Global Navigation Satellite Systems (GNSS) receivers are an essential sensor for modern global positioning, navigation, and precise timing applications. High precision, i.e., geodetic GNSS positioning applications are based on carrier-phase measurements, where understanding the signal electrical reception characteristics, i.e., receiver antenna phase center corrections (PCCs), is critical. With the main goal of determining the PCC models of GNSS receiver antennas, only a few antenna calibration systems are in operation or under development worldwide. The International GNSS Service (IGS) publishes type-mean PCC models for almost all geodetic-grade GNSS antennas. However, the type-mean models are not perfect and do not fully reflect the signal reception properties of individual GNSS receiver antennas. Published relevant scientific research has shown that the application of individual PCC models significantly improves the accuracy of GNSS positioning. In this paper, the automated GNSS receiver antenna calibration system, recently developed at the Faculty of Geodesy of the University of Zagreb in Croatia, is shortly presented. The developed system is an absolute field calibration system based on the utilization of a Mitsubishi MELFA RV-4FML-Q 6-axis industrial robot. During calibration, the robot performs precise antenna under test (AUT) rotations and tilting. The antenna PCC modelling is based on time-differenced double-difference carrier-phase observations and spherical harmonics (SH) expansion. Our early antenna calibration results, for the Global Positioning System (GPS) L1 frequency, show a sub-millimeter repeatability of the estimated PCC model and a sub-millimeter agreement with the IGS approved Geo++ GmbH values.

The 3D geometry and anisotropic properties of magnetic nanostructures has been found to have a direct impact on their magnetization properties due to spatial coordinates and larger surface areas, which sheds new light on next-generation materials for advanced applications in magnetic energy harvesting. Our work presents novel pathways for the synthesis and assembly of multifunctional anisotropic 3D nanomagnets with various shapes and sizes with key attention to their anisotropic morphologies. We investigated the excellent properties of these new anisotropic 3D nanomagnets for the design of magnetic actuator systems and nanogenerators by embedding the 3D nanomagnets in a piezoelectric polyvinylidene fluoride (PVDF) polymer matrix. The 3D nanomagnets-PDVF composites were found to exhibit the highly electroactive β-phase with enhanced piezoelectric sensitivity. Further, the 3D nanomagnets-PDVF thin films have outstanding magnetic responsiveness and actuation capacity ideal for the fabrication of magnetic nanogenerators. These types of materials have a great deal of potential to generate sustainable alternative energy sources through harvesting and conversion of ubiquitous and residual low-frequency environmental magnetic noise into usable electricity.

Deep Learning (DL) for monitoring slowly evolving degradation processes typically involves overcoming data drift, complexity, and unavailability issues resulting from dynamic and harsh conditions, and rarity of labeled failure patterns, respectively. While degradation patterns are mostly hidden in such complex data, observation-based DL lean towards producing uncertain predictions and/or overfit the model during training process. This problem is usually caused by the insignificance of certain data representations. Therefore, and particularly due to the sequential nature of data in such a degradation process, it is necessary to consider neighboring observations to judge the accuracy of its representation or improving it. In this context, instead of traditional observation-based learning philosophy, this paper presents data-driven sequential mapping, while health indices can also be represented as a vector of sequential data and not as a single regressor output changing the model’s architecture. Using a dataset generated from a mathematical model mimicking bearing degradation life cycles and responding to the aforementioned three main challenges, a comparative study is built on investigating observation-based and sequence-based learning paths. According to a well-defined visual and numerical evaluation criterion, a sequence-based methodology reflects a better understanding of data representations through parameter tuning reaching better approximation and generalization. Such results support the necessity to such learning mechanism, especially for sequential data, dealing with some sort of correlation, and degrade controversially.

The ranking and evaluation of surgeons surgical skills is an important factor in order to be able to appropriately assign patient cases according to the necessary level of surgeon competence, in addition to helping towards pinpointing the specific clinicians within the surgical cohort who require further developmental training. One of the more frequent means of surgical skills evaluation, has been seen to be via a qualitative assessment of a surgeons portfolio alongside other supporting pieces of information-a process of which is rather subjective.

The contribution presented as part of this paper, involves the use of a set of Delsys Trigno EMG wearable sensors which track and record the muscular activation patterns of a surgeon during a surgical procedure alongside computationally driven Artificial Intelligence(AI) methods towards the differentiation and ranking of the surgical skills of a clinician in a quantitative fashion. The participants for the research involved novice level surgeons, intermediate level surgeons and expert level surgeons in various simulated surgical cases. The results showed that the monitoring of a set of key anatomical muscles during the simulated surgical cases, can allow for an effective differentiation of a surgeons skillset based on an AI prediction.

Previous work has demonstrated the relevance of FT-IR investigation on gingival crevicular fluid (GCF) for monitoring orthodontic treatments [1]. Usually, FT-IR spectra are acquired in reflectance mode by dropping a few microliters of GCF on a reflecting support. A faster procedure for collecting GCF spectra can be obtained by exploiting a different acquisition geometry. The attenuated total reflection (ATR) approach allows the collection of good-quality infrared spectra from any solid or liquid sample with almost no sample preparation. The objective of this research is to validate the ATR approach for GCF investigation by comparing the spectra acquired using the GCF extracted by paper cones and examined using reflecting support compared to those collected employing the ATR accessory.

Patients aged between 13 and 21 years undergoing orthodontic treatment with fixed multibracket appliances were recruited. Two different paper supports were used: standard sterile absorbent paper cones and PerioPaper strips inserted 1 mm into the gingival crevice. GCF was extracted for the FT-IR measurements in reflection geometry, whereas PerioPaper supports were directly used for the ATR acquisition mode. A Perkin Elmer Spectrum One FT-IR spectrometer was used for FT-IR in specular-reflection mode and a Universal ATR accessory was used for the other type of measurements.

FT-IR spectra in the range of 4000 to 600 cm^-1 with 4 cm^-1 of the spectral resolution were obtained from GCF samples before starting the orthodontic treatments using the two different collection geometries. The characteristics of the spectra acquired in reflection and ATR mode were examined and the contribution of the different GCF components was clearly evidenced also in ATR mode spectra. This analysis confirmed that the ATR approach can allow a detailed biochemical characterization of GCF similar to the reflection acquisition mode. Using the ATR geometry, the measurement time is greatly shortened and there is no risk that the spectra can be affected by the extraction procedures. These findings are pivotal for future research in order to make the GCF analysis fast and easy for the monitoring of orthodontic tooth movement in complex cases.

[1] d’Apuzzo, F.; Nucci, L.; Delfino, I.; Portaccio, M.; Minervini, G.; Isola, G.; Serino, I.; Camerlingo, C.; Lepore, M. Application of vibrational spectroscopies in the qualitative analysis of gingival crevicular fluid and periodontal ligament during orthodontic tooth movement. J. Clin. Med. 2021, 10, 1405

Hydrogen (H₂) is the most common element on earth and is found mainly with oxygen in water and hydrocarbons. Hydrogen has applications in many industries. It is used in liquid to propel rockets, cryogenic research in superconducting studies, and oil refining. This paper aims to develop tapered optical fiber sensor coated with Graphene Oxide (GO) for H₂ sensing application. The optical fiber was fabricated by tapering a standard optical fiber via a cost-effective and simple tapering technique. The author has a varied parameter from waist diameter to 20 μm with a fixed length of 10 mm and an up / down taper of 5 mm. Micro-nano characterization techniques such as FESEM, EDX, AFM, and XRD were utilized to obtain detailed structural properties of these nanostructures and fundamentally understand their functions concerning optical sensor performance. The responses of the developed sensor toward H₂ were measured through the change in absorbance within the concentrations of 0.125% – 2.00% in synthetic air. The developed sensor indicated high sensitivity toward the change in H₂ concentration at 100 ^oC. The observed response and recovery time for 2.00% H₂ were calculated as 2 minutes and 11 minutes, respectively. The developed optical fiber sensor achieved high selectivity and excellent stability toward H₂ gas upon exposure to other gases such as NH₃ and CH₄.

One of the biggest cybersecurity challenges in recent years has been the risk that insiders pose. Internet consumers are susceptible to exploitation due to the exponential growth of network usage. Malware attacks are a major concern in the digital world. This occurrence indicates that threats necessitate specialized detection techniques and equipment, including the ability to facilitate accurate and rapid detection of an insider threat. In this research, we propose a machine learning algorithm using a neural network to enhance malware detection accuracy in response to this threat. A feature extraction, anomaly detection, and classification workflow is also proposed. We use the CERT4.2 dataset and preprocess the data by encoding text strings and differentiating threat and non-threat records. Our developed machine learning model incorporates multiple dense layers, ReLU activation functions, and dropout layers for regularization. The model attempts to detect and classify internal threats in the dataset with precision. We employed Random Forest, Naive Bayes, KNN, SVM, Decision Tree, Logical Regression, and the Gradient Boosting algorithm to compare our proposed model with other classification techniques. According to the results of the experiments, the proposed method functions properly and can detect malware more effectively and with 100% accuracy.

One of the core ideas of Industry 4.0 has been the use of Digital Twin Networks (DTN). DTN facilitates the co-evolution of real and virtual things through the use of DT modelling, interactions, computation, and information analysis systems. The DT simulates product lifecycles to forecast and optimizes manufacturing systems and component behaviour. Industry and Academia have been developing Digital Twin (DT) technology for real-time remote monitoring and control, transport risk assessment, and intelligent scheduling in the smart industry. This study aims to design and simulate a comprehensive digital twin model connecting three factories to a single server. It incorporates remote network control, IoT integration, advanced networking protocols, and security measures. The model utilizes the Open Shortest Path First (OSPF) routing protocol for seamless network connectivity within the interconnected factories. Authentication, authorization, and accounting (AAA) mechanisms ensure secure access and prevent unauthorized entry. The Digital Twin Model is simulated using Cisco Packet Tracer, validating its functionality in network connectivity, security, remote control, and motor efficiency monitoring. The results demonstrate the successful integration and operation of the model in smart industries. The networked factories exhibit improved operational efficiency, enhanced security, and proactive maintenance.

Safety related to pavement ageing is becoming a major issue, as cracks and holes in the road surface can lead to severe accidents. Although pavement maintenance is extremely costly, detecting a deterioration before its surface gets completely damaged remains a challenge. Current approaches still use wired sensors, which consume a lot of energy and are expensive; further than that, wired sensors may get damaged during installation. To avoid the use of cables, in this work a Zigbee-based wireless sensor network for pavement monitoring was developed and tested in the laboratory. The system consists of a slave sensor and a roadside unit: the slave sensor sends wireless acceleration data to the master, and the master saves the received acceleration data in a csv file. Further data or signal processing can be performed in the master based on this acceleration dataset. Two laboratory tests were carried out: the first one was to perform a dynamic calibration using a vibrating pot, and obtain the main characteristics of three micro electro-mechanical sensor (MEMS) accelerometers in terms of power consumption and sensitivity; the second one was to simulate the pavement response to a five-wheeled truck at different speeds by means of a vibrating table. Preliminary results showed that the Zigbee-based wireless sensor network of MEMS accelerometers is capable of capturing the required ranges of displacement/deformation, acceleration and frequency, ADXL354 becoming the best accelerometer with the lowest power consumption as small as 155 uA.

Machine intelligence models are effective at classifying datasets for data analytics and predicting insights that can be used to make clinical decisions. The models would aid in disease prognosis and preliminary disease investigation, both of which are necessary for effective treatment. In today's world, there is a high demand for the interpretability and explainability of decision models. XAI is an extension of artificial intelligence that extends ai's capability by explaining why it has made that prediction and whether we can rely on it or not. The goal of this paper is to predict heart disease prediction using the RHMIoT model in the XAI framework. The patient clinical data are gathered using medical IoT sensors and stored in a secure cloud storage using a lightweight block encryption and decryption approach. The model is used to predict the accuracy of heart disease and its severity level. The accuracy levels of cardiac disease are calculated using Deep Learning and auto-encoder-based methods. We present a novel strategy for identifying key features using machine learning and deep learning techniques in a secured cloud environment to improve the accuracy of CVD. A lightweight block encryption and decryption technique is provided for a secure RHMIoT. The outcomes were determined using several performance matrices. The performance of auto-encoder Kernel SVM model provided the greatest accuracy of 87.00%. The suggested RHMIoT system identifies the presence of heart disease in a patient and helps to get quick medical attention in case of an emergency situation.