Technologies, Volume 9, Issue 1 (March 2021) – 22 articles

In the last years, the Internet of Things (IoT) has emerged as a key application context in the design and evolution of technologies in the transition toward a 5G ecosystem. More and more IoT technologies have entered the market and represent important enablers in the deployment of networks of interconnected devices. As network and spatial device densities grow, energy efficiency and consumption are becoming an important aspect in analyzing the performance and suitability of different technologies. In this framework, this survey presents an extensive review of IoT technologies, including both Low-Power Short-Area Networks (LPSANs) and Low-Power Wide-Area Networks (LPWANs), from the perspective of energy efficiency and power consumption. Existing consumption models and energy efficiency mechanisms are categorized, analyzed and discussed, in order to highlight the main trends proposed in literature and standards toward achieving energy-efficient IoT networks. Current limitations and open challenges are also discussed, aiming at highlighting new possible research directions. Full article

Currently, 3D-printed aluminium alloy fabrications made by selective laser melting (SLM) offer a promising route for the production of small series of custom-designed support brackets and heat exchangers with complex geometry and shape and miniature size. Alloy composition and printing parameters need to be optimised to mitigate fabrication defects (pores and microcracks) and enhance the parts’ performance. The deformation response needs to be studied with adequate characterisation techniques at relevant dimensional scale, capturing the peculiarities of micro-mechanical behaviour relevant to the particular article and specimen dimensions. Purposefully designed Al-Si-Mg 3D-printable RS-333 alloy was investigated with a number of microscopy techniques, including in situ mechanical testing with a Deben Microtest 1-kN stage integrated and synchronised with Tescan Vega3 SEM to acquire high-resolution image datasets for digital image correlation (DIC) analysis. Dog bone specimens were 3D-printed in different orientations of gauge zone cross-section with respect to the fast laser beam scanning and growth directions. This corresponded to the varying local conditions of metal solidification and cooling. Specimens showed variation in mechanical properties, namely Young’s modulus (65–78 GPa), yield stress (80–150 MPa), ultimate tensile strength (115–225 MPa) and elongation at break (0.75–1.4%). Furthermore, the failure localisation and character were altered with the change in gauge cross-section orientation. DIC analysis allowed correct strain evaluation that overcame the load frame compliance effect and helped to identify the unevenness of deformation distribution (plasticity waves), which ultimately resulted in exceptionally high strain localisation near the ultimate failure crack position. Full article

CNNs (Convolutional Neural Networks) are becoming increasingly important for real-time applications, such as image classification in traffic control, visual surveillance, and smart manufacturing. It is challenging, however, to meet timing constraints of image processing tasks using CNNs due to their complexity. Performing dynamic trade-offs between the inference accuracy and time for image data analysis in CNNs is challenging too, since we observe that more complex CNNs that take longer to run even lead to lower accuracy in many cases by evaluating hundreds of CNN models in terms of time and accuracy using two popular data sets, MNIST and CIFAR-10. To address these challenges, we propose a new approach that (1) generates CNN models and analyzes their average inference time and accuracy for image classification, (2) stores a small subset of the CNNs with monotonic time and accuracy relationships offline, and (3) efficiently selects an effective CNN expected to support the highest possible accuracy among the stored CNNs subject to the remaining time to the deadline at run time. In our extensive evaluation, we verify that the CNNs derived by our approach are more flexible and cost-efficient than two baseline approaches. We verify that our approach can effectively build a compact set of CNNs and efficiently support systematic time vs. accuracy trade-offs, if necessary, to meet the user-specified timing and accuracy requirements. Moreover, the overhead of our approach is little/acceptable in terms of latency and memory consumption. Full article

The paper presents a new analytical model for thin structural adhesives in glued tube-to-tube butt joints. The aim of this work is to provide an interface condition that allows for a suitable replacement of the adhesive layer in numerical simulations. The proposed model is a nonlinear and rate-dependent imperfect interface law that is able to accurately describe brittle and ductile stress–strain behaviors of adhesive layers under combined tensile–torsion loads. A first comparison with experimental data that were available in the literature provided promising results in terms of the reproducibility of the stress–strain behavior for pure tensile and torsional loads (the relative errors were less than $6\%$) and in terms of failure strains for combined tensile–torsion loads (the relative errors were less than $14\%$). Two main novelties are highlighted: (i) Unlike the classic spring-like interface models, this model accounts for both stress and displacement jumps, so it is suitable for soft and hard adhesive layers; (ii) unlike classic cohesive zone models, which are phenomenological, this model explicitly accounts for material and damage properties of the adhesive layer. Full article

Hybrid models combine physics-based models and data-driven models. This combination is a useful technique to detect fault and predict the current degradation of equipment. This paper proposes a physics-based model, which will be part of a hybrid model, for a heating, ventilation, and air conditioning system installed in the passenger vehicle of a train. The physics-based model is divided into four main parts: heating subsystems, cooling subsystems, ventilation subsystems, and cabin thermal networking subsystems. These subsystems are developed when considering the sensors that are located in the real system, so the model can be linked via the acquired sensor data and virtual sensor data to improve the detectability of failure modes. Thus, the physics-based model can be synchronized with the real system to provide better simulation results. The paper also considers diagnostics and prognostics performance. First, it looks at the current situation of the maintenance strategy for the heating, ventilation, air conditioning system, and the number of failure modes that the maintenance team can detect. Second, it determines the expected improvement using hybrid modelling to maintain the system. This improvement is based on the capabilities of detecting new failure modes. The paper concludes by suggesting the future capabilities of hybrid models. Full article

This work proposes new interface conditions between the layers of a three-dimensional composite structure in the framework of coupled thermoelasticity. More precisely, the mechanical behavior of two linear isotropic thermoelastic solids, bonded together by a thin layer, constituted of a linear isotropic thermoelastic material, is studied by means of an asymptotic analysis. After defining a small parameter $\epsilon$, which tends to zero, associated with the thickness and constitutive coefficients of the intermediate layer, two different limit models and their associated limit problems, the so-called soft and hard thermoelastic interface models, are characterized. The asymptotic expansion method is reviewed by taking into account the effect of higher-order terms and defining a generalized thermoelastic interface law which comprises the above aforementioned models, as presented previously. A numerical example is presented to show the efficiency of the proposed methodology, based on a finite element approach developed previously. Full article

Any severe motor disability is a condition that limits the ability to interact with the environment, even the domestic one, caused by the loss of control over one’s mobility. This work presents RoboEYE, a power wheelchair designed to allow users to move easily and autonomously within their homes. To achieve this goal, an innovative, cost-effective and user-friendly control system was designed, in which a non-invasive eye tracker, a monitor, and a 3D camera represent some of the core elements. RoboEYE integrates functionalities from the mobile robotics field into a standard power wheelchair, with the main advantage of providing the user with two driving options and comfortable navigation. The most intuitive and direct modality foresees the continuous control of frontal and angular wheelchair velocities by gazing at different areas of the monitor. The second, semi-autonomous modality allows navigation toward a selected point in the environment by just pointing and activating the wished destination while the system autonomously plans and follows the trajectory that brings the wheelchair to that point. The purpose of this work was to develop the control structure and driving interface designs of the aforementioned driving modalities taking into account also uncertainties in gaze detection and other sources of uncertainty related to the components to ensure user safety. Furthermore, the driving modalities, in particular the semi-autonomous one, were modeled and qualified through numerical simulations and experimental verification by testing volunteers, who are regular users of standard electric wheelchairs, to verify the efficiency, reliability and safety of the proposed system for domestic use. RoboEYE resulted suitable for environments with narrow passages wider than 1 m, which is comparable with a standard domestic door and due to its properties with large commercialization potential. Full article

In this paper, a chaotic three dimansional dynamical system is proposed, that is a modification of the system in Volos et al. (2017). The new system has two hyperbolic sine nonlinear terms, as opposed to the original system that only included one, in order to optimize system’s chaotic behavior, which is confirmed by the calculation of the maximal Lyapunov exponents and Kaplan-Yorke dimension. The system is experimentally realized, using Bi-color LEDs to emulate the hyperbolic sine functions. An extended dynamical analysis is then performed, by computing numerically the system’s bifurcation and continuation diagrams, Lyapunov exponents and phase portraits, and comparing the numerical simulations with the circuit simulations. A series of interesting phenomena are unmasked, like period doubling route to chaos, coexisting attractors and antimonotonicity, which are all verified from the circuit realization of the system. Hence, the circuit setup accurately emulates the chaotic dynamics of the proposed system. Full article

Software-Defined Networking (SDN) is a new paradigm that revolutionizes the idea of a software-driven network through the separation of control and data planes. It addresses the problems of traditional network architecture. Nevertheless, this brilliant architecture is exposed to several security threats, e.g., the distributed denial of service (DDoS) attack, which is hard to contain in such software-based networks. The concept of a centralized controller in SDN makes it a single point of attack as well as a single point of failure. In this paper, deep learning-based models, long-short term memory (LSTM) and convolutional neural network (CNN), are investigated. It illustrates their possibility and efficiency in being used in detecting and mitigating DDoS attack. The paper focuses on TCP, UDP, and ICMP flood attacks that target the controller. The performance of the models was evaluated based on the accuracy, recall, and true negative rate. We compared the performance of the deep learning models with classical machine learning models. We further provide details on the time taken to detect and mitigate the attack. Our results show that RNN LSTM is a viable deep learning algorithm that can be applied in the detection and mitigation of DDoS in the SDN controller. Our proposed model produced an accuracy of 89.63%, which outperformed linear-based models such as SVM (86.85%) and Naive Bayes (82.61%). Although KNN, which is a linear-based model, outperformed our proposed model (achieving an accuracy of 99.4%), our proposed model provides a good trade-off between precision and recall, which makes it suitable for DDoS classification. In addition, it was realized that the split ratio of the training and testing datasets can give different results in the performance of a deep learning algorithm used in a specific work. The model achieved the best performance when a split of 70/30 was used in comparison to 80/20 and 60/40 split ratios. Full article

Availability, sustainability, and functionality pose particular challenges for the value chain of individual mobility. Therefore, in this work, the typical value chain is broken up and the traditional approaches are extended with an open-source perspective. The focus is on modularity and simplicity in order to enable the broadest possible applicability and fast implementation of local production. Nevertheless, the mobility concept should meet the current standards, especially with regard to safety. A modular vehicle frame is presented as a basis, which meets all regulations, can be built in self-assembly, and is available as open-source. Full article

The development of micromechanical models to predict the effective properties of multiphase composites is important for the design and optimization of new materials, as well as to improve our understanding about the structure–properties relationship. In this work, the two-scale asymptotic homogenization method (AHM) is implemented to calculate the out-of-plane effective complex-value properties of periodic three-phase elastic fiber-reinforced composites (FRCs) with parallelogram unit cells. Matrix and inclusions materials have complex-valued properties. Closed analytical expressions for the local problems and the out-of-plane shear effective coefficients are given. The solution of the homogenized local problems is found using potential theory. Numerical results are reported and comparisons with data reported in the literature are shown. Good agreements are obtained. In addition, the effects of fiber volume fractions and spatial fiber distribution on the complex effective elastic properties are analyzed. An analysis of the shear effective properties enhancement is also studied for three-phase FRCs. Full article

Weakly labeled data are inevitable in various research areas in artificial intelligence (AI) where one has a modicum of knowledge about the complete dataset. One of the reasons for weakly labeled data in AI is insufficient accurately labeled data. Strict privacy control or accidental loss may also cause missing-data problems. However, supervised machine learning (ML) requires accurately labeled data in order to successfully solve a problem. Data labeling is difficult and time-consuming as it requires manual work, perfect results, and sometimes human experts to be involved (e.g., medical labeled data). In contrast, unlabeled data are inexpensive and easily available. Due to there not being enough labeled training data, researchers sometimes only obtain one or few data points per category or label. Training a supervised ML model from the small set of labeled data is a challenging task. The objective of this research is to recover missing labels from the dataset using state-of-the-art ML techniques using a semisupervised ML approach. In this work, a novel convolutional neural network-based framework is trained with a few instances of a class to perform metric learning. The dataset is then converted into a graph signal, which is recovered using a recover algorithm (RA) in graph Fourier transform. The proposed approach was evaluated on a Fashion dataset for accuracy and precision and performed significantly better than graph neural networks and other state-of-the-art methods. Full article

Vehicles on the road are rising in extensive numbers, particularly in proportion to the industrial revolution and growing economy. The significant use of vehicles has increased the probability of traffic rules violation, causing unexpected accidents, and triggering traffic crimes. In order to overcome these problems, an intelligent traffic monitoring system is required. The intelligent system can play a vital role in traffic control through the number plate detection of the vehicles. In this research work, a system is developed for detecting and recognizing of vehicle number plates using a convolutional neural network (CNN), a deep learning technique. This system comprises of two parts: number plate detection and number plate recognition. In the detection part, a vehicle’s image is captured through a digital camera. Then the system segments the number plate region from the image frame. After extracting the number plate region, a super resolution method is applied to convert the low-resolution image into a high-resolution image. The super resolution technique is used with the convolutional layer of CNN to reconstruct the pixel quality of the input image. Each character of the number plate is segmented using a bounding box method. In the recognition part, features are extracted and classified using the CNN technique. The novelty of this research is the development of an intelligent system employing CNN to recognize number plates, which have less resolution, and are written in the Bengali language. Full article

In recent years, with the current advancements in Robotics and Artificial Intelligence (AI), robots have the potential to support the field of healthcare. Robotic systems are often introduced in the care of the elderly, children, and persons with disabilities, in hospitals, in rehabilitation and walking assistance, and other healthcare situations. In this survey paper, the recent advances in robotic technology applied in the healthcare domain are discussed. The paper provides detailed information about state-of-the-art research in care, hospital, assistive, rehabilitation, and walking assisting robots. The paper also discusses the open challenges healthcare robots face to be integrated into our society. Full article

In this article, a method for the synthesis of catalysts for methanol electrooxidation based on additive manufacturing and electroless metal deposition is presented. The research work was divided into two parts. Firstly, coatings were obtained on a flat substrate made of light-hardening resin dedicated to 3D printing. Copper was deposited by catalytic metallization. Then, the deposited Cu coatings were modified by palladium through a galvanic displacement process. The catalytic properties of the obtained coatings were analyzed in a solution of 0.1 M NaOH and 1 M methanol. The influence of the deposition time of copper and palladium on the catalytic properties of the coatings was investigated. Based on these results, the optimal parameters for the deposition were determined. In the second part of the research work, 3D prints with a large specific surface were metallized. The elements were covered with a copper layer and modified by palladium, then chronoamperometric curves were determined. The application of the proposed method could allow for the production of elements with good catalytic properties, complex geometry with a large specific surface area, small volume and low weight. Full article

Copolymer solution of vinylidene fluoride with tetrafluoroethylene (VDF-TeFE) was used for electrospinning of fluoropolymer scaffolds. Magnetron co-sputtering of titanium and copper targets in the argon atmosphere was used for VDF-TeFE scaffolds modification. Scanning electron microscopy (SEM) showed that scaffolds have a nonwoven structure with mean fiber diameter 0.77 ± 0.40 μm, mean porosity 58 ± 7%. The wetting angle of the original (unmodified) hydrophobic fluoropolymer scaffold after modification by titanium begins to possess hydrophilic properties. VDF-TeFE scaffold modification by titanium/copper leads to the appearance of strong antibacterial properties. The obtained fluoropolymer samples can be successfully used in tissue engineering. Full article

In the present work, the radiation properties of a graphene plasmonic patch antenna are investigated and enhanced in terms of efficiency, utilizing circuit-matching techniques. Initially, the reflection coefficient of graphene surface waves due to discontinuities is studied, while the power flow towards free-space is numerically extracted. This analysis indicates that the radiated power is increased for higher values of the chemical potential, although the surface wave is weakly confined and prone to degradation due to interference. For this reason, a graphene sheet that supports strongly confined surface waves is terminated via a matching layer, in order to enhance the radiating power. In particular, the matching layer consists of an appropriately selected larger chemical potential value, in order to minimize the reflection coefficient and boost the radiation performance. The numerical investigation of this setup validates the upgraded performance, since the radiating power is significantly increased. Then, a realistic setup that includes a graphene patch antenna is examined numerically, proving the augmentation of the radiation efficiency when the matching layer is utilized. Finally, the latter is designed with a graded increment in the chemical potential, and the computational analysis highlights the significant enhancement of the graphene plasmonic antenna gain towards the desired direction. Consequently, a more reliable framework for wireless communications between plasmonic devices at THz frequencies is established, which corresponds to the practical significance of the proposed methodology for improved radiation efficiency. All numerical results are extracted by means of an efficient modification of the Finite-Difference Time-Domain (FDTD) scheme, which models graphene accurately. Full article

Bug prediction aims at finding source code elements in a software system that are likely to contain defects. Being aware of the most error-prone parts of the program, one can efficiently allocate the limited amount of testing and code review resources. Therefore, bug prediction can support software maintenance and evolution to a great extent. In this paper, we propose a function level JavaScript bug prediction model based on static source code metrics with the addition of a hybrid (static and dynamic) code analysis based metric of the number of incoming and outgoing function calls (HNII and HNOI). Our motivation for this is that JavaScript is a highly dynamic scripting language for which static code analysis might be very imprecise; therefore, using a purely static source code features for bug prediction might not be enough. Based on a study where we extracted 824 buggy and 1943 non-buggy functions from the publicly available BugsJS dataset for the ESLint JavaScript project, we can confirm the positive impact of hybrid code metrics on the prediction performance of the ML models. Depending on the ML algorithm, applied hyper-parameters, and target measures we consider, hybrid invocation metrics bring a 2–10% increase in model performances (i.e., precision, recall, F-measure). Interestingly, replacing static NOI and NII metrics with their hybrid counterparts HNOI and HNII in itself improves model performances; however, using them all together yields the best results. Full article

Self-supervised learning has gained popularity because of its ability to avoid the cost of annotating large-scale datasets. It is capable of adopting self-defined pseudolabels as supervision and use the learned representations for several downstream tasks. Specifically, contrastive learning has recently become a dominant component in self-supervised learning for computer vision, natural language processing (NLP), and other domains. It aims at embedding augmented versions of the same sample close to each other while trying to push away embeddings from different samples. This paper provides an extensive review of self-supervised methods that follow the contrastive approach. The work explains commonly used pretext tasks in a contrastive learning setup, followed by different architectures that have been proposed so far. Next, we present a performance comparison of different methods for multiple downstream tasks such as image classification, object detection, and action recognition. Finally, we conclude with the limitations of the current methods and the need for further techniques and future directions to make meaningful progress. Full article

This work presents a unified approach to the analysis of contact problems with various interface laws that model the processes involved in contact between a deformable body and a rigid or reactive foundation. These laws are then used in the formulation of a general static frictional contact problem with unilateral constraints for elastic materials, which is governed by three parameters. A weak formulation of the problem is derived, which is in the form of an elliptic variational inequality, and the Tykhonov well-posedness of the problem is established, under appropriate assumptions on the data and parameters, with respect to a special Tykhonov triple. The proof is based on arguments on coercivity, compactness, and lower-semicontinuity. This abstract result leads to different convergence results, which establish the continuous dependence of the weak solution on the data and the parameters. Moreover, these results elucidate the links among the weak solutions of the different models. Finally, the corresponding mechanical interpretations of the conditions and the results are provided. The novelty in this work is the application of the Tykhonov well-posedness concept, which allows a unified and elegant framework for this class of static contact problems. Full article