useful for upgrading the existing MUAV products (with embedded cameras) already delivered to the customers only by updating their software. It is especially significant in the case when any necessary hardware upgrades would be economically unjustified or even impossible to be carried out. Full article

Even though individual mental and health status largely affects the safety in industrial sites, most studies for preventing industrial accidents are mainly focused on external factors such as regulations, education, etc. In this study, the effect of individual factors on safety (i.e., safety consciousness and safety commitment) was analyzed by collecting brainwave and pulse data at construction sites where industrial accidents have occurred with the highest percentage. The effects of brain stress, concentration, brain activity, and left and right brain imbalance on safety accidents were evaluated through brain wave measurements. In addition, the effects of cumulative fatigue, physical vitality, autonomic nerve health, and autonomic balance were identified through pulse wave measurements. Data were acquired for 180 construction workers at various construction sites, and the workers were classified into three grades according to factors that affected safety accidents at construction sites. Then, the safety consciousness and safety commitment levels of workers corresponding to each grade of the influence factors were evaluated by conducting a questionnaire on safety consciousness and safety commitment. As a result, the characteristics of brain and pulse waves required to improve safety consciousness and safety commitment ability of workers at construction sites were explored. Full article

In recent years, there has been a noticeable increase in interest in the possibilities of embedding additional data in the constellation of an already existing information signal in radio technology. This solution more precisely is based on adding a low power signal (or signals) to a stronger signal (cover). As will be described in the article, this technique is used in numerous radio communication areas, such as watermarking, covert channel creation, and multiple access techniques. Typically, those areas are considered as independent research topics. Our comparison suggests that these areas are closely related. In this article, a comprehensive survey of the implementation of signal superposition is conducted with an emphasis on the similarities and differences between individual solutions. Since the nature of the signal model entails certain problems in the security area, we provide the reader with a review of the state-of-the-art research on this topic, including the PLS (physical layer security) and LPD (low probability of detection) issues. Full article

In this paper, we investigate the performance of a vehicular visible light communications (VVLC) link with a non-collimated and incoherent light source (a light-emitting diode) as the transmitter (Tx), and two different optical receiver (Rx) types (a camera and photodiode (PD)) under atmospheric turbulence (AT) conditions with aperture averaging (AA). First, we present simulation results indicating performance improvements in the signal-to-noise ratio (SNR) under AT with AA with increasing size of the optical concentrator. Experimental investigations demonstrate the potency of AA in mitigating the induced signal fading due to the weak to moderate AT regimes in a VVLC system. The experimental results obtained with AA show that the link’s performance was stable in terms of the average SNR and the peak SNR for the PD and camera-based Rx links, respectively with <1 dB SNR penalty for both Rxs, as the strength of AT increases compared with the link with no AT. Full article

Motor learning is associated with functional brain plasticity, involving specific functional connectivity changes in the neural networks. However, the degree of learning new motor skills varies among individuals, which is mainly due to the between-subject variability in brain structure and function captured by electroencephalographic (EEG) recordings. Here, we propose a kernel-based functional connectivity measure to deal with inter/intra-subject variability in motor-related tasks. To this end, from spatio-temporal-frequency patterns, we extract the functional connectivity between EEG channels through their Gaussian kernel cross-spectral distribution. Further, we optimize the spectral combination weights within a sparse-based ${\ell }_2$-norm feature selection framework matching the motor-related labels that perform the dimensionality reduction of the extracted connectivity features. From the validation results in three databases with motor imagery and motor execution tasks, we conclude that the single-trial Gaussian functional connectivity measure provides very competitive classifier performance values, being less affected by feature extraction parameters, like the sliding time window, and avoiding the use of prior linear spatial filtering. We also provide interpretability for the clustered functional connectivity patterns and hypothesize that the proposed kernel-based metric is promising for evaluating motor skills. Full article

This paper presents a new technique for the extraction of high-order wave-damage interaction coefficients (WDIC) through modal decomposition. The frequency and direction dependent complex-valued WDIC are used to model the scattering and mode conversion phenomena of guided wave interaction with damage. These coefficients are extracted from the harmonic analysis of local finite element model (FEM) mesh with non-reflective boundaries (NRB) and they are capable of describing the amplitude and phase of the scattered waves as a function of frequency and direction. To extract the WDIC of each wave mode, all the possible propagating wave modes are considered to be scattered simultaneously from the damage and propagate independently. Formulated in frequency domain, the proposed method is highly efficient, providing an overdetermined equation system for the calculation of mode participation factors, i.e., WDIC of each mode. Case studies in a 6-mm aluminum plate were carried out to validate the WDIC of: (1) a through-thickness hole and (2) a sub-surface crack. At higher frequency, scattered waves of high-order modes will appear and their WDIC can be successfully extracted through the modal decomposition. Full article

Damage classification is an important topic in the development of structural health monitoring systems. When applied to wind-turbine foundations, it provides information about the state of the structure, helps in maintenance, and prevents catastrophic failures. A data-driven pattern-recognition methodology for structural damage classification was developed in this study. The proposed methodology involves several stages: (1) data acquisition, (2) data arrangement, (3) data normalization through the mean-centered unitary group-scaling method, (4) linear feature extraction, (5) classification using the extreme gradient boosting machine learning classifier, and (6) validation applying a 5-fold cross-validation technique. The linear feature extraction capabilities of principal component analysis are employed; the original data of 58,008 features is reduced to only 21 features. The methodology is validated with an experimental test performed in a small-scale wind-turbine foundation structure that simulates the perturbation effects caused by wind and marine waves by applying an unknown white noise signal excitation to the structure. A vibration-response methodology is selected for collecting accelerometer data from both the healthy structure and the structure subjected to four different damage scenarios. The datasets are satisfactorily classified, with performance measures over $99.9\%$ after using the proposed damage classification methodology. Full article

In this work, four sensor fusion algorithms for inertial measurement unit data to determine the orientation of a device are assessed regarding their usability in a hardware restricted environment such as body-worn sensor nodes. The assessment is done for both the functional and the extra-functional properties in the context of human operated devices. The four algorithms are implemented in three data formats: 32-bit floating-point, 32-bit fixed-point and 16-bit fixed-point and compared regarding code size, computational effort, and fusion quality. Code size and computational effort are evaluated on an ARM Cortex M0+. For the assessment of the functional properties, the sensor fusion output is compared to a camera generated reference and analyzed in an extensive statistical analysis to determine how data format, algorithm, and human interaction influence the quality of the sensor fusion. Our experiments show that using fixed-point arithmetic can significantly decrease the computational complexity while still maintaining a high fusion quality and all four algorithms are applicable for applications with human interaction. Full article

Face-routing is one of the reliable recovery schemes when geographic routing fails to transmit data packets. Although studies on face-routing can overcome the failure of the data transmission, they lead to much energy consumption due to frequent data transmissions between adjacent nodes for carrying out the rule of face-routing. To avoid the frequent data transmissions, several face-routing schemes have been recently proposed to transmit data packets to the farthest-neighbor node. However, they happen with many data retransmissions because the farthest-neighbor node has a relatively low transmission success ratio. To solve this problem, we propose a new face-routing scheme that determines the most appropriate neighbor node to balance the trade-off between energy efficiency and transmission reliability with two viewpoints. The first viewpoint focuses on how to increase the distance progress of the data delivery in one-hop range to enhance energy efficiency. After that, the second viewpoint focuses on how to increase the success ratio of the data delivery to enhance the transmission reliability. As a result of the simulation, it was confirmed that the proposed method achieves better performance in terms of energy efficiency than existing face-routing research, and it is better than recent face-routing research in terms of reliability and retransmission. Full article