Search

In this paper, a profitable strategy of utilizing solar and bio-energy available in a rural location for meeting the energy deficit is discussed. The proposed model can provide 24-hour continual power supply to the vicinity without the need for any battery. The land requirement is minimal in the proposed model making it cost-effective for implementation. The paper provides a step by step procedure to optimally size the solar and bio-power plants and schedule their operation for maximum utility. A programmable smart meter connected with a central system having a user-friendly control algorithm has also been proposed at every home for end-user profitability. The smart meter will indicate energy consumption in local home, its solar energy generation, vicinity power demand, current tariff of power usage, energy imported/exported and other required details. The proposed model is discussed in detail for comprising of two adjacent villages named Gaddikheri and Tajamajra in Rohtak (Haryana), North India (having 768 houses). The technical and financial analysis at this location is done using RETScreen expert software and the results are presented. The system is found to provide 24-hr power supply and also be profitable with the capital cost being recovered in less than 8 years time. Apart from this, the proposed system also provides other benefits like reduction in greenhouse gas emission (by 93%), auto power theft control, bio-fertilizers, and a clean environment.

With the development of mature communication network technologies such as 5G, edge computing and slicing, business applications in areas such as Telematics and intelligent transportation also need the support of vehicle-road collaboration technologies. Due to the mobility of vehicles and the heterogeneity of communication interfaces, it brings many challenges for data transmission and processing of Telematics. Therefore, network resources, roadside unit devices and sunken edge nodes need to work together to guarantee real-time network communication. Thus, they can jointly help improve the quality of network services and user experience, and show the level of vehicle-road collaboration wisdom in terms of safety, efficiency and service. In view of the future 5G network-related technologies and C-V2X development trend, this paper provides a comprehensive technical solution framework for Telematics based on end-edge collaboration, combining network resource deployment optimization and edge caching technologies for video transmission tasks in Telematics application scenarios. By adding 5G base station type RSU and UAV mobile base station to work jointly, network resources are better allocated for scalable coded video, and a relatively better edge caching strategy is obtained based on user mobility analysis. Aiming to improve the network system performance, the vehicle-side, road-side and edge-side are jointly considered to improve the utilization efficiency of limited network resources. In this paper, we hope to provide a practical and effective service architecture for Telematics video application scenarios through a convergence scheme of network resource allocation and edge caching configuration.

Pulswidth modulation (PWM) methods were developed to grant the voltage source inverter (VSI) output signal better quality measures. The PWM method that VSI relies on must consider the utilization of the direct current (DC) source, especially when the drive system works in the overmodulation region. In this paper, different zero-sequence harmonic injection PWM methods are proposed. Two essential objectives are optimized by minimizing total harmonic distortion and maximizing the DC bus utilization. All schemes’ results are compared with each other and with some of the commonly used PWM schemes. This paper concludes the usage of third, ninth, and fifteenth harmonic injections and any combination of them to reach a new PWM method that overtakes the disadvantages of other methods. Meanwhile, it provides a handy tool to operate VSI optimally. Ultimately, this work concludes to what extent is the zero-sequence injection is helpful. The work is evaluated experimentally, and the practical results match the simulation outcomes.

After reviewing the current state of explainable Artificial Intelligence (XAI) capabilities in artificial Intelligence (AI) systems developed for critical domains like criminology, engineering, governance, health, law and psychology, this paper proposes a domain-independent Accountable explainable Artificial Intelligence (AXAI) capability framework. The proposed AXAI framework extends the XAI capability to let AI systems share their decisions and adequately explain the underlying reasoning processes. The idea is to help AI system developers overcome algorithmic biases and system limitations through incorporation of domain independent AXAI capabilities. Moreover, existing XAI methods would neither separate nor quantify measures of comprehensibility, accuracy and accountability so incorporating and assessing XAI capabilities remains difficult. Assessment of the AXAI capabilities of two AI systems in this paper demonstrates that the proposed AXAI framework facilitates separation and measurement of comprehensibility, predictive accuracy and accountability. The AXAI framework allows for the delineation of AI systems in a three-dimensional AXAI space. It measures comprehensibility as the readiness of a human to apply the acquired knowledge. The system accuracy is measured as the ratio of the test and training data, training data size and the observed number of false-positive inferences. Finally, the AXAI framework measures accountability in terms of the inspect ability of the input cues, the processed data and the output information, for addressing any legal and ethical issues.

Brain diseases are mainly caused by abnormal growth of brain cells that may damage the brain structure, and eventually will lead to malignant brain cancer. An early diagnosis to enable decisive treatment using a Computer-Aided Diagnosis (CAD) system has major challenges, especially accurate detection of different diseases in the magnetic resonance imaging (MRI) images. In this paper, a three-step preprocessing is proposed to enhance the quality of MRI images, along with a new Deep Convolutional Neural Network (DCNN) architecture for effective diagnosis of glioma, meningioma, and pituitary. The architecture uses batch normalization for fast training with a higher learning rate and ease initialization of the layer weights. The proposed architecture is a computationally lightweight model with a small number of convolutional, max-pooling layers and training iterations. A demonstrative comparison between the proposed architecture and other discussed models in this paper is conducted. An outstanding competitive accuracy is achieved of 98.22% overall, 99% in detecting glioma, 99.13% in detecting meningioma, 97.3% in detecting pituitary and 97.14% in detecting normal images when tested on a dataset with 3394 MRI images. Experimental results prove the robustness of the proposed architecture which has increased the detection accuracy of a variety of brain diseases in a short time.

The preparation of a quantum superposition is the key to the success of many quantum algorithms and quantum machine learning techniques. The preparation of an incomplete or a non-uniform quantum superposition with certain properties is a non-trivial task. In this paper, an $n$-qubits variational quantum circuit using partial negation and controlled partial negation operators will be proposed to prepare an arbitrary quantum superposition. The proposed quantum circuit follows the symmetries of the unitary Lie group. The speed of the preparation process and the accuracy of the prepared superposition has a special importance to the success of any quantum algorithm. The proposed method can be used to prepare the required quantum superposition in $\mathcal{O}(n)$ steps and with high accuracy when compared with relevant methods in literature., Comment: 15 pages. This paper was supported financially by the Academy of Scientific Research and Technology (ASRT), Egypt, under initiatives of Science Up Faculty of Science Grant o 6558). (ASRT) is the 2nd affiliation of this research

Brain-inspired architectures are gaining increased attention, especially for edge devices to perform cognitive tasks utilizing its limited energy budget and computing resources. Hyperdimensional computing (HDC) paradigm is an emerging framework inspired by an abstract representation of neuronal circuits’ attributes in the human brain. That includes a fully holographic random representation, high-dimension vectors representing data, and robustness to uncertainty. The basic HDC pipeline consists of an encoding, training and comparison stages. The encoding algorithm maps different representations of inputs into a single class and stores them in the associative memory (AM) throughout the training stage. Later, during the inference stage, the similarity is computed between the query vector, which is encoded using the same encoding model, and the stored classes in the AM. HDC has shown promising results for 1D applications using less power, and lower latency than state-of-the-art digital neural networks (DNN). While in 2D applications, convolutional neural network (CNN) still achieves higher classification accuracy at the expense of more computations. In this paper, a comprehensive study on the HDC paradigm, main algorithms, and its implementation is presented. Moreover, the main state-of-the-art HDC architectures for 1D and 2D applications are highlighted. The article also analyzes two competing paradigms, namely, HDC and CNN, in terms of accuracy, complexity, and the number of operations. The paper concluded by highlighting challenges and recommendations for future directions on the HDC framework.

Most conventional data association based multi-target tracking (MTT) algorithms typically suffer from intractable computational complexities and could not perform in an environment where large number of closely spaced multiple targets move across each other in clutters. Unlike to the existing MTT systems, the linear multi-target (LM) algorithm modifies the measurement detection followed by neighbored tracks as a clutter, hence, it updates the track without the influence of other tracks. Thus, LM technique is a computationally efficient algorithm that allows the multi-target system to play like a single target tracking algorithm. Smoothing maximizes the state estimation accuracy and reduces estimation error based on future scan measurement. However, only few research paper focused on the LM algorithm without utilizing the benefits of the smoothing. This paper presents Rauch-Tung-Striebel Smoothing in the linear multi-target based on integrated probabilistic data association (RTS-LMIPDA). The RTS-LMIPDA algorithm fuses forward and backward LM track predictions to obtain the smoothing prediction which is required to calculate the smoothing multi-target state estimates in the forward track. Numerical analysis is presented to illustrate the estimation accuracy and false track discrimination (FTD) performances of RTS-LMIPDA in comparison to the existing MTT algorithms using the simulations.

Lately, gamification (i.e., employing game-design elements and game principles in non-game contexts) has gained massive popularity and widespread usage in various areas, including education. However, gamification deployment in education in general and remote education in particular still faces many challenges that mainly influence user quality of experience. These challenges include lack of dedicated communication-based systems, potential additional load on teachers, absence of customization and personalization for users, and no support for advanced technology-enhanced learning (TEL). This paper investigates the use of gamification for networked delivery of science, technology, engineering and mathematics (STEM) subjects. It proposes an innovative gamification framework, the NEWTON-enhanced gamification model (N-EGM), which was designed as part of the European Horizon 2020 project NEWTON. The paper also describes the proposed N-EGM model deployment in the gamification engine of a real learning management system and its associated communication and networking solution. The communication support provides easy-to-use gamification configuration functionality and efficient data collection and processing in a heterogeneous technology context. Finally, the paper evaluates the proposed N-EGM model as part of a NEWTON project pilot deployed in a Romanian school. The results demonstrate the effectiveness of the proposed gamification solution in improving both students’ learning experience and their engagement, while also increasing student knowledge gain.

For the last few years, cases of applying articial intelligence (AI) to engineering activitiestowards sustainability have been reported. Life Cycle Engineering (LCE) provides a potential to systematicallyreach higher and productivity levels, owing to its holistic perspective and consideration of economic andenvironmental targets. To address the current gap to more systematic deployment of AI with LCE (AI-LCE)we have performed a systematic literature review emphasizing the three aspects:(1) the most prevalent AItechniques, (2) the current AI-improved LCE subelds and (3) the subelds with highly enhanced by AI.A specic set of inclusion and exclusion criteria were used to identify and select academic papers fromseveral elds, i.e. production, logistics, marketing and supply chain and after the selection process describedin the paper we ended up with 42 scientic papers. The study and analysis show that there are manyAI-LCE papers addressing Sustainable Development Goals mainly addressing: Industry, Innovation, andInfrastructure; Sustainable Cities and Communities; and Responsible Consumption and Production. Overall,the papers give a picture of diverse AI techniques used in LCE. Production design and Maintenance andRepair are the top explored LCE subelds whereas logistics and Procurement are the least explored subareas.Research in AI-LCE is concentrated in a few dominating countries and especially countries with a strongresearch funding and focus on Industry 4.0; Germany is standing out with numbers of publications. Thein-depth analysis of selected and relevant scientic papers are helpful in getting a more correct picture ofthe area which enables a more systematic approach to AI-LCE in the future. Funding: Adapt 2030 (Adaptive lifecycle design by applying digitalization and AI techniques to production) (Grant Number: 2019-05589)10.13039/501100001858-VINNOVA, Sweden’s Innovation Agency Adapt 2030

Smart cities often rely on technological innovations to improve citizens’ safety and quality of life. This paper presents a novel smart mobility system that facilitates people’s access to public mobility while preserving their privacy. In contrast to several well-known smart mobility systems discussed in this paper, the one we propose combines privacy guarantees with user friendliness. Specifically, the system is based on a zero-interaction approach whereby a person can use public transport services without any need to perform explicit actions. Operations related to ticket purchases and validation were fully automated. The system is also designed with the privacy-by-design paradigm to preserve user privacy as much as possible. Throughout the paper, several technical details are discussed as well to describe a prototype version of the system that was implemented. The prototype was successfully tested in the city of Imola (Emilia Romagna, Italy) to prove the validity of the system in the field.

Targets of cyber crime are not exclusive to the private sector. Successful cyber attacks on nation-states have proved that cyber threats can jeopardize significant national interests. In response, nation-states have begun to handle cyber threats at the national defense level, which is titled ‘cyber defense.’ The cyber defense sector is related to national security, therefore requires robust security technology. Contrary to normal systems, blockchain provides strong security properties without a centralized control entity, and as such its application in the cyber defense field is under the spotlight. In this paper, we present opportunities blockchain provides for cyber defense, research and national projects, and limitations. We constructed a survey of government documents, interviews, related news, technical reports, and research papers from 2016 to 2021. As a result, our research contributes to reducing the gap in blockchain for cyber defense by systematically conducting research and analysis. In our research, we found that not only research but also government-led plans are actively promoting blockchain, which demonstrates that blockchain will play a remarkable role in cyber defense. This paper concludes with suggestions for future research in aspects of the blockchain technology, evaluation, and survey.

This paper focuses on the problem that the current path planning algorithm is not mature enough to achieve the expected goal in a complex dynamic environment. In light of the ant colony optimization (ACO) with good robustness and strong search ability, and the dynamic window method (DWA) with better planning effect in local path planning problems, we propose a fusion algorithm named DACO that can quickly and safely reach the designated target area in a complex dynamic environment. This paper first improves the ant colony optimization, which greatly improves the convergence performance of the algorithm and shortens the global path length. On this basis, we propose a second-level safety distance determination rule to deal with the special problem of the research object encountering obstacles with unknown motion rules, in order to perfect the obstacle avoidance function of the fusion algorithm in complex environments. Finally, we carry out simulation experiments through MATLAB, and at the same time conduct three-dimensional simulation of algorithm functions again on the GAZEBO platform. It is verified that the algorithm proposed in this paper has good performance advantages in path planning and dynamic obstacle avoidance.

DevOps is a set of practices and a cultural movement that aims to break down barriers between development and operation teams to improve collaboration and communication. Different organizations have embraced DevOps principles due to the massive potential, such as a much shorter time to production, increased reliability and stability. However, despite the widespread adoption of DevOps and its infrastructure, there is a lack of understanding and literature on the key concepts, practices, tools, and challenges associated with implementing DevOps strategies. The main goal of this research paper is to explore and discuss challenges related to DevOps culture and practices. Moreover, it describes how DevOps works in an organization, provides a detailed explanation of DevOps, and investigates the cultural challenges that organizations face when implementing DevOps. The proposed paper reveals ten critical challenges that need to be addressed in adopting the DevOps culture. The challenges are further analyzed on the basis of the various continents. According to the findings, the following critical challenges are considered during the implementation of a DevOps culture: lack of collaboration and communication, Lack of skill and knowledge, complicated infrastructure, Lack of management, Lack of DevOps approach, and trust confidence problems.

Conventional theory for designing strictly nonblocking networks, such as crossbars or Clos networks, assumes that these networks have a centralized topology. Many new applications, however, require networks to have a distributed topology, like 2-D mesh or torus. In this paper, we present a new theoretical framework for designing such nonblocking networks. The framework is based on a linear programming formulation originally proposed for solving the hose-model traffic routing problem. The main difference, however, is that the link bandwidths in our problem must be discrete. This makes the problem much more challenging. We show how to apply the developed theorems to tackle the problem of designing bufferless NoCs (networks-on-chip). The proposed bufferless NoCs are deadlock/livelock-free and consume significantly less power than their buffered counterparts. We also present a multi-slice technique to reduce node capacity variations. This can make the proposed NoC architecture more cost efficient in a VLSI (very large-scale integration) implementation. In addition, we offer a detailed delay/throughput performance evaluation of the proposed bufferless NoC in the paper.

Perception is the fundamental task of any autonomous driving system, which gathers all the necessary information about the surrounding environment of the moving vehicle. The decision-making system takes the perception data as input and makes the optimum decision given that scenario, which maximizes the safety of the passengers. This paper surveyed recent literature on autonomous vehicle perception (AVP) by focusing on two primary tasks: Semantic Segmentation and Object Detection. Both tasks play an important role as a vital component of the vehicle’s navigation system. A comprehensive overview of deep learning for perception and its decision-making process based on images and LiDAR point clouds is discussed. We discussed the sensors, benchmark datasets, and simulation tools widely used in semantic segmentation and object detection tasks, especially for autonomous driving. This paper acts as a road map for current and future research in AVP, focusing on models, assessment, and challenges in the field.

This paper first reviews the related concepts of agricultural supply chain and the current research status of agricultural inventory models; proposes the classification standard of supply chain structure from four different aspects: the complexity of the supply chain node enterprises, the reliability of the supply chain and the optimization goal. In addition, this article starts from the characteristics of agriculture itself, according to the actual needs of the agricultural supply chain, considers the feasibility of the solution strategy from different sides and gives related models, which solves the related problems in the agricultural supply chain to a certain extent. At present, there are many researches on resource allocation algorithms at the edge of mobile networks, but the existing resource allocation algorithms still have the problem of high computational complexity and can still solve the above problems in resource utilization optimization. In order to solve these problems, this paper proposes a system architecture of the agricultural industry Internet of Things based on edge computing. The main motivations of this article are to improve the stability and usability of the system, and to enhance agricultural wisdom, so as to provide a reference for the development of precision agriculture. According to this method, Firstly, from the perspective of how to improve the energy efficiency of mobile terminals, methods to improve the energy efficiency of mobile terminals are studied in multi-user systems where mobile terminals compete for MEC servers with limited computing resources. Then, this paper focuses on the waste of computing resources caused by user movement under the scenario of limited server coverage in MEC system, and studies the problem from the perspective of effective management of computing resources. Experimental results show that the proposed algorithm has good performance.

Based on people’s enthusiasm for swimming and their pursuit of swimming skill at a higher level, this paper proposes a non-contact motion guidance system based on optical motion capture technology, which aims to improve their swimming technique conveniently, scientifically and quickly. The swimming actions can be captured using one smartphone underwater, and the app on the phone will process the video and give suggestions based on the reference postures derived from statistics of professional swimmers for improvement of the swimming performance. In this system, the Canny operator was applied to detect the edge of the image captured from the swimming video, and the Hough line was implemented to recognize and locate the key position information which refers to the angle between knee and hip joints, and the angle between the trunk and horizontal plane. Furthermore, by comparing the key position information with the data of standard swimming, improvement measures can be given to improve the swimming skills of swimmers. As the result of the experiments, the system can accurately capture the swimmer’s movement state, locate the incorrect posture of the swimmer, and provide the corrective information in real-time. Although the paper only took breaststroke as an example, the proposed method can be extended to the other swimming styles.

This paper presents a new method for computationally effective implementation of a discrete-time fractional-order proportional–integral–derivative (FOPID) controller. The proposed method is based on a unique representation of the FOPID controller, where fractional properties are modeled by a specific finite impulse response (FIR) filter. The balanced truncation model order reduction method is applied in the proposed approach to obtain an effective, low-order model of the FOPID controller. The time-invariant FOPID controller implementation is presented first, and then the methodology is extended to the controller with time-varying gains. A comparative analysis shows that the proposed methodology leads to the effective modeling of discrete-time FOPID controllers. In addition to simulation runs, the effectiveness of the introduced methodology is confirmed in a real-life experiment involving the control of the DC motor servo system. The paper concludes with the implementation tools developed in the Matlab/Simulink environment.

Altres ajuts: Secreteraria d'Universitats i Recerca (Generalitat de Catatunya), European Social Fund for the FI Grant, Institució Catalana de Recerca i Estudis Avançats This paper presents a single-frequency reflective-mode phase-variation microwave sensor devoted to the dielectric characterization of materials. The device is implemented in coplanar waveguide (CPW) technology and consists of two parts: (i) the sensing region, a step-impedance resonator (SIR) as termination of a CPW transmission line, and (ii) the design region, a cascade of high/low impedance quarter-wavelength inverters, used to boost up the sensitivity. By placing the so-called material under test (MUT) on top of the sensing region, the capacitance of the SIR is altered due to the effects of the dielectric constant of the MUT. This modifies the phase of the reflection coefficient seen from the input port, the output variable. From a circuit analysis, it is demonstrated that the sensitivity for small perturbations in the vicinity of the dielectric constant of a reference (REF) material can be optimized by setting the operation frequency of the sensor to the resonance frequency of the SIR loaded with such REF material. The maximum sensitivity in one of the reported sensors is as high as 66.5°, and the main figure of merit, defined as the ratio between the maximum sensitivity and the area of the sensing region expressed in terms of the squared guided wavelength, is FoM=3643∘/λ2 . Such figure of merit represents a significant improvement as compared to the one of the equivalent sensor implemented by means of an open-ended quarter-wavelength sensing line. Such equivalence between the semi-lumped element (i.e., SIR-based) sensor and the fully distributed counterpart is also analyzed in the paper.

In modern electronic warfare, both the radar and the jammer are capable of changing their waveforms to achieve better performance. The conflict relationship between the radar and jammer can be modeled as a non-cooperative game. In this paper, the optimal strategies of a monostatic radar and a jammer are investigated in terms of both Stackelberg game and symmetric game. Instead of using the widely used mutual information as the utility function, signal-to-jamming-plus-noise ratio (SJNR) is used in formulating the utility functions which is directly related to the target detection performance. In Stackelberg game, the Stackelberg equilibria solutions are derived when the leader is the radar and the jammer, respectively. In symmetric game, the existence of Nash equilibrium (NE) is investigated and the corresponding strategies are obtained when Nash equilibrium exists. If Nash equilibrium does not exist, the Stackelberg equilibria strategies are still useful as they can be regarded as the robust strategies which could optimize the worst case performance. The relationship between the mutual information utility function and the proposed SJNR utility function is investigated as well. Simulation results show the effectiveness of the radar and jammer strategies. The investigation in this paper can provide optimal waveform solutions for the monostatic radar and the jammer in the confrontation scenario.

Studies on requirements engineering with Agile methods for software development have shown difficulties in managing the quality of the requirements and communicating with users. Some of these studies have proposed conceptual modeling as a solution to these problems. However, the effort that is required to create conceptual models conflicts with Agile values. In this paper, we propose an approach for using conceptual models in projects while adhering to Agile principles. This approach focuses on projects in which requirements are expressed as user stories that are the main artifacts of the requirements used for software development with Agile methods. First, the paper presents a literature review in which we have systematically searched for the challenges to requirements engineering with Agile methods. Next, we report on a survey study in which we interviewed 16 experts in the Agile methodology to confirm the identified challenges and find new ones that are not covered in the literature. Based on a thematic analysis of the challenges, we argue that most of them map to the two main purposes of using conceptual models in software development: improving communication and understanding requirements. To effectively use conceptual models in projects that use the Agile methodology, several conditions must be met, which we make explicit in the paper. The paper ends by illustrating how these conditions can be met demonstrating the models that can be automatically generated from a given set of user stories. This demonstration was subsequently used to obtain feedback from the experts on the perceived benefits of conceptual models in addressing the challenges of requirements engineering.

Particle swarm optimization (PSO) is one of the most well-regarded swarm-based algorithms in the literature. Although the original PSO has shown good optimization performance, it still severely suffers from premature convergence. As a result, many researchers have been modifying it resulting in a large number of PSO variants with either slightly or significantly better performance. Mainly, the standard PSO has been modified by four main strategies: modification of the PSO controlling parameters, hybridizing PSO with other well-known meta-heuristic algorithms such as genetic algorithm (GA) and differential evolution (DE), cooperation and multi-swarm techniques. This paper attempts to provide a comprehensive review of PSO, including the basic concepts of PSO, binary PSO, neighborhood topologies in PSO, recent and historical PSO variants, remarkable engineering applications of PSO, and its drawbacks. Moreover, this paper reviews recent studies that utilize PSO to solve feature selection problems. Finally, eight potential research directions that can help researchers further enhance the performance of PSO are provided.

It is generally noticed that increasing the number of convolutional layers in generic image classification procedures proves to be detrimental to model performance in terms of validation accuracy and loss. Apart from vanilla CNNs, we have state-of-the-art (SOTA) architectures such as ResNet50 (and its variants) which show that through the use of skip-connections, higher performance metrics are attainable through deeper architectures. However, most evaluative metrics converge on a log scale as we go deeper with diminishing gradient of the metrics’ curves. Given these two contrasting speculations, in this paper, we implement various vanilla and SOTA CNNs for the diagnosis of one of the most common forms of breast cancer - invasive ductal carcinoma (IDC) - to examine and understand the feasibility of implementation of SOTA CNNs through transferred weights when juxtaposed with vanilla CNNs (and LeNet-5) of varying configurations in terms of their performance metrics and other parameters. In this paper, we solve the dual-objective of studying behavioural aspects of avant-garde CNN models (more specifically, VGG16, VGG19, ResNet50, ResNet50V2, MobileNetV2, and DenseNet121) and proper diagnosis of IDC through intermediate neural activations to critically evaluate and theorize the performance of different models. We notice that among all the models, only VGG16, VGG19, LeNet-5 and a selected vanilla CNN through an optimization procedure were the ones to attain the best metrics, shared amongst them.

Wearing safety helmets can effectively protect workers safety on construction sites. However, workers often take off the helmets because of weak security-conscious and discomfort, then hidden dangers will be brought by this behaviour. Workers without safety helmets will suffer more injuries in accidents such as falling human body and vertical falling matter. Hence, detecting safety helmet wearing is a vital step of construction sites safety management and a safety helmet detector with high speed and accuracy is urgently needed. However, traditional manual monitor is labour intensive and methods of installing sensors on safety helmet are difficult to popularize. Therefore, this paper proposes a deep learning-based method to detect safety helmet wearing at a satisfactory accuracy with high detection speed. Our method chooses YOLO v5 as the baseline, then the fourth detection scale is added to predict more bounding boxes for small objects and the attention mechanism is adopted in the backbone of the network to construct more informative features for following concatenation operations. In order to overcome the defects caused by insufficient data, targeted data augmentation and transfer learning are used. Improvements caused by every modification are discussed in this paper. Finally, our model achieves 92.2% mean average precision, up 6.3% compared to the original algorithm, and it only takes 3.0 ms to detect an image at $640 \times 640$ . These results demonstrate the robustness and feasibility of our model. Meanwhile, the size of our trained model is only 16.3 m, which means the model is easy to be deployed. At last, after obtaining a satisfactory model, a graphical user interface (GUI) is designed to make our algorithm more user-friendly.

Handwritten Digit Recognition (HDR) is one of the most challenging tasks in the domain of Optical Character Recognition (OCR). Irrespective of language, there are some inherent challenges of HDR, which mostly arise due to the variations in writing styles across individuals, writing medium and environment, inability to maintain the same strokes while writing any digit repeatedly, etc. In addition to that, the structural complexities of the digits of a particular language may lead to ambiguous scenarios of HDR. Over the years, researchers have developed numerous offline and online HDR pipelines, where different image processing techniques are combined with traditional Machine Learning (ML)-based and/or Deep Learning (DL)-based architectures. Although evidence of extensive review studies on HDR exists in the literature for languages, such as English, Arabic, Indian, Farsi, Chinese, etc., few surveys on Bengali HDR (BHDR) can be found, which lack a comprehensive analysis of the challenges, the underlying recognition process, and possible future directions. In this paper, the characteristics and inherent ambiguities of Bengali handwritten digits along with a comprehensive insight of two decades of state-of-the-art datasets and approaches towards offline BHDR have been analyzed. Furthermore, several real-life application-specific studies, which involve BHDR, have also been discussed in detail. This paper will also serve as a compendium for researchers interested in the science behind offline BHDR, instigating the exploration of newer avenues of relevant research that may further lead to better offline recognition of Bengali handwritten digits in different application areas., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. 38 pages, 23 figures, 12 tables

In this paper, a fully decentralized local energy market based on peer-to-peer(P2P) trading is proposed for small-scale prosumers. In the proposed market, the prosumers are classified as buyers and sellers and can bilaterally engage in energy trading (P2P) with each other. The buyer prosumers are equipped with electrical storage and can participate in a demand response (DR) program while protecting their privacy. In addition to bilateral negotiating with the local sellers, these players can compensate for their energy deficiency from the upstream market as the retail market at hours without local generation. In this paper, the retail market price is assumed uncertain. Robust optimization is applied to model this uncertainty in the buyer prosumers model. The proposed decentralized robust optimization guarantees the solution’s existence for each realization of uncertainty components. Furthermore, it performs optimization to realize the hard worse case from uncertainty components. A fully decentralized approach known as the fast alternating direction method of multipliers (FADMM) is employed to solve the proposed decentralized robust problem. The proposed approach does not require third-party involvement as a supervisory node nor disclose the players’ private information. Numerical studies were carried out on a small distribution system with several prosumers. The numerical results suggested the operationality and applicability of the proposed decentralized robust framework and the decentralized solving method.

By bringing content close to end-users, proactive caching plays a vital role in improving the user experience in wireless networks. Caching content at the network edge proactively has been particularly effective in fast-changing vehicular networks. The objective of this paper is to address the proactive caching problem at the next roadside unit (RSU) in vehicular networks using reinforcement learning techniques. The paper proposes two proactive caching algorithms based on multi-armed bandit (MAB) learning, namely non-contextual MAB-based (MAB) and contextual MAB-based (cMAB). Additionally, the paper also investigates the uncertainty associated with proactive caching systems in the form of entropy with a specifically extended Subjective Logic framework, providing an insight into the underlying link between prediction accuracy and uncertainty. Two cities: Las Vegas, USA with grid road layout and Manchester, UK with more complex and historical layout, are considered in the simulation. Results have shown the generality of the proposed schemes in cities with different road layouts. Performance of the two proposed MAB-based systems is compared with two non-contextual baseline system: Equal Probability-based and Probability-based, and one contextual baseline system named Compact Prediction Tree+ based. Both proposed systems outperformed their counterparts. In terms of the prediction accuracy, cMAB has reached 75% and 80\% accuracy in Las Vegas and Manchester respectively, and MAB reaches over 50% in both testing cities. Regarding the benefits to the vehicular network, cMAB and MAB perform similarly in both cities irrespective of the road layout. Particularly, the paper shows that on average 75% and 81% content fragments are proactively served with cMAB and over 50% with MAB in Las Vegas and Manchester, which is consistent with the prediction accuracy associated with the schemes.

As an indispensable part of intelligent transportation system, a traffic-sensitive navigation system can assist drivers in avoiding traffic congestion by providing navigation services. The navigation service provider (NSP) utilizes road condition information from nearby vehicles collected by roadside units (RSUs) to guide a vehicle through an optimal route provided by the navigation services. This paper mainly focuses on tackling three key issues in such navigation system. Firstly, it is essential to ensure the privacy of the vehicles’ personal identifiable information (PII). Secondly, the location, start point destination and route of the vehicle should not be leaked to RSUs and also should not be linked to its PII by RSUs and NSP. Thirdly, the process of vehicles obtaining the navigation services should be lightweight. Because of such problems, this paper proposes a dual privacy-preserving lightweight navigation system for vehicular networks through designing a novel signature scheme and combining other cryptographic primitives to keep dual privacy including PII and route related information. And the correctness analysis of the proposed system, security proof of the designed signature scheme adopted in the proposed system, and privacy analysis of the system are thoroughly provided. In addition, the performance of the proposed system is evaluated and compared with the existing systems to illustrate that the proposed system is more efficient than the existing systems.

Wireless Sensor Networks (WSNs) represent a key component in emerging distributed computing paradigms such as IoT, Ambient Intelligence, and Smart Cities. In these contexts, the difficulty of testing, verifying, and monitoring applications in their intended scenarios ranges from challenging to impractical. Current simulators can only be used to investigate correctness at source code level and with limited accuracy. This paper proposes a system and a methodology to model and verify symbolic distributed applications running on WSNs. The approach allows to complement the distributed application code at a high level of abstraction in order to test and reprogram it, directly, on deployed network devices. The proposed intelligent architecture enables the execution of distributed applications and the verification of the supplied correctness conditions. This paper shows the feasibility of the proposed approach and its effectiveness even when networks include resource-constrained nodes with some sample applications and quantitative experiments measuring the overhead introduced by the monitoring operations.

Sparse general matrix multiplication (SpGEMM) is an important and expensive computation primitive in many real-world applications. Due to SpGEMM's inherent irregularity and the vast diversity of its input matrices, developing high-performance SpGEMM implementation on modern processors such as GPUs is challenging. The state-of-the-art SpGEMM libraries (i.e., $nsparse$ and $spECK$) adopt several algorithms to tackle the challenges of global load balance, local load balance, and allocation of the result matrix. While these libraries focus on the high-level algorithm design for SpGEMM, they neglect several low-level architecture-specific optimizations, which causes inefficient implementations in their libraries. In this paper, we classify their inefficient implementations into seven categories. Based on our observations, we propose a highly optimized SpGEMM library called $OpSparse$. The optimizations in $OpSparse$ include 1) optimizing the binning method by improving the utilization of the shared memory, 2) optimizing the hashing method by reducing the access to the hash table, 3) improving the trade-off between hash collision rate and hardware utilization in the hashing method by setting appropriate binning ranges, 4) reducing the overheads of global memory utilization by minimizing the global memory usage of the metadata, and 5) improving the execution parallelism by overlapping global memory allocation with kernel execution. Performance evaluations with 26 commonly used matrices on an Nvidia Tesla V100 GPU show that $OpSparse$ achieves up to $27.8\times$, $1.81\times$, and $2.04\times$ performance speedup over three state-of-the-art libraries: $cuSPARSE$, $nsparse$, and $spECK$, respectively., This paper has been submitted to the IEEE Access since May 7, 2022, and is currently under review by IEEE Access. 20 pages, 11 fgures, 5 tables

This paper describes the lived experiences of experts who have taken a holistic approach to driving transformation in various organizations. The primary goal of the paper is to outline, at a high level, key insights and observations that are critically important to implementing Digital Business Transformation (DBT). The authors have undertaken primary interviews in which interviewees were asked to share their experience and opinions regarding organizational prerequisites for embarking on DBT, misconceptions about DBT, and the challenges to implementing DBT. These insights were analyzed and clustered and this paper articulates six lessons they have learned during the transformations. These lessons relate to Chief Executive Officer (CEO) commitment and the need for change leadership, the best team to drive the implementation successfully, the perception and behavior of stakeholders before, during and after the transformation, the importance of choosing the best digital platform to suit the business and the increased importance of relationships in the business ecosystem. We present the findings in a holistic framework and organizations seriously considering DBT implementation in their own business may consider these insights to be a good starting point. Refereed/Peer-reviewed

5G wireless communication systems provide massive system capacity with high data rates, very short low-latency, and ultra-high reliability, in addition to high connection density with a positive experience on smart cities and the Internet of Things (IoT). Future networks are anticipated to revolutionize typical applications such as the enhanced mobile broadband services (EMBB), ultra-reliable low latency communication (uRLLC), and massive machine-type communications (mMTC) anywhere and everywhere. This rationalizes the need to investigate the sustainable elements of 5G networks in smart cities to understand how 5G networks can be more environmentally- friendly and energy-efficient. This paper aims to investigate how 5G networks can act as key enablers in achieving sustainability in smart cities, using a macroscopic review. An overview of 5G communication networks and several 5G technologies used in smart city applications to enhance sustainability is presented. This is followed by investigating the indicators that measure sustainability in 5G networks across the environmental, social, and economic dimensions; and sub-dimensions such as energy efficiency, power consumption, carbon footprint, pollution, cost, health, safety, and security. The results show that the majority of research papers focus on the environmental dimensions of sustainability (42%) when attempting to address sustainability in 5G systems and smart cities. The findings also showed a huge interest in the economic (37%) and social (21%) dimensions as well. Further, when examining the sub-dimensions, it was found that most of the studies focused on energy efficiency (20%), power consumption (17%), and cost (15%).

This work presents a comprehensive study, from an industrial perspective, of the process between the collection of raw data, and the generation of next-item recommendation, in the domain of Video-on-Demand (VoD). Most research papers focus their efforts on analyzing recommender systems on already-processed datasets, but they do not face the same challenges that occur naturally in industry, e.g., processing raw interactions logs to create datasets for testing. This paper describes the whole process between data collection and recommendation, including cleaning, processing, feature engineering, session inferring, and all the challenges that a dataset provided by an industrial player in the domain posed. Then, a comparison on the new dataset of several intent-based recommendation techniques in the next-item recommendation task follows, studying the impact of different factors like the session length, and the number of previous sessions available for a user. The results show that taking advantage of the sequential data available in the dataset benefits recommendation quality, since deep learning algorithms for session-aware recommendation are consistently the most accurate recommenders. Lastly, a summary of the different challenges in the VoD domain is proposed, discussing on the best algorithmic solutions found, and proposing future research directions to be conducted based on the results obtained.

This paper focuses on the unsupervised segmentation of images, which is an essential topic in the field of computer vision. In the absence of prior knowledge, it is challenging to generate semantic segmentation regions based on image content automatically. In this paper, we consider unsupervised image segmentation from the perspective of sub-region clustering and graph convolution. We over-segment the source image into disjoint sub-regions and generate multiscale representative maps for each sub-region. To explore the potential contextual correlation between different sub-regions, we build a graph model that establishes the dependency relationship and use a graph convolution network to transfer long-range contextual information. The segmentation results are obtained by combining sub-regions clustering and graph convolution training. We conduct extensive experiments on three image datasets, and the results show that the proposed method can provide consistent and meaningful segmentation results.

Relational databases are storage for a massive amount of data. Knowledge of structured query language is a prior requirement to access that data. That is not possible for all non-technical personals, leading to the need for a system that translates text to SQL query itself rather than the user. Text to SQL task is also crucial because of its economic and industrial value. Natural Language Interface to Database (NLIDB) is the system that supports the text-to-SQL task. Developing the NLIDB system is a long-standing problem. Previously they were built based on domain-specific ontologies via pipelining methods. Recently a rising variety of Deep learning ideas and techniques brought this area to the attention again. Now end to end Deep learning models is being proposed for the task. Some publicly available datasets are being used for experimentation of the contributions, making the comparison process convenient. In this paper, we review the current work, summarize the research trends, and highlight challenging issues of NLIDB with Deep learning models. We discussed the importance of datasets, prediction model approaches and open challenges. In addition, methods and techniques are also summarized, along with their influence on the overall structure and performance of NLIDB systems. This paper can help future researchers start having prior knowledge of findings and challenges in NLIDB with Deep learning approaches.

Self-Sovereign Identity (SSI) is a novel and emerging, decentralized digital identity approach that enables entities to control and manage their digital identifiers and associated identity data while enhancing trust, privacy, security, and the many other properties identified and analyzed in this paper. The paper provides an overview and classification of the SSI properties, focusing on an in-depth analysis, furthermore, presenting a comprehensive collection of SSI properties that are important for the implementation of the SSI system. In addition, it explores the general SSI process flow, and highlights the steps in which individual properties are important. After the initial purification and classification phase, we then validated properties among experts in the field of Decentralized and Self-Sovereign Identity Management using an online questionnaire, which resulted in a final set of classified and verified SSI properties. The results can be used for further work on definition and standardization of the SSI field.

Safety is paramount in AV deployment. Traditionally, AV safety is incorporated during development by identifying the failure of vehicular components using risk and scenarios-based validation approaches with risk indicators. The main challenge of having a comprehensive risk assessment for AV safety is to include all potential complex environments that could occur in real-time, which is more critical for higher AV automation levels. Real-world real-time risk assessment research addresses this shortcoming by providing an advanced warning to the AV during deployment either at the vehicle level or at the infrastructure level. This paper proposes a risk tagging methodology to quantitatively risk tag the severity rating of the real-world environment for real-time risk assessment of AV using the existing roadside infrastructure. The proposed methodology - Spatial-Temporal Risk Estimation Ensemble Technique (STREET), provides advanced risk indicators in the form of pedestrian risk tag figure and time to collision value to the AV. This paper includes the evaluation of STREET, tested on four events over a pre-defined uncontrolled traffic scene from the infrastructure and validated using ground truth and heatmap of pedestrian occurrence. This methodology includes three different algorithms developed to emphasize different events depending on the AV risk and safety management strategy. The STREET reduces the bandwidth needed compared to traditional approaches of streaming video images for lightweight integration of AV risk assessment. The outcome of the pedestrian risk tag from STREET can be used as a severity rating for the existing real-time risk assessment of AV via cooperative mode.

Made available in DSpace on 2022-05-01T15:13:32Z (GMT). No. of bitstreams: 0 Previous issue date: 2022-01-01 This paper proposes a new design and practical implementation of a robust {\mathcal H} {\infty } control applied to some mechanical systems considering in the controller design full access to the state variables and in the equivalent controller implementation only the feedback of all positions or of all velocities of the plant. An important characteristic, though, is that limitations are considered for the state feedback. Two methods are presented, one that uses feedback only from state variables related to positions of the system and another that uses feedback only from state variables related to velocities. The control strategy is based on Linear Matrix Inequalities (LMIs), using the theory of \mathcal D -stability, which allows the designer to allocate the closed loop system eigenvalues in a negative complex semi-plane region, which, in addition to ensuring stability, also allows to attend certain performance requirements of the feedback system. The paper motivation is to provide satisfactory control results with limited states access, without any kind of estimation of the state variables that are not available. Therefore, the proposed control systems are interesting options as alternatives for the design of full-order state feedback for plants with uncertain parameters using only output feedback, considering that it is not required to build an observer to estimate any plant state variable. Furthermore, their implementations are relatively cheaper because it is not necessary to measure all state variables of the plant. Department of Electrical Engineering UNESP - São Paulo State University IFMS - Federal Institute of Education Science and Technology of Mato Grosso do Sul Department of Electrical Engineering UNESP - São Paulo State University

Radio frequency energy harvesting (RFEH) and wireless power transmission (WPT) are two emerging alternative energy technologies that have the potential to offer wireless energy delivery in the future. One of the key components of RFEH or WPT system is the receiving antenna. The receiving antenna’s performance has a considerable impact on the power delivery capability of an RFEH or WPT system. This paper provides a well-rounded review of recent advancements of receiving antennas for RFEH and WPT. Antennas discussed in this paper are categorized as low-profile antennas, multi-band antennas, circularly polarized antennas, and array antennas. A number of contemporary antennas from each category are presented, compared, and discussed with particular emphasis on design approach and performance. Current design and fabrication challenges, future development, open research issues of the antennas and visions for RFEH and WPT are also discussed in this review.

In view of the lack of data association in spatiotemporal information analysis and the lack of spatiotemporal situation analysis in knowledge graphs, this paper combines the semantic web of the geographic knowledge graph with the visual analysis model of spatial information and puts forward the comprehensive utilization of the related technologies of the geographic knowledge graph and big data visual analysis. Then, it realizes the situational analysis of COVID-19 (Coronavirus Disease 2019) and the exploration of patient relationships through interactive collaborative analysis. The main contributions of the paper are as follows. (1) Based on the characteristics of the geographic knowledge graph, a patient entity model and an entity relationship type and knowledge representation method are proposed, and a knowledge graph of the spatiotemporal information of COVID-19 is constructed. (2) To analyse the COVID-19 patients’ situations and explore their relationships, an analytical framework is designed. The framework, combining the semantic web of the geographic knowledge graph and the visual analysis model of geographic information, allows one to analyse the semantic web by using the node attribute similarity calculation, key stage mining, community prediction and other methods. (3) An efficient epidemic prevention and anti-epidemic method, which is of referential significance, is proposed. It is based on experiments and the collaborative analysis of the semantic web and spatial information, allowing for real-time situational understanding, the discovery of patients’ relationships, the analysis of the spatiotemporal distribution of patients, super spreader mining, key node analysis, and the prevention and control of high-risk groups.

While the performance of mmWave links has been thoroughly investigated by simulations or testbeds, the behavior of this technology in real-world commercial setups has not yet been thoroughly documented. In this paper, we address this gap and present the results of an empirical study to determine the actual performance of a commercial 5G mmWave cell through on-field measurements. We evaluate the signal and beam coverage map of an operational network as well as the end-to-end communication performance of a 5G mmWave connection, considering various scenarios, including human body blockage effects, foliage-caused and rain-induced attenuation, and water surface effects. To the best of our knowledge, this paper is the first to report on a commercial deployment while not treating the radio as a black box. Measurement results are compared with 3GPP's statistical channel models for mmWave to check the possible gaps between simulated and actual performance. This measurement analysis provides valuable information for researchers and 5G verticals to fully understand how a 5G mmWave commercial access network operates in real-world, under various operational conditions, with buildings, humans, trees, water surfaces, etc., 12 pages, 11 figures, 7 tables

The performance of the solar photovoltaic system is affected by the unpredictable phenomenon of partial shading. This causes the mismatch losses that suppress the power generation of healthy PV modules in it. The objective of the proposed work in this paper is to bring out the maximum power from each PV module in the PV array by reducing the mismatch losses. A new array configuration method is proposed in this paper, which follows the screw pattern in the row formation. Each PV row is created with distinct PV modules from the rows of the conventional array configuration. This proposed work allows the PV system to operate with minimum mismatch losses by even shade dispersion over the PV array. The proper mathematical expression with all necessary constraints was derived for the array formation of the proposed work. The output analysis is been validated in the simulation of the $9\times 9$ PV array in MATLAB/Simulink Ⓡ. The mismatch loss generation and output power enhancement are measured and compared with the various conventional array configuration methods under six kinds of partial shading patterns. The proposed array configuration is 40% more efficient than the conventional series-parallel array configuration and also performs better than the total cross-tied and sudoku puzzle pattern methods. The shade dispersion rate of the proposed array configuration has highly reduced the mismatch losses in the PV system and hence, it improves the power output.

Novel personalized and affordable approaches are needed in order to provide efficient therapeutic interventions for people with autism spectrum disorder. In this paper, we introduce a new category of socially assistive robots, i.e., socially assistive drones (SADs) for therapeutic interactions with humans. SADs autonomously perform and engage humans in dance movement therapy (DMT), which has proven to be highly effective for people with autism spectrum disorder when it is personalized and adapted carefully. The main focus of this paper is on developing adaptive, personalized, and interactive control approaches based on fuzzy logic, which efficiently deals with nonlinear dynamics and directly introduces the expert (linguistic) knowledge into the control system. The developed approaches are implemented via a small quadcopter in real-life interactions with 10 participants. The results of the case study prove the excellent performance of the SAD in adapting and personalizing its decisions with respect to each user and thus keeping them engaged in the proposed DMT plans.

The paper investigates cyber threats and potential solutions for protecting industrial control systems (ICS). On the cyber threats side, different off-the-shelf offensive solutions, both hardware and software, are analysed and tested. The goal of the paper is to increase cyber threat awareness by showing how such off-the-shelf solutions, well known to IT security experts, can be utilised as (or inspire) attack vectors to gain access to generally unprotected industrial plants. After obtaining an accessing point, Man-in-the- Middle (MITM) and Legal-Client-to-Server (LCSA) types of attacks from reconnaissance, client-to- server and server-to-client categories are demonstrated. For this purpose, a Modbus communication protocol implemented in a real compressor station is used as basis. Regarding potential protection solutions, the paper proposes a simple-to-implement and cheap hardening methodology applicable inside almost any industrial plant. A novel, PLC-based ICS cyber security protection method, made of a signal validity monitoring mechanism and a control system integrity check mechanism is also discussed and demonstrated. Both penetration testing and hardening methodology are verified experimentally, using real PLC and HMI devices.

The paper presents an empirical study conducted in the context of an advanced power electronics design course in a technical university. In selected points of the lectures formative assessments were introduced and supported by the use of clickers. The purpose of the article is to examine students’ perceptions of the effectiveness of clickers use. Individual, semi-structured student interviews conducted which were analysed using thematic content analysis. The findings reveal that the perceptions of the students with respect to the use of clickers were associated to: sustain student attention during the lecture, the provision of confirmatory and elaborative instructional feedback, the anonymity of the students’ answers, asking questions appropriate for this teaching/learning method, students’ motivation, students own suggestions for improvement, and the perceived existence of game-like elements. While in general the findings confirm the previous literature, there are a few points which are unique for this study, such as the relative importance of enhancing the students’ attention compared to other studies, as well as the importance of using appropriate questions from the subject-matter. Besides, the drawbacks of previous research on clickers are not in line with the perceived ones in the present study. The paper is concluded with providing research and practical implications. The former touch upon the potential of this study on informing the research on didactics in the field of electrical engineering in Higher Education, whereas the latter pertains to its contribution with respect to learning design in similar learning contexts.

Load forecasting is a critical aspect for power systems planning, operation and control. In this paper, as part of research efforts of an ambitious project at Memorial University of Newfoundland in St. John’s, Canada, to achieve more energy efficient and environmental friendly “Sustainable Campus”, we present a day-ahead load forecasting approach for the energy management system of the project. The hourly load consumption dataset from January 1, 2016 to March 31, 2020 is used in the paper, which was collected from two power meters on campus. Using the load consumption dataset along with the collected meteorological dataset, a total of 19 regression model-based day-ahead load forecasting algorithms for Memorial University of Newfoundland’s campus load are developed and evaluated in this paper. These 19 models belong to five families of regression models in MATLAB Regression Toolbox: Linear Regression, Regression Trees, Support Vector Machines (SVM), Gaussian Process Regression (GPR), and Ensemble of Trees. It is found that the family of GPR models shows the best load forecasting performance because they are nonparametric kernel-based probabilistic models. Two GPR models, Rational Quadratic GPR and Exponential GPR, are recommended as the best models for load forecasting through this study.

In this paper, we propose a novel speech enhancement paradigm which can effectively solve the problem of retrieving a desired speech signal in a multi-talker environment. The proposed speech enhancement paradigm involves a three-step procedure consisting of separation, ranking, and enhancement. First, a speech separation system – which could be a conventional spatial filter bank or more advanced separation systems – separates mixtures of speech signals captured by microphones into speech signals from candidate speakers. Next, novel ranking algorithms – proposed in this paper – are applied to determine the talker-of-interest amongst the separated speech signals. Finally, the speech signal of the talker-of-interest is estimated as a linear combination of the separated signals, whose weights are determined by the ranking algorithms. We propose ranking algorithms, which exploit turn-taking patterns between conversational partners in order to determine the talker-of-interest amongst competing speakers. Unlike some existing solutions, our ranking algorithms do not require access to additional sensors, e.g., EEG electrodes, cameras, etc., but only rely on microphone signals. Specifically, the proposed algorithms rank the separated speech signals based on the probability of speech overlaps and gaps with the user’s own voice. The speech signal with highest ranking is the talker with minimum probability of speech overlap and gap with the user’s own voice. The proposed ranking algorithms are shown highly effective at determining the talker-of-interest, since conversational partners, i.e., the user and the talker-of-interest, behaviorally avoid speech overlaps and gaps. We evaluate the proposed speech enhancement paradigm in two practical hearing aid related applications, where the objective is to enhance a speech signal of a conversational partner in a multi-talker environment. The results of the evaluation demonstrate that the proposed speech enhancement systems in both applications significantly outperform conventional speech enhancement systems.

Waste incineration plants are complex dynamical systems that rely on expert human operators to maintain steady combustion, by observing real-time in-chamber video feeds. Real-time plant forecasting provides vital operational support in decision making, and applying machine learning to automatically learn dynamics forecast models from video feeds is an attractive means to realise this. However, learning complex dynamics in systems that requires cost-efficiency remains an open research problem. Specifically, modelling plant dynamics in real-time is challenging due to uncertainties caused by inhomogeneous waste inputs, requiring complex learning that impedes real-time modelling. To address this, this paper presents a real-time data-driven framework for generating video forecasts, by incorporating task-relevant domain-knowledge, during learning. Specifically, this method combines dynamics modelling and forecasting using dynamic mode decomposition, with Fourier transformations informed by expert operator heuristic knowledge for encoding task-relevant frequency information inside the learning process. Experiments in this paper demonstrate that the proposed framework captures intuitive physical aspects of the underlying physiochemical process, with a greatly reduced computational runtime in comparison to standard approaches, allowing for application in real-time domains. Forecasted video predictions are accurate over short time horizons, and capture important system characteristics over longer time periods.

In this paper, I respond to a critique of one of my papers previously published in this journal, entitled "Bell's Theorem Versus Local Realism in a Quaternionic Model of Physical Space." That paper presents a local-realistic model of quantum correlations based on a quaternionic 3-sphere, taken as a physical space in which we are confined to perform all our experiments. The critique, on the other hand, considers two entirely different models within a flat Euclidean space, neither related to my quaternionic 3-sphere model. It then criticizes its own flat space models and claims that it has thereby criticized the model presented in my paper. Along the way, without providing evidence or proof, it claims that the results in my paper are based on mistakes. I demonstrate that there are no mistakes of any kind in my paper. On the contrary, I bring out a number of elementary mathematical and conceptual mistakes from the critique and the critiques it relies on., Comment: 12 pages, 4 figures. This paper is published in response to arXiv:2103.00225 [quant-ph]