Heterogeneous Devices Research Articles

IBUST: An intelligent behavioural trust model for securing industrial cyber-physical systems

To meet the demand of the world’s largest population, smart manufacturing has accelerated the adoption of smart factories—where autonomous and cooperative instruments across all levels of production and logistics networks are integrated through a Cyber-Physical Production System (CPPS). However, these networks are comprised of various heterogeneous devices with varying computational power and memory capabilities. As a result, many secure communication protocols – that demand considerably high computational power and memory – can not be verbatim employed on these networks, and thereby, leaving them more vulnerable to security threats and attacks over conventional networks. These threats can largely be tackled by employing a Trust Management Model (TMM) by exploiting the behavioural patterns of nodes to identify their trust class. In this context, ML-based models are best suited due to their ability to capture hidden patterns in data, learning and improving the pattern detection accuracy over time to counteract and tackle threats of a dynamic nature, which is absent in most of the conventional models. However, among the existing ML-based solutions in detecting attack patterns, many of them are computationally expensive, require a long training time, and a considerably large amount of training data—which are seldom available. An aid to this is the association rule learning (ARL) paradigm, whose models are computationally inexpensive and do not require a long training time. Therefore, this paper proposes an ARL-based intelligent Behavioural Trust Model (iBUST) for securing the CPPS. For this intelligent TMM, a variant of Frequency Pattern Growth (FP-Growth), called enhanced FP-Growth (EFP-Growth) algorithm is developed by altering the internal data structures for faster execution and by developing a modified exponential decay function (MEDF) to automatically calculate minimum supports for adapting trust evolution characteristics. In addition, a new optimisation model for finding optimum parameter values in the MEDF and an algorithm for transmuting a 1D quantitative feature into a respective categorical feature are developed to facilitate the model. Afterwards, the trust class of an object is identified employing the Naïve Bayes classifier. This proposed model is evaluated on a trust evolution-supported experimental environment along with other compared models taking a benchmark dataset into consideration, where it outperforms its counterparts.

Network management schemes for IoT environment towards 6G: A comprehensive review

Usage of smart devices, expansion of Internet of Things (IoT) network, and its adoption in variety of applications grows exponentially day-by-day. It leads to various issues such as network privacy, security management, unreliable network management (NM) scheme, resource utilization, energy utilization, decentralization, transparency, and data interoperability. Development of the NM platform/framework for secure data dissemination in IoT networks is tedious due to heterogeneity and imbalanced IoT protocol stack. Recently, traditional network scheduling methods, Software Defined Networks (SDN), Network Function Virtualization (NFV), and hybrid approaches are widely used for managing the IoT network. But, most of these approaches are controlled by a centralized server, which may not be appropriate due to the heterogeneity of the IoT devices. Motivated from the aforementioned discussion, In this paper, we present a systematic survey on various NM schemes used to secure data dissemination for the IoT environment. We present a detailed taxonomy of all the existing NM schemes for IoT networks. Pros and cons of the traditional NM methods applicable for IoT environment are discussed in detail. The paper discusses a state-of-art survey on integrating blockchain (BC) technology in the existing IoT NM techniques. In this paper, our prime focus is to understand the potential of BC for enabling secured IoT management. Finally, a case study on BC implication for Unmanned Aerial Vehicle (UAV) network using 6G is also discussed along with future directions.

In vitro oxygen imaging of acellular and cell-loaded beta cell replacement devices

Type 1 diabetes (T1D) is an autoimmune disease that leads to the loss of insulin-producing beta cells. Bioartificial pancreas (BAP) or beta cell replacement strategies have shown promise in curing T1D and providing long-term insulin independence. Hypoxia (low oxygen concentration) that may occur in the BAP devices due to cell oxygen consumption at the early stages after implantation damages the cells, in addition to imposing limitations to device dimensions when translating promising results from rodents to humans. Finding ways to provide cells with sufficient oxygenation remains the major challenge in realizing BAP devices’ full potential. Therefore, in vitro oxygen imaging assessment of BAP devices is crucial for predicting the devices’ in vivo efficiency. Electron paramagnetic resonance oxygen imaging (EPROI, also known as electron MRI or eMRI) is a unique imaging technique that delivers absolute partial pressure of oxygen (pO2) maps and has been used for cancer hypoxia research for decades. However, its applicability for assessing BAP devices has not been explored. EPROI utilizes low magnetic fields in the mT range, static gradients, and the linear relationship between the spin–lattice relaxation rate (R1) of oxygen-sensitive spin probes such as trityl OX071 and pO2 to generate oxygen maps in tissues. With the support of the Juvenile Diabetes Research Foundation (JDRF), an academic-industry partnership consortium, the “Oxygen Measurement Core” was established at O2M to perform oxygen imaging assessment of BAP devices originated from core members’ laboratories. This article aims to establish the protocols and demonstrate a few examples of in vitro oxygen imaging of BAP devices using EPROI. All pO2 measurements were performed using a recently introduced 720 MHz/25 mT preclinical oxygen imager instrument, JIVA-25™. We began by performing pO2 calibration of the biomaterials used in BAPs at 25 mT magnetic field since no such data exist. We compared the EPROI pO2 measurement with a single-point probe for a few selected materials. We also performed trityl OX071 toxicity studies with fibroblasts, as well as insulin-producing cells (beta TC6, MIN6, and human islet cells). Finally, we performed proof-of-concept in vitro pO2 imaging of five BAP devices that varied in size, shape, and biomaterials. We demonstrated that EPROI is compatible with commonly used biomaterials and that trityl OX071 is nontoxic to cells. A comparison of the EPROI with a fluorescent-based point oxygen probe in selected biomaterials showed higher accuracy of EPROI. The imaging of typically heterogenous BAP devices demonstrated the utility of obtaining oxygen maps over single-point measurements. In summary, we present EPROI as a quality control tool for developing efficient cell transplantation devices and artificial tissue grafts. Although the focus of this work is encapsulation systems for diabetes, the techniques developed in this project are easily transferable to other biomaterials, tissue grafts, and cell therapy devices used in the field of tissue engineering and regenerative medicine (TERM). In summary, EPROI is a unique noninvasive tool to experimentally study oxygen distribution in cell transplantation devices and artificial tissues, which can revolutionize the treatment of degenerative diseases like T1D.

Energy-efficient Personalized Federated Search with Graph for Edge Computing

Federated Learning (FL) is a popular method for privacy-preserving machine learning on edge devices. However, the heterogeneity of edge devices, including differences in system architecture, data, and co-running applications, can significantly impact the energy efficiency of FL. To address these issues, we propose an energy-efficient personalized federated search framework. This framework has three key components. Firstly, we search for partial models with high inference efficiency to reduce training energy consumption and the occurrence of stragglers in each round. Secondly, we build lightweight search controllers that control the model sampling and respond to runtime variances, mitigating new straggler issues caused by co-running applications. Finally, we design an adaptive search update strategy based on graph aggregation to improve personalized training convergence. Our framework reduces the energy consumption of the training process by lowering the training overhead of each round and speeding up the training convergence rate. Experimental results show that our approach achieves up to 5.02% accuracy and 3.45× energy efficiency improvements.

A review on client selection models in federated learning

AbstractFederated learning (FL) is a decentralized machine learning (ML) technique that enables multiple clients to collaboratively train a common ML model without them having to share their raw data with each other. A typical FL process involves (1) FL client(s) selection, (2) global model distribution, (3) local training, and (4) aggregation. As such FL clients are heterogeneous edge devices (i.e., mobile phones) that differ in terms of computational resources, training data quality, and distribution. Therefore, FL client(s) selection has a significant influence on the execution of the remaining steps of an FL process. There have been a variety of FL client(s) selection models proposed in the literature, however, their critical review and/or comparative analysis is much less discussed. This paper brings the scattered FL client(s) selection models onto a single platform by first categorizing them into five categories, followed by providing a detailed analysis of the benefits/shortcomings and the applicability of these models for different FL scenarios. Such understanding can help researchers in academia and industry to develop improved FL client(s) selection models to address the requirement challenges and shortcomings of the current models. Finally, future research directions in the area of FL client(s) selection are also discussed.This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence

Modeling and Analysis of Malware Propagation for IoT Heterogeneous Devices

With the rapid development and application of Internet of Things (IoT) technology, a large number of various types of IoT devices with security vulnerabilities have been connected to the public network. Attackers only need to use malware to successfully infect a few types of IoT devices to generate large-scale propagation in the network and capture a large number of broilers, which in turn lays the foundation for subsequent attacks. In this study, an innovative malware propagation model [i.e., IoT-susceptible-latent-propagated-recovered (IoT-SLPR)] based on epidemiological theory is developed for malware propagation in multiple IoT heterogeneous devices. The proposed model incorporates the propagation mechanism of malware in real IoT networks, and considers heterogeneous devices with different infection rates and recovery rates. This study calculates and proves the basic reproductive rate of malware and the stability of equilibrium points for different types of devices. In addition, this study designs a dynamic recovery rate optimal defense strategy, which comprehensively considers the cost and the overall infection rate of the device. The experimental results show that the proposed propagation model can effectively reflect the propagation dynamics of malware in multiple heterogeneous devices, and the achieved dynamic recovery rate defense strategy performs better than the static strategy regarding the overall results.

Joint Gradient Sparsification and Device Scheduling for Federated Learning

Federated learning is an attractive distributed learning framework where edge server coordinates edge devices to collaboratively train an artificial intelligence (AI) model while training data stays in edge devices. However, implementing federated learning in an energy-efficient way still faces some problems in practice: i) uploading high-dimensional parameters of edge devices is energy-costly and causes communication congestion; ii) The heterogeneity of edge devices, i.e. non-i.i.d. data, different software/hardware configurations, and time-varying channel conditions, affects training accuracy and convergence rate of federated learning. To address these issues, in this paper, we aim at minimizing energy use of the heterogeneous edge devices while maintaining the learning performance. Note that the effects of sparsification and the number of participating devices are highly coupled, and they have opposite effects on both learning performance and energy consumption. Therefore, the gradient sparsification and device scheduling should be jointly optimized for federated learning. We consider the case where a fraction of heterogeneous devices are scheduled to upload their sparsified local gradients for global aggregation at the server. First, a tradeoff between the sparsification and the number of participating devices is revealed through the convergence rate analysis. Then, a joint optimization problem is formulated to minimize energy consumption under the convergence guarantee. An efficient algorithm is proposed to find the globally optimal solution of the considered non-convex problem. Simulations verify the efficiency of the proposed scheme.

Data-Driven Participant Selection and Bandwidth Allocation for Heterogeneous Federated Edge Learning

Federated edge learning (FEEL) is a rapidly growing distributed learning technique for next-generation wireless edge systems. Smart systems across various application domains face challenges, such as data heterogeneity, limited wireless resources, and device heterogeneity, which necessitate intelligent participant selection schemes that accelerate convergence rates. Consequently, this article presents joint participant selection and bandwidth allocation schemes to address these challenges. First, we formulate an optimization problem that considers communication and computation latencies, as well as imbalanced data distribution, while meeting round deadlines and bandwidth constraints. To address the combinatorial problems of participant selection, we employ a relaxation method followed by a proposed priority selection algorithm to select near-optimal participants. The proposed algorithm initially prioritizes participants with larger datasets, effective channel states, and better CPU speeds. To address data heterogeneity, we propose a randomized deadline-controlling algorithm that diversifies updates by allowing the edge server to include different participants with fewer data samples in training rounds. The proposed algorithms offer near-optimal performance compared to the brute-force method. Experiments demonstrate that our proposed scheme accelerates the convergence rate by up to 55% under extensive non-IID settings compared to benchmarks. Furthermore, the deadline-controlling algorithm improves performance at high levels of data heterogeneity, resulting in faster FEEL systems.

Communitychain: Toward a Scalable Blockchain in Smart Home

Rapid development of smart homes in recent years has led to the production of an increasing number of smart devices that meet people’s daily needs. However, these smart devices may belong to different vendors and provide different functions and services, making it difficult to apply a centralized solution to a smart home system. Although blockchain has the potential to address the potential problems of smart homes because of its features such as nontampering, decentralization, and security, scalability remains a key challenge when integrating blockchain and smart homes, which is mainly reflected in the horizontal expansion, throughput and latency of the system. To address this challenge, Communitychain—a new scalable blockchain architecture suitable for smart home systems composing heterogeneous devices—is proposed in this study. A community model based on sharding that enables the architecture to adapt to an increase in the number of nodes is designed, which includes new miner nodes and leader node selection algorithms. Efficient cross-shard routing and sideBlock schemes are adopted to improve the throughput and latency and enhance the scalability of the system. Furthermore, lightweight authorization and authentication processes running in each community ensure secure access to device resources. The experimental results indicate that the proposed blockchain architecture can effectively improve the scalability of blockchain-based smart homes and adapt well to the system dynamics.

Navigating Routing Challenges in Internet of Things (IoT): A Focus on Routing Protocols and Wireless Sensor Networks

This paper states the routing issues in the Internet of Things (IoT). The main point of focus is especially on routing protocols in IoT and wireless sensor networks (WSN). “Everything which is connected to the internet is alive”, is going to be the new rule for the future. The future is the Internet of Things (IoT), the world is changing every day with new technologies and inventions. The Internet of Things (IoT) has known as a future scenario of technology in human life. IoT expands the concept of the Internet from a network of rather homogenous devices such as computers to a network of heterogeneous devices such as home appliances and the conventional network is mostly replaced with IoT. Devices are the main users of the IoT framework an important part is IEEE 802.15.4, a standard that operates at the PHY layer and MAC layer of IoT devices. The IoT Ecosystem includes a large number of integrated heterogeneous low-power, low-cost devices, and secure connections are needed between them [15]. They communicate with each other to gather, share and forward the information in a multi-hop manner, this requires a complex routing between them. WSN, new technology in the IoT system comprised of a large collection of sensors, a mesh network can be used to individually gather data and send data through a router to the internet in an IoT system [6]. an enormous number of sensors are used and its growth is also fast, they are collecting inexperienced data and transforming it into meaningful and useful information. considering the full usability of these devices and a huge number of them, produce routing challenges [16]. Routing in IoT is the most critical part because we are working with devices that have restricted device resources. the optimum goal is to have enhanced routing protocols and efficient performance with such limited resource devices [15, 18].

Overview of the Integration of Communications, Sensing, Computing, and Storage as Enabling Technologies for the Metaverse over 6G Networks

The metaverse, as an envisioned paradigm of the future internet, aims to establish an immersive and multidimensional virtual space in which global users can interact with one another, as in the real world. With the rapid development of emerging technologies—such as digital twins (DT), blockchain, and artificial intelligence (AI)—the diverse potential application scenarios of the metaverse have attracted a great deal of research attention and have created a prosperous market. The demand for ubiquitous communications, pervasive sensing, ultra-low latency computing, and distributed storage has consequently surged, due to the massive heterogeneous devices and data in the metaverse. In order to achieve the metaverse, it is essential to establish an infrastructure system that integrates communications, sensing, computing, and storage technologies. Information about the physical world can be obtained by pervasive sensing, computing resources can be scheduled in a reasonable manner, quick data access can be achieved through the coordination of centralized and distributed storage, and, as the bridge, mobile communications systems connect communications, sensing, computing, and storage in a new system, which is the integration of communications, sensing, computing, and storage (I-CSCS). Following this trend, this paper discusses the requirements of the metaverse for spectrum resources, ultra-reliable transmission, seamless coverage, and security protection in wireless mobile communications systems, and analyzes the fundamental supporting role of the sixth-generation mobile communications system (6G) in the metaverse. Then, we explore the functions and roles of the integrated sensing and communications technologies (ISAC), as well as the integration of communications, computing, and storage technologies for the metaverse. Finally, we summarize the research directions and challenges of I-CSCS in the metaverse.

(Gordon E. Moore Medal for Outstanding Achievement in SSS&T) Moore’s Law Sustained by Non-Lithographic Technologies

Since G. Moore’s historical observation [1] that the functionality per chip (bits, transistors) as well as MPU performance (clock frequency in MHz × instructions per clock = millions of instructions per second) doubles every 1.5 to 2 years, the semiconductor industry has continued to grow to over 600 G$ in 2022 [2].The IRDS 2022 Roadmap [3] catches the scaling challenges faced for the upcoming decades by the term ‘3D Power Scaling’. This period is the third in a sequence of eras that started early on with straightforward geometrical scaling by continuous shortening of the wavelengths (from regular UV to deep UV/immersion) used in the lithographic patterning of planar transistor structures (Fig. 1a). In the second, almost past ‘equivalent scaling’ era, new superior material properties and critical dimensions nearing single-digit nanometer values could still be realized by cost-effective technology solutions, often in spite of the delay in (EUV) lithography solutions. Ever more complex device architectures requiring extreme edge placement accuracy, layer conformality and shape fidelity in all processing steps (deposition, etching) could be fully 3D-integrated into vertical intra- and inter-chip concepts (Fig. 1b) thus alleviating the need for new capital-intensive lithography tools capable of higher resolution (~ 40% of total tool costs).Figure 2 illustrates the development in the past 65 years of the non-lithographic technologies that have sustained Moore’s Law, as driven by continuous downscaling of CMOS devices. Two extra trends have set in half way (~1993): single-wafer processing and heterogeneous device integration. The latter development, often coined as More than Moore, a term that was introduced in the 2005 ITRS Roadmap, and refers to the silicon-compatible integration of devices with functionalities that do not necessarily scale according to Moore's Law, but provide additional value in different ways. The More-than-Moore approach allows for the incorporation of non-digital functionalities (e.g., RF communication, passive components, power control, sensors, actuators) that can flexibly move from the system board-level into the package (SiP) or onto the chip (SoC).We will discuss a kaleidoscopic view of the author’s activities in Rapid Thermal Processing, RIE etching for TSVs, RF-SIP integration of passives, and Atomic Layer Processing (deposition, etch, and cleaning). In sub-10 nm scaling and fabrication of 3D architectures, especially the techniques of ALD and ALE have manifested to cost-effectively bridge the record incubation time needed to bring EUV technology from prototype to commercial use. More importantly, these unique techniques can be used to create advanced devices in dedicated isotropic (thermal and radical-enhanced) and anisotropic (directional and ion-enhanced) processing modes. Here, energetic species (radicals and/or ions in a plasma) are used in one or two steps, with the ions yielding anisotropic profiles (used in FinFET logic and 3D NAND memory), and neutrals and radicals yielding isotropic profiles used to deposit or etch the features in horizontal nanowires, nanosheets and ‘forksheets’ in GAA-FETs.1. G. Moore, Cramming more components onto integrated circuits, Electronics 38(8), (1965) 114-118.2. https://www.semiconductors.org/global-semiconductor-sales-increase-24-year-to-year-in-october-annual-sales-projected-to-increase-26-in-2021-exceed-600-billion-in-2022/3. International Roadmap for Devices and Systems (IRDS™) 2022 Edition - IEEE IRDS™. Fig. 1. a) Different scaling ages for device manufacturing; b) the planar-to-GAA transition; after [3]. Fig. 2. Historic and future perspectives of non-lithographic key technologies in IC scaling. Figure 1

Distributed Fixed-Time Energy Management for Port Microgrid Considering Transmissive Efficiency

To enhance the efficiency of a port microgrid, this paper proposes an energy management method and a topology construction mechanism considering the convergence rate and information transmission distances, respectively. Firstly, a distributed fixed-time energy management method is proposed to solve an energy management problem in a known time and guarantee the efficiency of the port microgrid. Secondly, to address the challenge of heterogeneous devices with multiple communication protocols, information exchange between different devices is facilitated through a polymorphic network. To obtain a connected communication topology that can ensure the implementation of the distributed energy management method, a connected networking mechanism is proposed. This mechanism minimizes the total communication distance to reduce the effect of the information transmission distance on communication effectiveness. Finally, the effectiveness of both algorithms is demonstrated by simulation, and the advantages of the distributed fixed-time energy management method on the convergence rate are reflected through a comparison with other methods.

Fog computing out of the box: Dynamic deployment of fog service containers with TOSCA

AbstractThe conventional cloud‐centric Internet of Things (IoT) application fails to meet the latency requirement of time‐critical applications. The idea of edge and fog computing arrived to distribute workloads across the fog devices located in the local area. However, achieving seamless interoperability, platform independence, and automatic deployment of services becomes the major challenge over heterogeneous fog devices. This paper proposes an integrated and standards‐based fog computing federation framework, FogDEFT, that adapts OASIS–Topology and Orchestration Specification for Cloud Applications (TOSCA) for service deployment in fog. The framework standardizes the distributed application design with TOSCA Service Template to deploy Docker Containers in Swarm mode and manages interoperability over heterogeneous fog devices. The framework uses a lightweight TOSCA compliant orchestrator to dynamically deploy various fog applications (user‐developed services) on the fly.

Deep Learning-Based Geomagnetic Navigation Method Integrated with Dead Reckoning

Accurate location information has significant commercial and economic value as they are widely used in intelligent manufacturing, material localization and smart homes. Magnetic sequence-based approaches show great promise mainly due to their pervasiveness and stability. However, existing geomagnetic indoor localization methods are facing the problems of location ambiguity and feature extraction deficiency, which will lead to large localization errors. To address these issues, we propose a coarse-to-fine geomagnetic indoor localization method based on deep learning. First, a multidimensional geomagnetic feature extraction method is presented which can extract magnetic features from spatial and temporal aspects. Then, a hierarchical deep neural network model is devised to extract more accurate geomagnetic information and corresponding location clues for more accurate localization. Finally, localization is achieved through a particle filter combined with IMU localization. To evaluate the performance of the proposed methods, we carried out several experiments at three trial paths with two heterogeneous devices, Vivo X30 and Huawei Mate30. Experimental results demonstrate that the proposed algorithm can achieve more accurate localization performance than the state-of-the-art methods. Meanwhile, the proposed algorithm has low cost and good pervasiveness for different devices.

When edge intelligence meets cognitive buildings: The COGITO platform

Future buildings are complex systems that aim at improving the quality of life of their inhabitants and increasing safeness, security, and efficiency. In order to reach these goals, they require their own self-management and self-adaptation capabilities, thus becoming cognitive entities. However, developing cognitive buildings that exploit advanced Artificial Intelligence (AI) techniques in a distributed fashion is still a challenge. Indeed, they need to continuously collect and process a variety of environmental parameters, learn and predict the users’ needs and preferences, and then control a large number of heterogeneous devices. These operations may leverage the dynamic availability of edge and cloud computing resources.This paper discusses the challenges and requirements tied to the development of cognitive buildings and proposes COGITO, an Internet of Things platform specifically suited to design and implement cognitive buildings. COGITO enables edge intelligence and the exploitation of all the available resources in the computing continuum between the edge and the cloud. All of this is to support holistic services that guarantee energy efficiency, security, and user comfort. For validation purposes, a case study involving the use of COGITO in an office building is presented, together with a performance evaluation analysis.

Neural Plasticity Changes Induced by Motor Robotic Rehabilitation in Stroke Patients: The Contribution of Functional Neuroimaging.

Robotic rehabilitation is one of the most advanced treatments helping people with stroke to faster recovery from motor deficits. The clinical impact of this type of treatment has been widely defined and established using clinical scales. The neurofunctional indicators of motor recovery following conventional rehabilitation treatments have already been identified by previous meta-analytic investigations. However, a clear definition of the neural correlates associated with robotic neurorehabilitation treatment has never been performed. This systematic review assesses the neurofunctional correlates (fMRI, fNIRS) of cutting-edge robotic therapies in enhancing motor recovery of stroke populations in accordance with PRISMA standards. A total of 7, of the initial yield of 150 articles, have been included in this review. Lessons from these studies suggest that neural plasticity within the ipsilateral primary motor cortex, the contralateral sensorimotor cortex, and the premotor cortices are more sensitive to compensation strategies reflecting upper and lower limbs' motor recovery despite the high heterogeneity in robotic devices, clinical status, and neuroimaging procedures. Unfortunately, the paucity of RCT studies prevents us from understanding the neurobiological differences induced by robotic devices with respect to traditional rehabilitation approaches. Despite this technology dating to the early 1990s, there is a need to translate more functional neuroimaging markers in clinical settings since they provide a unique opportunity to examine, in-depth, the brain plasticity changes induced by robotic rehabilitation.

A Comprehensive Empirical Study of Heterogeneity in Federated Learning

Federated learning (FL) is becoming a popular paradigm for collaborative learning over distributed, private datasets owned by non-trusting entities. FL has seen successful deployment in production environments, and it has been adopted in services such as virtual keyboards, auto-completion, item recommendation, and several IoT applications. However, FL comes with the challenge of performing training over largely heterogeneous datasets, devices, and networks that are out of the control of the centralized FL server. Motivated by this inherent challenge, we aim to empirically characterize the impact of device and behavioral heterogeneity on the trained model. We conduct an extensive empirical study spanning nearly 1.5K unique configurations on five popular FL benchmarks. Our analysis shows that these sources of heterogeneity have a major impact on both model quality and fairness, causing up to 4.6× and 2.2× degradation in the quality and fairness, respectively, thus shedding light on the importance of considering heterogeneity in FL system design.

A collaborative management mechanism for UPIDs operation against MAD attacks

With the user-side heterogeneous power IoT devices (UPIDs) integrating into power systems, the cyber security of UPIDs brings a hidden danger to the power system due to their high-wattage terminals, and there is a new attack plan that can leverage such terminal botnets in order to manipulate the power demand, referred to as Manipulation of demand (MAD) attacks. To this end, a collaborative management mechanism for UPIDs operation is needed to mitigate the malicious impact of MAD attacks. Firstly, taking into account the vulnerability of the deployed cyber security defense measures, a cyber security vulnerability assessment method for UPIDs is proposed, which provides a criterion for the security control strategy. Besides, the security control strategy is constructed considering the security factors in the cyber domain and the physical domain, ensuring the reliability of power systems. The effectiveness of the proposed collaborative management mechanism is verified by simulations on a modified IEEE 14-bus system, highlighting the importance of the active defense scheme against MAD attacks.

Crack mechanism correlated with Sn grain orientations on Ni metal surface subjected to 1000 thermal shocks

With the rapid development of 3D packaging, it is of great practical significance to understand the failure mechanism of solder joints in heterogeneous integration devices. In this study, the grain orientation and microstructure evolution of SAC305 solder joints were investigated in thermal shock tests for fan-out wafer level packaging. The heterogenous evolution of microstructure in Sn–3Ag-0.5Cu (SAC305) solder joints was investigated using electron backscattered diffraction (EBSD) and nanoindentation analysis. When the thermal shock reaches 1000 cycles, the local deformation caused the heterolattices in the joint to rotate toward the corners due to the thermal expansion. During deformation, sub-grain formation is accompanied by an increase in dislocations, the accumulation of plastic slip, and the activation of slip systems. The energy stored by this deformation is released in the form of cracks. The work in this paper provides new insights into the early stages of grain orientation, microstructural deformations and the accumulation of stored energy associated with these deformations, providing a scientific basis for the application and failure analysis of large size fan-out welded joints in complex working conditions.