Search

Holographic-type communication (HTC) is on the verge of revolutionizing current communication paradigms. It will seamlessly connect people from distant locations through a fully immersive experience that will engage all five senses: sight, hearing, touch, smell, and taste. However, the practical realization and widespread adoption of HTC impose significant demands on current networks and end-user devices. While recent studies have primarily focused on discussing the HTC challenges and potential research directions, this paper takes a further step by evaluating the capabilities of wireless networks to meet HTC requirements. Specifically, it highlights the limitations of the fifth-generation (5G) networks by identifying HTC-related Key Performance Indicators (KPIs) and explores the potential of the sixth-generation (6G) networks. Moreover, it not only proposes potential research approaches for HTC enhancement but also analyzes their impact on the challenges in HTC implementation. Finally, the paper questions the ubiquitous potential of 6G, suggesting that a coordinated approach leveraging artificial intelligence (AI) for jointly optimizing user sites and communication networks is the most promising strategy, rather than solely relying on the capabilities of specific network generations.

This paper presents a systematic bibliometric analysis of the artificial intelligence (AI) domain to explore privacy protection research as AI technologies integrate and data privacy concerns rise. Understanding evolutionary patterns and current trends in this research is crucial. Leveraging bibliometric techniques, the authors analyze 8,322 papers from the Web of Science (WoS) database, spanning 1990 to 2023. The analysis highlights IEEE Transactions on Knowledge and Data Engineering and IEEE Access journals as highly influential, the former being an early contributor and the latter emerging as a pivotal source. The study demonstrates substantial disparities in scientific productivity across countries. Specifically, the top 10 countries collectively accounted for 74.8% of the articles, with China and the USA making up nearly half of the total contribution (46.1%). In contrast, regions in Africa and South America exhibited lower scientific production. The evolution of privacy preservation research is reflected, shifting from an algorithm-oriented approach to a focus on data orientation, and subsequently, to privacy solutions centered around Cloud Computing. In recent years, there has been a shift towards embracing Federated Learning and Differential Privacy. The analysis brings to light emerging themes and identifies research gaps, notably a global disparity in research output and a lag in ethical and legal inquiry. It asserts that enhanced interdisciplinary collaboration is imperative to formulate comprehensive privacy solutions for AI. Specifically, the paper imparts invaluable insights that are pivotal for effectively addressing the evolving privacy concerns in the era of AI and big data.

Metaverse is a virtual world platform that enables users to create and explore 3D environments. Recently, this technology has gained traction due to its potential to revolutionize the way people interact and share information. This paper provides an overview of the current state of Metaverse technologies. It begins by introducing their core components and surveys recent research and development activities in the area. Next, the paper examines the potential applications of Metaverse technology in different domains, such as gaming, social networking, training, education, healthcare, and marketing. It then outlines the challenges and opportunities associated with the development and deployment of Metaverse technologies, including scalability, privacy, and security issues. Finally, the paper concludes with a discussion of the implications of Metaverse technologies for the future of digital communication and information sharing.

The potential of digital twin technology to generate virtual models and simulations of real assets and processes has contributed to the widespread adoption of this technology across a variety of business sectors. As a result, performance, cost savings, and overall efficiency have all seen significant improvements. By conducting a study of the academic literature and previous research, this article offers readers an introduction to the origins, evolution, and applications of digital twin technology. This paper looks at the different kinds of digital twins, the advantages, and disadvantages of using them, and the causes that have contributed to their rising popularity. Case examples illustrating effective implementations of digital twin technology in industry and healthcare are included in the article. An overview of digital twin technology and its prospective applications in several industries is provided in this paper. This paper intends to contribute to the current understanding of digital twin technology and its potential impact by discussing the advantages, limitations, and prospects for further study and development.

As the number of smart connected devices increases day by day, a massive amount of tasks are generated by various types of Internet of Things (IoT) devices. Intelligent edge computing is a promising enabler in next-generation wireless networks to execute these tasks on proximate edge servers instead of smart devices. Additionally, regarding the execution of tasks in edge servers, smart devices could provide a low-latency environment to the end users. Within this paper, an artificial intelligence (AI)-empowered fast task execution method in heterogeneous IoT applications is proposed to reduce decision latency by taking into account different system parameters such as the execution deadline of the task, battery level of devices, channel conditions between mobile devices and edge servers, and edge server capacity. In edge computing scenarios, the number of task requests, resource constraints of edge servers, mobility of connected devices, and energy consumption are the main performance considerations. In this paper, the AI-empowered fast task decision method is proposed to solve the multi-device edge computing task execution problem by formulating it as a multi-class classification problem. The extensive simulation results demonstrate that the proposed framework is extremely fast and precise in decision-making for offloading computation tasks compared to the conventional Lyapunov optimization-based algorithm results by ensuring the guaranteed quality of experience.

With the rapid transformation of the energy sector towards modern power systems represented by smart grids (SGs), microgrids (MG), and distributed generation, blockchain (BC) technology has shown the capability for solving security, privacy, and reliability challenges that hinder progress. Currently, the energy structure is forming a decentralized system that prioritizes customer satisfaction. BC technology undertakes power network stockholders in a secure energy market, transparent transactions, and fair competition and offers promising energy solutions. This paper is a comprehensive review of energy applications using BC integration. Firstly, we introduce the drivers of BC leverage that make it a potentially important component of the power network. Following that, we provide background information on BC and its application in areas other than the energy sector. Subsequently, we discuss studies and sort potential energy applications from various recent papers and surveys that have already adopted BC technology in the energy sector. Then, we summarize the pricing infrastructure for applying BC in the energy sector and identify the requirements to build it. Finally, energy security and privacy challenges based on BC are highlighted, along with potential drawbacks and concerns related to the pricing infrastructure.

Nowadays, health insurance has become an essential part of people’s lives as the number of health issues increases. Healthcare emergencies can be troublesome for people who can’t afford huge expenses. Health insurance helps people cover healthcare services expenses in case of a medical emergency and provides financial backup against indebtedness risk. Health insurance and its several benefits can face many security, privacy, and fraud issues. For the past few years, fraud has been a sensitive issue in the health insurance domain as it incurs high losses for individuals, private firms, and governments. So, it is essential for national authorities and private firms to develop systems to detect fraudulent cases and payments. A high volume of health insurance data in electronic form is generated, which is highly sensitive and attracts malicious users. Motivated by these facts, we present a systematic survey for Artificial Intelligence (AI) and blockchain-enabled secure health insurance fraud detection in this paper. This paper presents a taxonomy of various security issues in health insurance. We proposed a blockchain and AI-based secure and intelligent system to detect health insurance fraud. Then, a case study related to health insurance fraud is presented. Finally, the open issues and research challenges in implementing the blockchain and an AI-empowered health insurance fraud detection system is presented.

The research on chatbots has gained momentum over the past few years. Academics and practitioners investigate how these tools for communication with customers or internal team can be improved in terms of their performance, acceptance, and deployment. Although there is a plethora of recent studies available, not all of them deal with the digital business transformation implications of chatbots. The main aim of the research presented in this paper was to conduct a systematic literature review of high-quality journal research papers in order to summarise the current state of research on chatbots, identify their role in digital business transformation and suggest the areas warranting further attention. 74 papers were included in the research. Topical (focus and applications), methodological (methods used, sample size, sample type, and countries studied) and bibliometric (publication outlet, citations, and Altmetric Attention Score) aspects are evaluated and described. Scholars and practitioners can use the results to identify topics, areas, and applications that are intensely discussed in the literature and require further attention, select a methodology for their research that is well established in the field or is emerging, identify the most influential publications not to be missed in their research or identify publication outlets for publishing their research on chatbots.

Predicting the time, location and magnitude of an earthquake is a challenging job as an earthquake does not show specific patterns resulting in inaccurate predictions. Techniques based on Artificial Intelligence (AI) are well known for their capability to find hidden patterns in data. In the case of earthquake prediction, these models also produce a promising outcome. This work systematically explores the contributions made to date in earthquake prediction using AI-based techniques. A total of 84 scientific research papers, which reported the use of AI-based techniques in earthquake prediction, have been selected from different academic databases. These studies include a range of AI techniques including rule-based methods, shallow machine learning and deep learning algorithms. Covering all existing AI-based techniques in earthquake prediction, this article provides an account of the available methodologies and a comparative analysis of their performances. The performance comparison has been reported from the perspective of used datasets and evaluation metrics. Furthermore, using comparative analysis of performances the paper aims to facilitate the selection of appropriate techniques for earthquake prediction. Towards the end, it outlines some open challenges and potential research directions in the field.

Currently, Machine Learning (ML) is becoming ubiquitous in everyday life. Deep Learning (DL) is already present in many applications ranging from computer vision for medicine to autonomous driving of modern cars as well as other sectors in security, healthcare, and finance. However, to achieve impressive performance, these algorithms employ very deep networks, requiring a significant computational power, both during the training and inference time. A single inference of a DL model may require billions of multiply-and-accumulated operations, making the DL extremely compute- and energy-hungry. In a scenario where several sophisticated algorithms need to be executed with limited energy and low latency, the need for cost-effective hardware platforms capable of implementing energy-efficient DL execution arises. This paper first introduces the key properties of two brain-inspired models like Deep Neural Network (DNN), and Spiking Neural Network (SNN), and then analyzes techniques to produce efficient and high-performance designs. This work summarizes and compares the works for four leading platforms for the execution of algorithms such as CPU, GPU, FPGA and ASIC describing the main solutions of the state-of-the-art, giving much prominence to the last two solutions since they offer greater design flexibility and bear the potential of high energy-efficiency, especially for the inference process. In addition to hardware solutions, this paper discusses some of the important security issues that these DNN and SNN models may have during their execution, and offers a comprehensive section on benchmarking, explaining how to assess the quality of different networks and hardware systems designed for them.

Demand is growing for more accountability regarding the technological systems that increasingly occupy our world. However, the complexity of many of these systems—often systems-of-systems—poses accountability challenges. A key reason for this is because the details and nature of the information flows that interconnect and drive systems, which often occur across technical and organizational boundaries, tend to be invisible or opaque. This paper argues that data provenance methods show much promise as a technical means for increasing the transparency of these interconnected systems. Specifically, given the concerns regarding ever-increasing levels of automated and algorithmic decision-making, and so-called “algorithmic systems” in general, we propose decision provenance as a concept showing much promise. Decision provenance entails using provenance methods to provide information exposing decision pipelines: chains of inputs to, the nature of, and the flow-on effects from the decisions and actions taken (at design and run-time) throughout systems. This paper introduces the concept of decision provenance, and takes an interdisciplinary (tech-legal) exploration into its potential for assisting accountability in algorithmic systems. We argue that decision provenance can help facilitate oversight, audit, compliance, risk mitigation, and user empowerment, and we also indicate the implementation considerations and areas for research necessary for realizing its vision. More generally, we make the case that considerations of data flow, and systems more broadly, are important to discussions of accountability, and complement the considerable attention already given to algorithmic specifics.

With the rise of artificial intelligence (AI) and deep learning techniques, fake digital contents have proliferated in recent years. Fake footage, images, audios, and videos (known as deepfakes) can be a scary and dangerous phenomenon and can have the potential of altering the truth and eroding trust by giving false reality. Proof of authenticity (PoA) of digital media is critical to help eradicate the epidemic of forged content. Current solutions lack the ability to provide history tracking and provenance of digital media. In this paper, we provide a solution and a general framework using Ethereum smart contracts to trace and track the provenance and history of digital content to its original source even if the digital content is copied multiple times. The smart contract utilizes the hashes of the interplanetary file system (IPFS) used to store digital content and its metadata. Our solution focuses on video content, but the solution framework provided in this paper is generic enough and can be applied to any other form of digital content. Our solution relies on the principle that if the content can be credibly traced to a trusted or reputable source, the content can then be real and authentic. The full code of the smart contract has been made publicly available at Github.

This paper presents an intelligent assistant for preventing the injuries during motion gaming. In our previous work, we invented UKI, a middleware that allows its users to play any existing game by using body motions. The goal of this paper is to add an intelligent module to the UKI for promoting healthy motion gameplay. According to the literature in sports science and medicine, several exercise injuries are caused by the unbalanced use of body segments and repetitive movement. To prevent such injuries, we present a module that monitors and promotes two health metrics of the player: balancedness in use of body segments and variability in body movement. The proposed module recommends motions that will improve the two health metrics during gameplay. We evaluate effectiveness of the module on 45 participants with normal health condition. Two methods for motion recommendation were compared: recommendation according to the health fitness functions (HFF) and recommendation by the random motion selection. Our results indicate significant improvement in the two health metrics when HFF is used.

Citizens of the digital world are threatened. The digital systems that surround them are increasingly able to make autonomous decisions over and above them and on their behalf. They feel that their moral rights, as well as the social, economic, and political spheres, can be affected by the behavior of such systems. Although unavoidable, the digital world is becoming uncomfortable and potentially hostile to its users as human beings and as citizens. Notwithstanding the introduction of the GDPR and of initiatives to establish criteria on software transparency and accountability, users feel vulnerable and unprotected. In this paper, we present EXOSOUL, an overarching research framework that aims at building a software a personalized exoskeleton that enhances and protects users by mediating their interactions with the digital world according to their own ethics of actions and privacy of data. The exoskeleton disallows or adapts the interactions that would result in unacceptable or morally wrong behaviors according to the ethics and privacy preferences of the users. With their software shield, users will feel empowered and in control, and more in the balance of forces with the other actors of the digital world. To reach the breakthrough result of automatically building a personalized exoskeleton, EXOSOUL identifies multidisciplinary challenges never touched before: 1) defining the scope for and inferring citizen's ethical preferences; 2) treating privacy as an ethical dimension managed through the disruptive notion of active data; and 3) automatically synthesizing ethical actuators, i.e., connector components that mediate the interaction between the user and the digital world to enforce her ethical preferences. In this paper, we discuss the research challenges of EXOSOUL in terms of their feasibility and risks.

In the cloud-based Internet of Things (IoT) environments, quantifying uncertainty is an important element input to keep the acceptable level of reliability in various configurations. In this paper, we aim to address the pricing model of delivering data over the cloud while taking into consideration the dynamic uncertainty factors such as network topology, transmission/reception energy, nodal charge and power, and computation capacity. These uncertainty factors are mapped to different nodes with varying capabilities to be processed using Artificial Intelligence (AI)-based algorithms. Accordingly, we aim to find a way to calculate and predict the price per big data service over the cloud using AI and deep learning. Therefore, in this paper, we propose a framework to address big data delivery issues in cloud-based IoT environments by considering uncertainty factors. We compare the performance of the framework using two AI-based techniques called Genetic Algorithm (GA) and Simulated Annealing Algorithm (SAA) in both centralized and distributed versions. The use of AI techniques can be applied in multilevel to provide a kind of deep learning to further improve the performance of the system under study. The results reveal that the distributed algorithm outperforms the centralized one. In addition, the results show that the GA has lower running time compared to the SAA in all the test cases such as 68% of improvement in the centralized version, and 66% of improvement in the distributed version in case when the size of uncertainty array is 256. Moreover, when the size of uncertainty array increases, the results show 60% speed up in the distributed GA compared to its centralized version. The improvements achieved would help the service providers to actually improve their profit using the proposed framework.

Today’s rail industry deploys innovative applications for real-time monitoring of trains and railway infrastructures and remote control of trains to enhance the safety and security of passengers and train operation. These applications can require multiple constraints satisfactions in terms of latency, data rate, reliability, coverage, mobility, security, etc. In order to meet the desired quality of users’ (trains, passengers, controllers, etc.,) experience, it is essential to apply and combine various enabling wireless technologies for the railway communications architecture such as end-to-end (from the cloud hosting applications to the user client, traversing communication network infrastructure) Network Slicing (NS), Software Defined Network (SDN), Network Function Virtualization (NFV), Multi-access Edge Computing (MEC), Blockchain, Artificial Intelligence (AI) tools, etc. This paper aims to propose a global unified control framework to automatically compute deploy, and manage necessary resources on the different network infrastructures (cloud, core, transport, and multiple-access networks). In this context, this article proposes the first study in the railway sector to link application constraints and enabling technologies. After a brief overview of the railway context, the main applications requirements, and the technologies that could be applied, we propose a high-level architecture for railway applications incorporating the various technologies identified: MEC, NFV, SDN, AI tools, and Blockchain. We also study the impact that these different technologies could have on existing rail applications and identify the challenges and future directions for rail networks in the era of 5G and future 6G.

Sports multimedia is among the most prominent types of content distributed across social media today, and the retargeting of videos for diverse aspect ratios is essential for a suitable representation on different social media platforms. In this respect, ice hockey is quite challenging due to its agile movement pattern and speed, and because the main reference point (puck) is very small. In this paper, we introduce a novel pipeline for intelligent video cropping tailored for ice hockey. Our main goal is to identify regions of interest in video frames by detecting and tracking the hockey puck using state-of-the-art AI models. Our pipeline employs scene detection, object detection, outlier detection, and smoothing as key features. Our proposed pipeline called SmartCrop-H is not only highly efficient and configurable with respect to target aspect ratios, but also addresses the automation needs in this domain. Our comprehensive evaluation, comprising objective and subjective measures, shows the overall efficiency of the entire pipeline, including assessments of both the individual components and the end-to-end system performance. We also demonstrate the practical applicability of SmartCrop-H with a user study, which indicates that our framework performs on par with, or even surpasses, professional tools in terms of output quality.

In recent times, both geopolitical challenges and the need to counteract climate change have led to an increase in generated renewable energy as well as an increased demand for clean electrical energy. The resulting variability of electricity production and demand as well as an overall demand increase, put additional stress on the existing grid infrastructure. This leads to strongly increased maintenance demands for distribution system operators (DSOs). Today, condition monitoring is used to address these challenges. Researchers have already explored solutions for monitoring critical assets like switchgear and circuit breakers. However, with a shrinking knowledgeable technical workforce and increasing maintenance requirements, mere monitoring is insufficient. Already today, DSOs ask for actionable recommendations, optimization strategies, and prioritization methods to manage the growing task backlog effectively. In this paper we propose a vision of a grid-level cognitive assistance system that translates the outcome of diagnosis and prognosis systems into actionable work tasks for the grid operator. The solution is highly interdisciplinary and based on empirical studies of real-world requirements. We also describe the related work relevant to the multi-disciplinary aspects and summarize the research gaps that need to be closed over the next years.

Open innovation, which blurs the boundaries of organizations using inflows and outflows of knowledge to boost their innovation processes, has transformed the innovation paradigm, evolving from higher degrees of protectionism to cooperative relationships. Nevertheless, frequently the management of the huge amount of information and data generated in an open innovation ecosystem requires the use of information and communication technologies. In this context, artificial intelligence can be a major help to profit from all the opportunities derived from open innovation. Considering the growing body of academic literature dealing with the use of artificial intelligence tools in the context of open innovation environments, the objective of this article is revealing the main references, the academic trends and the hottest topics dealing with this subject, disentangling the knowledge structure of the research through a bibliometric analysis carried out over 63 papers selected from Web of Science database, using both co-word analysis and bibliographic coupling. The recent burst in the academic production anticipates a potentially massive interest in this topic, which is studied by the literature at three different levels (operational, managerial, and social). This study reveals the existence of relevant research opportunities, specially related with the management of the potential conflicts that may stem from the fuzzy ownership of the data generated by an artificial intelligence, and the roles of the different agents in such context.

The demand for Artificial Intelligence (AI) based solutions is exponentially increasing in all application fields, including low-power devices on the edge. However, due to their limited computational capabilities, these devices, which run Central Processing Units (CPUs) tailored to embedded applications, are typically not optimized to run complex neural networks. Providing ad-hoc extensions to the instruction set architecture of a RISC-V processor can be a viable solution to address this issue. In this work, we propose the use of the PyTorch Graph Lowering (Glow)–LLVM toolchain to understand the impact of compiled code of AI models on a RISC-V machine and extend its instruction set to improve runtime performance. This approach allows code profiling, detection of computational bottlenecks, and provisioning of the necessary CPU enhancements to be implemented in the LLVM backend before hardware implementation. After profiling known Artificial Neural Networks (ANNs) quantized to int8 (particularly, a single perceptron, RESNET18, VGG11, and LENET5), we have identified and devised three additional instructions, we named LWM, LWA and LWS respectively indicating Load Word-and-Multiply, -Add, and -Subtract. As a result, we obtained an edge AI-oriented, significantly improved processor description in terms of inference time and program density, ready to be hardware-designed. For $128\times 128$ RGB images, the custom extensions enabled up to $13\times $ speed up compared to RV32I and $5\times $ compared to RV32IM, with a maximum of 11.7% lower code. Together with these findings the paper systematically highlights the main methodological steps to include new instructions in an LLVM backend.

Major countries around the world are leveraging information and communications technology (ICT), such as artificial intelligence, big data, cloud computing and cybersecurity, to strengthen their militaries. These technological advancements are driving the changes in weapon systems, strategies, and tactics. In 2021, the South Korean Ministry of National Defense introduced the concept of “military meta power” to describe the power generated in the cognitive military domain utilizing ICT. This concept is significant in that it differentiates between cognitive and physical military power, emphasizing the importance of the cognition-based power in future warfare. Additionally, the concept consolidates various cognitive military capabilities, including AI capability, cyber capability, space capability, and C4ISR (Command, Control, Communications, Computers, Intelligence, Surveillance, and Reconnaissance), into a unified ICT-generated power. This study aims to contribute to the literature on “military meta power” by providing a comprehensive overview of its characteristics, composition, and attributes. To achieve this objective, the research begins with an exploration of the philosophical insights on technology related to ICT, followed by a comprehensive review of the perspectives of existing academic papers. Finally, this study provides an in-depth analysis of various military strategies.

As reliance on disruptive applications based on Artificial Intelligence (AI) and Blockchain grows, the need for secure and trustworthy solutions becomes ever more critical. Whereas much research has been conducted on AI and Blockchain, there is a shortage of comprehensive studies examining their integration from a security perspective. Hence, this survey addresses such a gap and provides insights for policymakers, researchers, and practitioners exploiting AI and Blockchain’s evolving integration. Specifically, this paper analyzes the potential benefits of the integration of AI and Blockchain as well as the related security concerns, identifying possible mitigation strategies, suggesting regulatory measures, and describing the impact it has on public trust.

In order to leverage the advantages of the large sample capacity in the era of big data and improve the performance of contemporary news recommendation methods, we have characterized complete information such as news textual details, explicit themes, and implicit themes, and thoroughly portrayed users’ mixed full term interest including long-term and short-term interests. As a result, we propose an artificial intelligence (AI)-News Personalization system that combines Complete news Content Characterization and user Full term interest portrayal, i.e., NP-3C-FIP. The NP-3C-FIP system first utilizes Latent Dirichlet Allocation (LDA) to extract the implicit theme distribution from the news textual content. Then, it learns the unified news characterizations based on the headline, summary, category, subcategory, and implicit themes. Using these news characterizations, the proposed method transforms the historical clicked news into the representation vectors. Subsequently, the obtained sequence of news representation vectors is fed into a Gate Recurrent Unit (GRU) network to capture the sequential interest features of the user. Furthermore, this paper introduces a personalized attention mechanism to model the stable tastes. Finally, the system concatenates the portrayals of the full term to obtain a unified user representation vector, and calculates the click probability for candidate articles using vector dot product. Experimental results demonstrate the effectiveness of the proposed method in improving news recommendation performance.

Touch-enabled sensation and actuation are expected to be the most promising, straightforward, and important uses of the B5G/6G communication networks. In light of the next generation (6G) systems’ prerequisite for low latency, the infrastructure should be reconfigurable, intelligent, and interoperable in the real-time existing wireless network. It has a drastic impact on society due to its high precision, accuracy, reliability, and efficiency, combined with the ability to connect a user from remote areas. Hence, the touch-enabled interaction is primarily concerned with the real-time transmission of tactile-based haptic information over the internet, in addition to the usual audio, visual, and data traffic, thus enabling a paradigm shift towards a real-time control and steering communication system. The existing system latency and overhead often have delays and limitations on the application’s usability. In light of the aforementioned concerns, the study proposes an intelligent touch-enabled system for B5G/6G and an IoT-based wireless communication network, incorporating AR/VR technologies. The tactile internet and network-slicing serve as the backbone of touch technology and incorporates intelligence from techniques such as artificial intelligence and machine/deep learning. The survey also introduces a layered and interfacing architecture with its E2E solution for the intelligent touch-based wireless communication system. It is anticipated for the upcoming 6G system to provide numerous opportunities for various sectors to utilize AR/VR technology in robotics and healthcare facilities to help in addressing several problems faced by society. Conclusively the article presents a few use cases concerning the deployment of touch infrastructure in automation, robotics, and intelligent healthcare systems, assisting in the diagnosis and treatment of the prevailing Covid-19 cases. The paper concludes with some considerable future research aspects of the proposed system with a few ongoing projects concerning the development and incorporation of the 6G wireless communication system.

Radio frequency sensing (RF-sensing) is an emerging field assisting vision technology for object detection, tracking, and various such use cases. RF-Sensing technology uses radio signals and their reflections in order to capture the surrounding details and then applies artificial intelligence algorithms to sense the objects. RF-sensing has advantage over the existing vision technology when it comes to low light conditions, privacy concerns, and far-range scenarios. In this paper, we propose different RF-sensing methods for person identification (PI), human activity identification (HAI), and surrounding-location identification (SLI). Here, we make use of data generated from millimeter-Wave sensors and pre-process it using signal processing techniques before feeding it to deep learning (DL) based classifiers. For PI and SLI, we propose a DL network consisting of convolutional neural network (CNN) and long short term memory (LSTM) blocks. Here, first the input is passed through the CNN layers and then followed by the LSTM. For HAI, we propose a hierarchical classifier consisting of two stages. First, similar activities are clubbed together and the coarse level classifier is trained to distinguish among the non-similar activities. This is followed by fine level classifiers, which are trained to distinguish between those activities which are deemed to be similar while training the coarse level classifier. While testing, we employ the coarse level classifier unconditionally. The fine level classifier is employed only when the coarse level classifier’s output class corresponds to that of similar activity. We use the output of the hierarchical classifier to determine the activity. Further, we show that the proposed algorithms for the PI and SLI are able to predict accurately among 5 persons and among 5 different surrounding-locations respectively and the proposed hierarchical classifier for HAI, which is used to classify among 11 different activities, shows an improvement of $\approx 4 \%$ compared to other DL based architectures in the literature.

The COVID-19 pandemic has revealed several limitations of existing healthcare systems. Thus, there is a surge in healthcare innovation and new business models using computer-mediated virtual environments to provide an alternative healthcare system. Today, digital transformation is not limited to virtual communication alone but encompasses digitalizing the network of social connections in the healthcare industry using metaverse technology. The metaverse is a universal and immersive virtual world facilitated by virtual reality (VR) and augmented reality (AR). This paper presents the first effort to offer a comprehensive survey that examines the latest metaverse developments in the healthcare industry, which covers seven domains: telemedicine, clinical care, education, mental health, physical fitness, veterinary, and pharmaceuticals. We review metaverse applications and deeply discuss technical issues and available solutions in each domain that can help develop a self-sustaining, persistent, and future-proof solution for medical healthcare systems. Finally, we highlight the challenges that must be tackled before fully embracing the metaverse for the healthcare industry.

Recent advances in artificial intelligence (AI) and continuous monitoring of patients using wearable devices have enhanced the accuracy of diagnosing various arrhythmias, from the captured Electrocardiogram (ECG) signals. Achieving high accuracy when using Deep Neural Network (DNN) for ECG classification is accomplished at the cost of compute and memory intensive operations, thus limiting its deployment to devices with high computing capabilities, and makes it unsuitable for wearable edge devices. To facilitate the deployment of deep neural networks on wearable mobile edge devices with limited resources, a lightweight Convolution Neural Network (CNN) model based on the ShuffleNet architecture is proposed and implemented as a solution in this paper. A sliding window of variable stride is used to increase the number of under-represented classes in the database. Moreover, a novel encoding scheme is employed for labelling and training test set samples, allowing the model to detect multiple classes in one ECG segment. A loss function (Focal loss) that proved to be effective when applied for DNN training on an imbalanced dataset was also explored in this work. The proposed model outperformed traditional CNN with 9x less trainable parameters and improved the F1-score by 2%.

There has been an enormous increase in information flow and communication data due to the rapid rise in the number of Internet of Everything (IoE) devices and the development of cutting-edge technologies such as the rollout of the Sixth Generation (6G) network. The rising and inevitable off-loading requirements of IoE devices have resulted in an unprecedented increase in the reliance on edge and cloud paradigms. However, such a reliance on far-end technologies to access already scarce resources can often result in increased latency and unstable connection issues due to limited bandwidth. In this paper, we investigate the solution for such a stringent network design by presenting a conceptual cloud architecture based on key components such as resource allocation, scheduling and task off-loading for IoE devices. The IoE devices utilise a scheduler to access resources from nearby higher resourced IoE devices for their task computation, where the scheduler allocates incoming requests according to the availability of resources within a cluster of devices or to other devices in nearby clusters. Motivated by these design characteristics, we propose a design of a novel Main Task Off-loading Scheduling Algorithm (MTOSA) for efficient task allocation and dissemination. We present a theoretical analysis of five different scheduling policies namely Round Robin (RR), Strongest Channel (SC), Max Rate (MR), Proportional Fair (PF) and Priority Base (PB) scheduling to find an optimal technique for task off-loading in futuristic networks. Furthermore, we compare the performance of these five scheduling policies with the two existing scheduling policies from the literature. It is shown through various experiments that the proposed MTOSA algorithm performs better when compared with the existing schemes for different performance parameters.

The rapid diagnosis and surgical prediction of necrotizing enterocolitis (NEC) remain a challenge because its complex pathogenesis has not been completely elucidated, and no single medical examination is specific for diagnosing NEC. Artificial intelligence (AI) has proven the robustness of multivariate analysis and been widely used in the diagnosis of complex diseases in the past decade. In this paper, a new multimodal AI system including feature engineering, machine learning (ML), and deep learning (DL) was constructed based on abdominal radiographs (ARs) and clinical data. A total of 4,535 ARs from 1,823 suspected NEC patients were analyzed by transfer learning, and then medical images and clinical parameters from 827 suspected NEC patients were used to train, validate, and test the AI system. Our results demonstrated that the system was effective in diagnosing NEC. In addition, the clinical datasets obtained one week before surgery from 379 NEC patients were studied by the multimodal AI system, and the results showed that it was capable of predicting which NEC patients required surgery. We compared the results in external test sets with those made by clinicians and found that the diagnostic and surgical predictive ability of the AI system was equivalent to that of experienced clinicians. This multimodal AI system can help clinicians improve diagnostic efficiency, reduce the number of missed diagnoses, and facilitate early diagnosis and treatment to prevent disease progression or even death.

Artificial Intelligence (AI) refers to systems designed by humans, interpreting the already collected data and deciding the best action to take, according to the pre-defined parameters, in order to achieve the given goal. Designing, trial and error while using AI, brought ethics to the center of the dialogue between tech giants, enterprises, academic institutions as well as policymakers. Ethical challenges in AI brought ethical AI framework in place in an attempt to regulate people’s lives and interactions, used for the benefit of society, for the human rights’ protection as well as for the respect of individual’s privacy and autonomy. The paper aims to summarize and critically evaluate the basic principles for the use of AI, with emphasis to the General Data Protection Regulation’s (GDPR) approach, concerning data subject’s consent, data protection principles and data subject’s rights in a context of ‘privacy by design’ architecture.

The advent of the Internet of Things (IoT) allows the Cyber-Physical System (CPS) components to communicate with other devices, and to interact with safety-critical systems, posing new research challenges in security, privacy, and reliability. Efficient power measurement in smart IoT devices has become one of the key research topics. In this paper, we design and develop a CPS to detect IoT security threats via behavioral power profiling of a heterogeneous wireless sensor device using a Raspberry Pi and a smartphone. Experimentation and verification have been conducted on a group of smart IoT devices with different test scenarios, including the device in an idle and active state with distributed denial-of-service (DDoS) and a man-in-the-middle (MitM) attack. We propose to use the device power consumption rate to predict and detect a security threat using statistical signal processing and multivariate regression model. The proposed system can detect a potential security threat with an average accuracy of 80% and a device high of 89%.

Unmanned merchant ships are ships that carry goods and engage in commercial activities without manual operations on board. In contrast to traditional merchant ships, unmanned merchant ships have great advantages in economy, society, and safety. However, we observe that commercial applications of unmanned ships are still at an exploratory stage. We believe that unmanned merchant ships will be widely adopted and popular in the near future. There is a need to have a good survey of current technologies, foundation, and obstacles of implementation of future unmanned merchant ships. Such a survey is beneficial to ship builders, researchers, owners, and students. Therefore, in this paper, we present a comprehensive survey of ship structure and technology, navigation, automation, algorithms, communication, Internet of Things (IOT), etc. We introduce traditional ships and analyze the role of crew and navigation operations on board. We summarize classifications, benefits, and core technologies of unmanned merchant ships. We review architecture, communication standards, security, essential technologies of IOT for unmanned merchant ships. Moreover, we introduce intelligent awareness, data fusion, applications, and E-navigation for unmanned merchant ships. We also present several future research directions at the end.

Orthogonal frequency division multiplexing (OFDM), usually with sufficient cyclic prefix (CP), has been widely applied in various communication systems. The CP in OFDM consumes additional resource and reduces spectrum and energy efficiency. However, channel estimation and signal detection are very challenging for CP-free OFDM systems. In this paper, we propose a novel artificial intelligence (AI)-aided receiver (AI receiver) for a CP-free OFDM system. The AI receiver includes a channel estimation neural network (CE-NET) and a signal detection neural network based on orthogonal approximate message passing (OAMP), called OAMP-NET. The CE-NET is initialized by the least-square channel estimation algorithm and refined by a linear minimum mean-squared error neural network. The OAMP-NET is established by unfolding the iterative OAMP algorithm and adding several trainable parameters to improve the detection performance. We first investigate their performance under different channel models through extensive simulation and then establish a real transmission system using a 5G rapid prototyping system for an over-the-air (OTA) test. Based on our study, the AI receiver can estimate time-varying channels with a single training phase. It also has great robustness to various imperfections and has better performance than those competitive algorithms, especially for high-order modulation. The OTA test further verifies its feasibility to real environments and indicates its potential for future communications systems.