Edge Layer Research Articles

Secure and Reliable Big-Data-Based Decision Making Using Quantum Approach in IIoT Systems

Nowadays, the industrial Internet of things (IIoT) and smart factories are relying on intelligence and big data analytics for large-scale decision making. Yet, this method is facing critical challenges regarding computation and data processing due to the complexity and heterogeneous nature of big data. Smart factory systems rely primarily on the analysis results to optimize production, predict future market directions, prevent and manage risks, and so on. However, deploying the existing classical solutions such as machine learning, cloud, and AI is not effective anymore. Smart factory systems and industries need novel solutions to sustain their development. On the other hand, with the fast development of quantum information systems (QISs), multiple sectors are studying the opportunities and challenges of implementing quantum-based solutions for a more efficient and exponentially faster processing time. To this end, in this paper, we discuss the implementation of quantum solutions for reliable and sustainable IIoT-based smart factory development. We depict various applications where quantum algorithms could improve the scalability and productivity of IIoT systems. Moreover, we design a universal system model where smart factories would not need to acquire quantum computers to run quantum algorithms based on their needs; instead, they can use quantum cloud servers and quantum terminals implemented at the edge layer to help them run the desired quantum algorithms without the need of an expert. To prove the feasibility of our model, we implement two real-world case studies and evaluate their performance. The analysis shows the benefits of quantum solutions in different sectors of smart factories.

The anticancer effects of graphene oxide against digestive system cancers: Short review

In recent years, advances in nanotechnology have enabled researchers in the field of diagnosis and treatment to use nanoparticles or complex multifunctional nanosystems for the diagnosis and targeted treatment of several diseases, including cancers. Graphene oxide is one of the most studied materials in the world. Due to its unique properties, it has been in the center of researchers' attention and has been named as a material for the future. Graphene oxide is made up of only carbon atoms, where each carbon atom is bonded to three other carbon atoms with sp2 hybridized orbitals, creating a honeycomb and hexagonal lattice. Graphene oxide is a layered carbon structure with oxygen-containing functional groups that are attached to both sides of the edge layer and also to the edges of the graphene oxide sheet. Due to its two-dimensional planar structure, chemical/mechanical stability, excellent sensitivity to light, excellent conductivity, high surface area and good biocompatibility, graphene oxide nanoparticles have been placed as a suitable candidate in the treatment of cancers. Graphene oxide can be used to make a drug delivery system for anti-cancer drugs such as paclitax, doxorubicin and methotrexate, and the drug can be directed to the desired location by loading it. This short review aimed to have an overview on the anticancer effects of graphene oxide on gastrointestinal cancers.

A Hierarchical Federated Learning Algorithm Based on Time Aggregation in Edge Computing Environment

Federated learning is currently a popular distributed machine learning solution that often experiences cumbersome communication processes and challenging model convergence in practical edge deployments due to the training nature of its model information interactions. The paper proposes a hierarchical federated learning algorithm called FedDyn to address these challenges. FedDyn uses dynamic weighting to limit the negative effects of local model parameters with high dispersion and speed-up convergence. Additionally, an efficient aggregation-based hierarchical federated learning algorithm is proposed to improve training efficiency. The waiting time is set at the edge layer, enabling edge aggregation within a specified time, while the central server waits for the arrival of all edge aggregation models before integrating them. Dynamic grouping weighted aggregation is implemented during aggregation based on the average obsolescence of local models in various batches. The proposed algorithm is tested on the MNIST and CIFAR-10 datasets and compared with the FedAVG algorithm. The results show that FedDyn can reduce the negative effects of non-independent and identically distributed (IID) data on the model and shorten the total training time by 30% under the same accuracy rate compared to FedAVG.

A TRUSTED COMPUTING RESOURCES OPTIMAL SCHEDULING ALGORITHM IN INDUSTRIAL INTERNET AND HEALTHCARE INTEGRATING DRL, BLOCKCHAIN AND END-EDGE-CLOUD

With the rapid development of the Internet of Things (IoT) and Internet technology, the product of the combination of the two, the Industrial Internet, has also received extensive attention and there are more and more research achievements related to the Industrial Internet. In the industrial Internet system, the communication network system composed of sensors, communication nodes, controllers and other intelligent devices can realize efficient and convenient data interaction between people and machines, providing an important infrastructure and technical support for industrial big data analysis and intelligent production. However, in the current industrial Internet system, industrial equipment users generally have the problem of low computing energy efficiency, and the collected industrial data has a high-security risk in the transmission, processing and other processes. At the same time, the size and scale of the industrial Internet equipment group is huge, and the lack of rational resource allocation leads to excessive waste of computing resources in the system, which is also a prominent problem of the current industrial Internet system. In response to the above questions, this paper, on the basis of reading a large number of documents, integrates the improved DRL algorithm, End-Edge-Cloud architecture and blockchain to form a new industrial Internet architecture. The architecture realizes computing offload through the three-tier structure of terminal layer, edge layer and cloud layer, and guarantees the security of industrial data through the decentralized feature of blockchain, ultimately achieving the goal of reducing energy consumption, computing overhead and trusted computing. In the architecture proposed in this paper, the dynamic unloading of industrial data and computing tasks is achieved through a three-tier architecture. The MDP is used to build an optimization problem model, and the improved DRL algorithm is used to iteratively solve the optimal computing resource scheduling strategy. The main research contents of this paper include (1) Using MDP to model optimization problems; (2) Propose an industrial Internet system architecture that integrates and improves DRL, “end edge cloud” and blockchain; (3) The MDP problem is solved iteratively based on deep reinforcement learning. The simulation results show that the proposed architecture has more advantages than the existing six architectures in terms of computing cost, equipment energy consumption and total working time.

Distributed artificial intelligence: Taxonomy, review, framework, and reference architecture

Artificial intelligence (AI) research and market have grown rapidly in the last few years, and this trend is expected to continue with many potential advancements and innovations in this field. One of the emerging AI research directions is Distributed Artificial Intelligence (DAI). It has been motivated by technological advances in communication, networking, and hardware, together with the nature of data being generated from connected, distributed, and diverse objects. DAI is expected to create a fertile environment for innovative, advanced, robust, and scalable approaches for AI supporting the vision of smart societies. In this paper, we explore state of the art on DAI and identify the opportunities and challenges of provisioning distributed AI as a service (DAIaaS). We provide a taxonomy and a comprehensive review covering the literature from 2016 to 2022. It comprises various aspects of DAI, including AI workflow, distribution paradigms, supporting infrastructure, management techniques, and applications. Based on the gained insights from the conducted review, we propose Imtidad, a framework for provisioning DAIaaS over the cloud, fog, and edge layers. We refine this framework and propose the Imtidad software Reference Architecture (RA) for designing and deploying DAI services. In addition, we extended the framework and developed a future networking infrastructure transformation framework, as it is one of the main enablers for DAI. This framework and RA can be used as guidance facilitating the transition to the future DAI, where DAI is decoupled from the design and development of smart applications. This paper, including the proposed framework, RA, taxonomy, and detailed review, is expected to have an extensive impact on DAI research and accelerate innovations in this area.

Influence of collisions on the validation of global gyrokinetic simulations in the edge and scrape-off layer of TCV

Understanding and predicting turbulent transport in the edge and scrape-off-layer (SOL) of magnetic confinement fusion devices is crucial for developing feasible fusion power plants. In this work, we present the latest improvements to the gyrokinetic turbulence code GENE-X and validate the extended model against experimental results in the TCV tokamak (“TCV-X21”). GENE-X features a full-f electromagnetic gyrokinetic model and is specifically targeted for edge and SOL simulations in diverted geometries. GENE-X can model the effect of collisions using either a basic Bhatnagar–Gross–Krook (BGK) or more sophisticated Lenard–Bernstein/Dougherty (LBD) collision operator. We present the results of a series of GENE-X simulations using the BGK or LBD collision models, contrasting them to collisionless simulations. We validate the resulting plasma profiles, power balance, and SOL heat flux against experimental measurements. The match to the experiment significantly improves with the fidelity of the collision model chosen. We analyze the characteristics of the turbulence and find that in almost all cases in the confined region the turbulence is driven by trapped electron modes (TEM). Both the simulations without collisions and those with the BGK collision operator do not accurately describe turbulence driven by TEMs. The more sophisticated LBD collision operator presents a minimum requirement for accurate gyrokinetic edge turbulence simulations.

EIDDM: Edge and Internet Layer Distributed DoS Threats Detection and Mitigation for Internet of Things Wireless Communications

EIDDM: Edge and Internet Layer Distributed DoS Threats Detection and Mitigation for Internet of Things Wireless Communications

Smart healthcare system using integrated and lightweight ECC with private blockchain for multimedia medical data processing.

Cloud-based Healthcare 4.0 systems have research challenges with secure medical data processing, especially biomedical image processing with privacy protection. Medical records are generally text/numerical or multimedia. Multimedia data includes X-ray scans, Computed Tomography (CT) scans, Magnetic Resonance Imaging (MRI) scans, etc. Transferring biomedical multimedia data to medical authorities raises various security concerns. This paper proposes a one-of-a-kind blockchain-based secure biomedical image processing system that maintains anonymity. The integrated Healthcare 4.0 assisted multimedia image processing architecture includes an edge layer, fog computing layer, cloud storage layer, and blockchain layer. The edge layer collects and sends periodic medical information from the patient to the higher layer. The multimedia data from the edge layer is securely preserved in blockchain-assisted cloud storage through fog nodes using lightweight cryptography. Medical users then safely search such data for medical treatment or monitoring. Lightweight cryptographic procedures are proposed by employing Elliptic Curve Cryptography (ECC) with Elliptic Curve Diffie-Hellman (ECDH) and Elliptic Curve Digital Signature (ECDS) algorithm to secure biomedical image processing while maintaining privacy (ECDSA). The proposed technique is experimented with using publically available chest X-ray and CT images. The experimental results revealed that the proposed model shows higher computational efficiency (encryption and decryption time), Peak to Signal Noise Ratio (PSNR), and Meas Square Error (MSE).

A Super-Efficient TinyML Processor for the Edge Metaverse

Although the Metaverse is becoming a popular technology in many aspects of our lives, there are some drawbacks to its implementation on clouds, including long latency, security concerns, and centralized infrastructures. Therefore, designing scalable Metaverse platforms on the edge layer can be a practical solution. Nevertheless, the realization of these edge-powered Metaverse ecosystems without high-performance intelligent edge devices is almost impossible. Neuromorphic engineering, which employs brain-inspired cognitive architectures to implement neuromorphic chips and Tiny Machine Learning (TinyML) technologies, can be an effective tool to enhance edge devices in such emerging ecosystems. Thus, a super-efficient TinyML processor to use in the edge-enabled Metaverse platforms has been designed and evaluated in this research. This processor includes a Winner-Take-All (WTA) circuit that was implemented via a simplified Leaky Integrate and Fire (LIF) neuron on an FPGA. The WTA architecture is a computational principle in a neuromorphic system inspired by the mini-column structure in the human brain. The resource consumption of the WTA architecture is reduced by employing our simplified LIF neuron, making it suitable for the proposed edge devices. The results have indicated that the proposed neuron improves the response speed to almost 39% and reduces resource consumption by 50% compared to recent works. Using our simplified neuron, up to 4200 neurons can be deployed on VIRTEX 6 devices. The maximum operating frequency of the proposed neuron and our spiking WTA is 576.319 MHz and 514.095 MHz, respectively.

Distributed PV Management Based on Edge-cloud Computing Model of Smart Distribution Transformer Terminal Unit

With the growing installation of distributed PV at a low voltage level and the resulting concerns about good power quality, power facilities see a greater need for energy management programs, especially for flexible distributed PV management. In the dawn of a new generation of ICT technologies, an edge-cloud computing model widely applied to smart distribution transformer terminal units (SD-TTU) is proposed to achieve lifecycle management of power facilities. Firstly, existing distribution management systems’ current practices and limitations are presented. According to the development trend, the reconstructed computing model involves three layers, and data interaction among the cloud layer, edge layer, and device layer is described. To support the huge scale of sensed data integration and transmission, the architecture of the edge node is established. Then the technical details for the distributed PV management are presented, including equipment configuration, main monitoring information, and specific procedures.

Latency-Aware Task Scheduling in Software-Defined Edge and Cloud Computing With Erasure-Coded Storage Systems

The collaborative edge and cloud computing system has emerged as a promising solution to fulfill the unprecedented high requirements of 5G application scenarios. Due to vendor variations, it is often difficult to manage hardware facilities in such a collaborative system. Moreover, the amount of data generated and tasks requested by end devices are increasing exponentially, which introduces storage and computation bottlenecks. To address these issues, a novel systematic framework called software-defined edge and cloud computing (SD-ECC) is designed to manage the underlying physical resources of edge and cloud layers via software.SD-ECC is combined with an erasure-coded storage system, for which a task scheduling problem is formulated by considering data access and task processing steps. Then, a joint data access and task processing (JDATP) algorithm is proposed to minimize the task response time including data access latency and task processing latency. A practical SD-ECC platform is developed on OpenStack, OpenDaylight, and Kubernetes. Experiments are conducted with real-world datasets. The experimental results demonstrated that our proposed JDATP algorithm can reduce 20.87% of the task response time and increase 14.16% of the remaining storage space on average by comparing it with alternative schemes.

Edge Computing-Assisted IoT Framework With an Autoencoder for Fault Detection in Manufacturing Predictive Maintenance

The Industrial Internet of Things (IIoT) enables intelligent predictive maintenance in smart manufacturing by incorporating IoT technologies, big data techniques, artificial intelligence, cloud computing and other ever-developing enabling technologies. Although a large body of research has been conducted on IIoT based predictive maintenance, most work focuses on addressing only a part of the problem. However, predictive maintenance involves an ecosystem from ingesting data from sensors to displaying the results on a dashboard for engineers to visualize. This paper proposes a complete and optimized IoT big data ecosystem embedded into a three-layer architecture, consisting of an edge layer, a cloud layer and an application layer. On top of the architecture, different layers are integrated seamlessly to address reliability, scalability, and reliability issues. An edge computing assisted autoencoder is introduced and enabled by being deployed in a distributed manner to improve both performance and efficiency. For practical interest, we also provide an application programming interface (API)-oriented implementation guideline for readers. To verify the proposed ecosystem, a real case study from industry is conducted to demonstrate the performance gain of edge computing-based IoT systems in conjunction with the autoencoder-based deep learning technique.

Lightweight multi-hop routing protocol for resource optimisation in edge computing networks

Data transmission over power line communication (PLC) infrastructure will proliferate lightweight Internet of Things (IoT) nodes in 5G and 6G networks. Consequently, a corresponding lightweight multi-hop routing protocol (LMRP) with reduced path loss and computational complexities will be required at the edges of PLC networks to connect the cloud sinks. Using a multi-layered system architecture, we present an LMRP for optimal routing and highlight the components of a smart PLC network comprising edge power pool orchestration, edge layer service provisioning, fog latency layer, and cloud resilient backbone. The LMRP reduces path loss and node failure states at the edge while optimising throughputs, minimum cost flow, and signal stability. A multi-hop deterministic testbed is designed and applied in three different locations to estimate path loss leveraging TelosB IoT node, Raspberry Pi (RPI) with NesC, and Java scripted logger application. Three different testbeds of varying path loss characteristics at the Federal University of Technology Owerri (FUTO) are used while the analysis was completed at Manchester Metropolitan University engineering LAB. The result of PL mitigation in Location 1 (sonic FUTO) shows 33.89%, 33.25%, and 32.77% with genetic algorithm (GA), particle swarm optimisation (PSO), and the proposed LMRP, respectively. In Location 2 (Old SEET Complex, FUTO), the PL obtained are 33.81%, 33.57%, and 32.62%, while Location 3 (New SEET Complex, FUTO) yields 33.65%, 33.41%, and 32.74% in PL mitigation for GA, PSO, and LMRP, respectively. Despite improved PL mitigation, the results also show that the proposed scheme offers a lightweight routing performance of at least 76.30% compared to similar schemes.

Built-up edge formation and flank wear evolution in conventional machining of AZ91 magnesium alloy

Magnesium alloys are largely desirable in numerous industries today because of their intrinsically attractive properties. Low density and high strength coupled with significant biocompatibility are the main advantages. Machining of magnesium alloys is challenging as a result of the persistent chip ignition risk as well as material build-up on the tool due to their softness. Therefore, magnesium alloys’ machinability has not been fully assessed in the available literature. The present work aims to address this aperture by assessing flank wear progression and the prevalence of adhesion-driven built-up edge (BUE) and built-up layer (BUL) in different tool coating materials during the dry short-interval conventional orthogonal turning of AZ91 magnesium alloy. A higher and lower combination of feed rates and cutting speeds were utilized for all tools to compare the possible range of maximum flank wear land width, average flank wear land width, and chip contact length while capturing and assessing said parameters through digital and optical microscopy. The data presented a large prevalence of adhesion wear across all tool coating types which is portrayed by the cyclic behaviour of the flank wear land width plots, which resulted from the repeated accumulation and wear of deposited workpiece material. The TiCN-coated tool presented the least amounts of BUL and BUE throughout both levels of testing, while the TiN-coated tool presented the largest amounts of average flank wear land width due to recurrent aberrative local deposition accumulation and extended streak line depositions. TiN-coated, Al2O3-coated, and uncoated tools presented higher average flank wear land width than the TiCN-coated tool by 8.5%, 3.1%, and 27.2% during level 1 cutting conditions and 84.8%, 21.3%, and 37.2% during level 2 cutting conditions. This implies the TiCN coating's increased resistance to elevated cutting temperatures, and in turn, reduced deposition of material through adhesion.

Correlation Anomaly Detection With Multiple Primary Attributes in Collaborative Device–Edge–Cloud Network

Anomaly detection is playing an increasingly important role in Internet of Things applications since anomalous events may cause some damage to the physical–social environment monitored by different kinds of smart object devices. In some cases, the occurrence of an anomalous event is caused by the fusion impact derived from several primary monitoring factors. Considering this, we propose a novel anomaly detection mechanism for the events with multiple decisive primary attributes in a collaborative device–edge–cloud architecture, in which a propagation and influence-based correlation is further explored in the edge layer for improving the detection efficiency. During the detection process, multiple primary attributes first cooperate to detect an anomaly in the edge layer in advance. If an anomaly occurs in the subregion managed by an edge device, social-aware interaction relationships between edge devices are further integrated to give a guidance on the detection of correlative anomaly in neighbor subregions. The cloud further analyzes the primary attributes information and the interaction relationship to determine the secondary attributes that are helpful in identifying the final anomaly. A large number of experiments show that our method is superior to the alternative methods in terms of energy consumption, detection time, and accuracy.

A fine-grained task scheduling mechanism for digital economy services based on intelligent edge and cloud computing

Digital economy is regarded countries as an inevitable choice to promote economic growth and provides new opportunities and new paths for the high-quality development of economy. For the Chinese market, the strong base behind cloud computing is the unstoppable development trend of the digital economy. In digital economy, the cloud as infrastructure becomes the base of the pyramid to build the digital economy. To relieve the pressure on the servers of the digital economy and develop a reasonable scheduling scheme, this paper proposes a fine-grained task scheduling method for cloud and edge computing based on a hybrid ant colony optimization algorithm. The edge computing task scheduling problem is described, and assumptions are set to simplify the difficulty of a scheduling solution. The multi-objective function is solved by using a hybrid ant colony optimization algorithm which solves computational problems by finding the optimal solution with the help of graphs. Ant colony optimization algorithm is easy to use and effective in scheduling problems. The proposed scheduling model includes an end-device layer and an edge layer. A terminal device layer consists of devices used by the clients that may generate computationally intensive tasks and are sometime uncapable to complete the tasks. The proposed scheduling policy migrates these tasks to a suitable place where they can be completed while meeting the latency requirements. The CPUs of the idle users on the end-device layer are used for other CPU-overloaded terminals. The simulation results, in terms of energy consumptions, and task scheduling delays, show that the task scheduling performance is better under the application of this method and the obtained scheduling scheme is more reasonable.

Filamentary transport in global edge-SOL simulations of ASDEX Upgrade

The understanding and the predictive capability for turbulence in the plasma edge and scrape-off layer (SOL) are crucial for the development of magnetic confinement fusion reactors. To this end, we characterise turbulent transport across the edge and SOL of the diverted ASDEX Upgrade tokamak in attached L-mode conditions by means of validated, global simulations. The collisionality is controlled by the divertor neutrals density, as their ionization increases the plasma density and decreases the temperature. The radial E×B particle and heat transport, quantified by effective diffusivities, rises strongly with collisionality. The modest increase in fluctuation amplitudes is not a sufficient explanation. The reason is shown to be the destabilisation of resistive drift-ballooning modes, resulting in larger phase shifts between the pressure and electrostatic potential. The transport varies both radially and poloidally. Due to its ballooning nature, radial transport is close to zero on the inboard mid-plane. On the outboard mid-plane, significant transport is driven in the SOL by large filaments (blobs) with amplitudes of up to 250% of the mean, propagating ballistically from the separatrix to the wall. This non-local transport leads to large radial variations of diffusivities, as they do not necessarily correlate with the local gradient. Ion temperature fluctuations in the plasma edge are shown to be involved in blob seeding at the separatrix, and are the largest in the SOL. Radial diffusivities peak at the top and bottom of the device, since gradients are flatter there due to the flux expansion while the cross-field flow is sustained by streamers—a feature which should be considered in mean-field transport modeling. The increase of SOL E×B transport with collisionality is likely fostered by the simultaneously decreasing radial electric field, resulting from a flattened electron temperature profile. Large amplitude blobs are a hazard for plasma facing components of fusion reactors, but they could be restrained by control of SOL collisionality.

An Efficient Security Model for Industrial Internet of Things (IIoT) System Based on Machine Learning Principles

This paper presents a security paradigm for edge devices to defend against various internal and external threats. The first section of the manuscript proposes employing machine learning models to identify MQTT-based (Message Queue Telemetry Transport) attacks using the Intrusion Detection and Prevention System (IDPS) for edge nodes. Because the Machine Learning (ML) model cannot be trained directly on low-performance platforms (such as edge devices),a new methodology for updating ML models is proposed to provide a tradeoff between the model performance and the computational complexity. The proposed methodology involves training the model on a high-performance computing platform and then installing the trained model as a detection engine on low-performance platforms (such as the edge node of the edge layer) to identify new attacks. Multiple security techniques have been employed in the second half of the manuscript to verify that the exchanged trained model and the exchanged data files are valid and undiscoverable (information authenticity and privacy) and that the source (such as a fog node or edge device) is indeed what it it claimed to be (source authentication and message integrity). Finally, the proposed security paradigm is found to be effective against various internal and external threats and can be applied to a low-cost single-board computer (SBC).

Secure-by-design serverless workflows on the Edge–Cloud Continuum through the Osmotic Computing paradigm

Nowadays, many Cloud companies adopt serverless computation based on the Function as a Service (FaaS) paradigm. Such an approach, built on the dynamic provisioning of serverless functions, has significantly altered the design of Cloud applications. Indeed, complex systems may now be viewed as a graph of serverless functions (i.e., workflow) distributed over the Cloud and Edge layers. However, building applications in such a complicated context and introducing security by design is still a non-trivial problem. Meanwhile, Osmotic Computing has advanced as a novel computational paradigm in recent years. This may serve as a suitable foundation for strengthening security in serverless applications throughout the Cloud–Edge Continuum. In this work, we rely on Osmotic Computing principles for modifying the architecture of OpenWolf, a novel serverless engine, and improving the execution time of ciphertext data by about 65%.

Morphological and mechanical properties of foamed thick-walled Wood-Plastic-Composite structures

Microcellular wood fiber reinforced polymers offer the possibility to reduce the use of fossil raw materials. In particular, thick-walled structures with thicknesses greater than 6 mm offer a high potential for weight savings. This study investigates the cell structures and mechanical properties of injection-molded test specimens. The influence of different thicknesses (6–10 mm) along with different chemical blowing agents (endothermic, exothermic) with varying dosages (0–2 wt%) is analyzed. The investigations reveal that exothermic chemical blowing agents form finer cells consistently to thin-walled structures than endothermic ones. Higher foaming agent content leads to higher pore fractions, with many small cells coalescing into a large open-pore cell network. The mechanical properties depend mainly on the pore content of the sample. The specific tensile properties deteriorate with the use of chemical blowing agents (CFA), whereas the sandwich structure produced with compact edge layers has a positive influence on the specific flexural properties.