Search

In vehicular edge computing, both edge and cloud can provide computing services (i.e., tasks). The edge can reduce vehicular task delay by processing data nearby, but is resource‐constrained and cannot handle too many tasks simultaneously. The resource‐rich cloud can handle massive tasks, but is far from users and has low quality of service and energy efficiency. Currently, some work has been done on task offloading for edge‐cloud collaboration. However, either the collaboration among multiple devices is not considered and the load is easily imbalanced; or edge‐cloud collaboration is required for offloading decisions with too many parameters, affecting tasks to be processed efficiently. To address these issues, we learn from the swarm intelligence evolution of sparrow foraging, improve a sparrow search algorithm by integrating three strategies of flyer refine producer update, sin/cos perturbation follower update, and adaptive adjustment agitator update, to optimize the vehicular task offloading location for edge‐cloud collaboration. Furthermore, we combine delay relaxation variables and delay‐energy penalty operator to design a lightweight heuristic task offloading algorithm, and greedily compare the task preoffload location sets with different delay constraints based on the improved sparrow search algorithm, to obtain the optimal task offloading with integrated delay and energy. Simulation experiments verify that our approach can improve the algorithm's optimization accuracy, convergence speed, and robustness. Moreover, the average task delay, total task energy consumption, and load balance degree of our approach outperform the five benchmark algorithms, and can comprehensively optimize task delay and energy consumption, and achieve edge devices load balancing. [ABSTRACT FROM AUTHOR]

With the fast development of Mobile Internet, data traffic generated by end devices is anticipated to witness substantial growth in the future years. However, processing tasks locally will cause latency due to the limited resources of the end devices. Edge-cloud collaboration, an effective solution for latency-sensitive applications, is attracting greater attention from both industry and academia. It combines the advantages of the cloud center with abundant computing resources and edge nodes with low-latency capabilities. In this paper, we propose a two-stage task offloading framework with edge-cloud collaboration to assist end devices processing latency-sensitive tasks either on the edge servers or in the cloud center. As for homogeneous task offloading, in the first stage, the competitive end devices offload tasks to the edge gateways. We formulate the selfish task offloading problem among end devices as a potential game. In the second stage, the edge nodes request resources from the cloud center to process end devices tasks due to their limited resources. Then, we consider the heterogeneous task offloading problem and use intelligent optimization algorithm to obtain the optimal offloading strategy. Simulation results show that the service prices of edge nodes influence the decisions and task offloading costs of end devices. We also verify the intelligent optimization algorithm can achieve optimal performance with low complexity and fast convergence. [ABSTRACT FROM AUTHOR]

Leveraging the natural cross-platform accessibility, Web3D has shown wide application prospects in digital content fields with a browser as a carrier, including virtual reality, 3D modeling and visualization, to name a few. Nonetheless, constructing a diversified and shared immersive Web3D service platform is still in its infancy. How to ensure seamless communication, computing, and storage services has become a core challenge in dealing with the dynamic changes in network environments, limitations in browser-based rendering capabilities, and the impending surge in interactive demands. The progression of cutting-edge technologies such as mobile Internet and 5G networks, coupled with the emergence of edge computing, has propelled Web3D service modes from solitary terminal or cloud-based rendering toward a collaborative edge-cloud paradigm. Such a paradigm effectively integrates the resources from the remote cloud, edge nodes, and end users, thereby fulfilling the requirements of rapid response, high-speed data transmission, and real-time interaction. Focusing on the application of edge-cloud collaboration frameworks in immersive Web3D services, how this computing mode tackle the intricate issues encountered in establishing Web3D platforms was discussed. Firstly, a Web3D scenario-oriented edge-cloud collaboration framework and the key challenges faced in constructing immersive Web3D service platforms were discussed. Secondly, the optimization strategies for edge-cloud collaboration-driven Web3D were discussed from the perspectives of resource scheduling, transmission and loading, pre-caching schemes, and model lightweighting, respectively. Finally, open research questions were explored. [ABSTRACT FROM AUTHOR]

To realize transparent monitoring and resilience improvement of low- voltage distribution network, both the data acquisition scope and frequency have been greatly expanded. Cloud-edge collaboration leverages the edge server's real-time response capabilities and the cloud server's robust data processing power to enhance the performance of high-frequency data acquisition processing. Nonetheless, it continues to confront challenges such as the entanglement of optimization variables, the presence of uncertain information, and a lack of awareness regarding acquisition frequencies. In this paper, we propose a machine learning-based cloud-edge collaborative data processing optimization algorithm to minimize the weighted sum of data processing delay and device energy consumption for distribution network resilience improvement. The joint optimization problem is decoupled into device-edge data offloading subproblem and edge-cloud data splitting subproblem, which are solved by the proposed upper confidence bound (UCB) based frequency-aware device-edge data offloading optimization algorithm and the exponential-weight algorithm for exploration and exploitation (EXP3) based edge-cloud data splitting optimization algorithm, respectively. Simulation results show that the proposed algorithm is superior to existing algorithms in performances of energy consumption and total processing delay. [ABSTRACT FROM AUTHOR]

With the continuous maturity and adoption of mobile devices enabled by wireless communication technology, people are more apt to record their sport exercise data or healthcare data through various lightweight and smart devices, e.g., mobile phones and smart watches. Meanwhile, massive sport data or healthcare data keep being produced with time, which forms a main source of big healthcare data. Deep mining and analysis of such healthcare data are of positive significance for accurately recognizing the real-time health condition of mobile users and further recommend appropriate sport items to them. However, traditional centralized healthcare data mining and recommendation approaches require mobile users to transmit their health data collected by mobile devices to a remote cloud platform, which often involves heavy data transmissions from mobile devices to cloud platform. As a consequence, the transmission cost is high and the time delay is long. Moreover, long-distance data transmissions are prone to disclose user privacy. Considering these limitations, we bring forth a novel time-efficient and privacy-preserving healthcare data integration and mining approach for sport item recommendation, based on edge-cloud collaboration mechanism. At last, we design a group of simulation experiments to validate the effectiveness and efficiency of our approach. Experimental comparisons indicate a good balance between different evaluation metrics. [ABSTRACT FROM AUTHOR]

Compared with cloud computing, edge–cloud collaboration can avoid long transmitting delay to cloud since tasks are close to edge, which makes edge–cloud collaboration suitable for delay-sensitive applications. However, the complex environment of edge–cloud poses new challenge to task scheduling. A Collaborative Scheduling strategy based on task Admission and Delay Evaluation (CSADE) is proposed to deal with the challenge and ensure the quality of service (QoS). In order to schedule maximum tasks to edge and guarantee QoS, a new dynamic delay model in resource manager is proposed to accurately estimate the average delay. Based on the average delay and execution time, task evaluator prevents the impossible tasks to avoid waste of resources on both edge and cloud. The scheduling policy in task scheduler fully leverages the conditions of edge, cloud and tasks to guarantee the QoS. The fault-tolerant mechanism would launch and adjust the scheduling strategy when task Scheduling in emergencies and resource node failures. Thus CSADE ensures the QoS for delay-sensitive applications from three levels, i.e., the accurate quantitative system delay in resource manager, the strict task admission evaluation in task evaluator, the edge-first elastic scheduling strategy and the fault-tolerant mechanism in task scheduler. Comparative experimental results on simulated datasets and real datasets verify that CSADE can reduce average delay time and QoS violation rate obviously. [ABSTRACT FROM AUTHOR]

Emerging computing paradigms provide field‐level service responses for users, for example, edge computing, fog computing, and MEC. Edge virtualization technologies represented by Docker can provide a platform‐independent, low‐resource‐consumption operating environment for edge service. The image‐pulling time of Docker is a crucial factor affecting the start‐up speed of edge services. The layer reuse mechanism of native Docker cannot fully utilize the duplicate data of node local images. In this paper, we propose a chunk reuse mechanism (CRM), which effectively targets node‐local duplicate data during container updates and reduces the volume of data transmission required for image building. We orchestrate the CRM process for cloud and remote‐cloud nodes to ensure that the resource overhead from container update data preparation and image reconstruction is within an acceptable range. The experimental results show that the CRM proposed in this paper can effectively utilize the node local duplicate data in the synchronous update of containers in multiple nodes, reduce the volume of data transmission, and significantly improve container update efficiency. [ABSTRACT FROM AUTHOR]

Low-Power Wide-Area Networks (LPWANs), extensively utilized for connecting billions of IoT devices, encounter wireless interference challenges in unlicensed frequency bands. Cutting-edge research suggests employing Received Signal Strength Indication (RSSI) sequences for error detection to mitigate interference-related issues. Nevertheless, the effectiveness of this method significantly declines under low signal-to-noise ratios (SNRs). Additionally, long-range communication often results in low SNR received signals, sometimes even below the noise floor. Targeting this fundamental issue, this article proposes the LPWAN packet technique, broadly applicable across diverse scenarios through edge–cloud collaboration. On the edge side, we propose an innovative architecture that fully exploits spatial distribution and interference independence in the field. Rather than utilizing resource-intensive RSSI-based error detection, we leverage a lightweight coding scheme for error detection at the Long Range (LoRa) edge, forwarding correct frames to the cloud. On the cloud side, packet recovery is achieved utilizing group-weighted voting. We design and implement ECRLoRa with commercially available devices (SemTech's SX1278 and SX1302 LoRa chipsets) and assess its performance in low SNR environments. Our thorough evaluation demonstrates that our approach attains a Packet Recovery Ratio of 96% with low SNR (i.e., below −10 dB), resulting in 1.8× throughput, 7.5× faster recovery time, and 4.92× average accuracy compared to state-of-the-art cloud-optimized application layer solutions. [ABSTRACT FROM AUTHOR]

Resistance spot welding poses potential challenges for automotive manufacturing enterprises with regard to ensuring the real-time and accurate quality detection of each welding spot. Nowadays, many machine learning and deep learning methods have been proposed to utilize monitored sensor data to solve these challenges. However, poor detection results or process interpretations are still unaddressed key issues. To bridge the gap, this paper takes the automotive bodies as objects, and proposes a resistance spot welding quality online detection method with dynamic current and resistance data based on a combined convolutional neural network (CNN), long short-term memory network (LSTM), and an attention mechanism. First, an overall online detection framework using an edge–cloud collaboration was proposed. Second, an online quality detection model was established. In it, the combined CNN and LSTM network were used to extract local detail features and temporal correlation features of the data. The attention mechanism was introduced to improve the interpretability of the model. Moreover, the imbalanced data problem was also solved with a multiclass imbalance algorithm and weighted cross-entropy loss function. Finally, an experimental verification and analysis were conducted. The results show that the quality detection accuracy was 98.5%. The proposed method has good detection performance and real-time detection abilities for the in-site welding processes of automobile bodies. [ABSTRACT FROM AUTHOR]

The rapid advancement of renewable energy technologies necessitates innovative solutions for the efficient deployment and management of microgrid systems. This paper presents a detailed study on the implementation of edge–cloud collaboration-based plug and play (PnP) and topology identification for microgrids, focusing on the Jingshan AC/DC Microgrid Cluster System (JS-MCS) in Hubei, China. Firstly, the paper elucidates the structure of JS-MCS's physical system and integral information system, comprising the cloud platform, communication network, edge unit, and terminal unit. The following sections introduce the methods of PnP, describing the overall procedure and specific implementations between cloud and edge and cloud and terminal. Then, a novel approach for topology identification is further explored based on correlation analysis. An exemplary analysis of the studied system, JS-MCS, accentuates practical insights into the application of PnP and topology identification methods. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) is an important component of the 5th generation mobile network, which is being deployed worldwide. To achieve the goal of "connecting everything", a great amount of devices are connected. New scenarios and applications are developed for IoT, some of which require intensive computations and low response delay. This makes edge computing and cloud computing two main technologies to back the Internet of Things. Edge-cloud computing was proposed to utilize the advantages of both these two computing forms. In this paper, we construct a secure edge-cloud computing (SECC) framework. Sensor nodes and applications can interact with the framework through unified interfaces. We implement the edge server as a collection of services, including Orchestration, Data Processing and Storage, Communication Management, Authentication and Authorization, and Situation Awareness. Through a daisy-chain approach, our framework can be secured for heterogeneous security needs of different parts of the system. We also build a prototype of the proposed framework and demonstrate the efficacy and security of the SECC framework through experiments and comprehensive analysis. [ABSTRACT FROM AUTHOR]

Federated Learning (FL) is suitable for the application scenarios of distributed edge collaboration of the Internet of Things (IoT). It can provide data security and privacy, which is why it is widely used in the IoT applications such as Industrial IoT (IIoT). Latest research shows that the federated learning framework is vulnerable to poisoning attacks in the case of an active attack by the adversary. However, the existing backdoor attack methods are easy to be detected by the defence methods. To address this challenge, we focus on edge‐cloud synergistic FL clean‐label attacks. Unlike common backdoor attack, to ensure the attack's concealment, we add a small perturbation to realize the clean label attack by judging the cosine similarity between the gradient of the adversarial loss and the gradient of the normal training loss. In order to improve the attack success rate and robustness, the attack is implemented when the global model is about to converge. The experimental results verified that 1% of poisoned data could make an attack successful with a high probability. Our method maintains stealth while performing model poisoning attacks, and the average Peak Signal‐to‐Noise Ratio (PSNR) of poisoning images reaches over 30 dB, and the average Structural SIMilarity (SSIM) is close to 0.93. Most importantly, our attack method can bypass the Byzantine aggregation defence. [ABSTRACT FROM AUTHOR]

Owing to the strong protection of data privacy, federated learning (FL) has become a key method to achieve intelligent decision making in smart homes. However, under realistic conditions, such as differentiated requirements and heterogeneous service environments, FL in smart homes faces the problems of non-independent and identically distributed (non-IID) data and uneven computing power, which lead to the poor adaptability of global models. To address this issue, this study proposes a cluster FL architecture based on edge-cloud collaboration. Firstly, a Gaussian mixture model-based cluster FL is proposed to improve the model accuracy by clustering features on the FL dataset and ensuring an independent identical distribution of the data. Subsequently, a model training strategy based on edge-cloud collaboration is proposed to achieve the sharing of edge-cloud computing power by split learning, which provides sufficient computing power for model training. The simulation results show that the proposed architecture improves the accuracy of global models, meanwhile ensures the provision of normal network service.

By integrating communications in different domains, integrated radio and optical networks can serve a wider range of applications and services. Integrated radio and optical network scenarios will involve more weak-computation-ability network nodes, such as small-cell base stations. To pursue efficient integrated radio and optical networks, more efficient ways to conduct transmission under the demand of edge and cloud collaboration are required. The lack of forward-looking resource allocation may easily lead to a waste of network resources without an expected return. Therefore, an efficient resource allocation scheme needs to consider certain issues: 1) a comprehensive perspective of traffic prediction; 2) a release of pressure on the transmission pipeline during the prediction process; and 3) a reduction of loss of edge nodes due to the computation. In this paper, benefiting from machine learning, we propose a resource allocation with edge-cloud collaborative traffic prediction (TP-ECC) in integrated radio and optical networks, where an efficient resource allocation scheme (ERAS) is designed based on the prediction results with the gated recurrent unit model. We maximize the utilization of limited resources to improve the awareness of network status. We present three evaluation indicators and build a network architecture to evaluate our resource allocation scheme. Through edge-cloud collaboration, our proposal can improve traffic prediction accuracy by 9.5% compared with single-point traffic prediction, and resource utilization is also improved by edge-cloud collaborative traffic prediction.

As a “plug-and-play” nuclear system, the technology of fission battery enables nuclear reactor systems to function as batteries. Undoubtedly, advanced intelligent technologies have to be adopted to realize the excellent attributes of fission batteries, i.e., economic, standardized, installed, unattended, and reliable, and its commercial applications. This article proposed the edge-cloud collaboration technology for fission batteries to satisfy its intelligent requirements. A detailed framework for a fission battery management system has been developed. To develop such a management system, four types of concrete transmission information for the edge-cloud collaboration (i.e. resource and data, AI technologies, application and services, safety and strategies) are identified. The edge and cloud designs are also discussed in detail, including their hardware and software. A functional framework of edge-cloud collaboration network platform for fission battery management system is developed preliminarily. The framework is based on a three-layer structure, i.e., perception layer, technical layer and application layer. Although no detailed algorithm or verification is introduced (since the edge-cloud collaboration framework has been widely used in some industry fields but not in nuclear engineering), as primarily work for the future detailed study of fission battery, it is necessary and extremely important.

Both accurate and fast mobile recommendation systems based on click behaviors analysis are crucial in ebusiness. Deep learning has achieved state-of-the-art accuracy and the traditional wisdom often hosts these computation-intensive models in powerful cloud centers. However, the cloud-only approaches put significant computational pressure on cloud servers and increase the latency in heavy-load scenarios. Moreover, existing work often adopts RNN structures to model behaviors that suffer from low processing speed for under-utilization of parallel devices such as GPUs. In this work, we propose an efficient internet behaviorbased recommendation framework with edge-cloud collaboration on deep CNNs (CoRec) to improve both the accuracy and speed for mobile recommendation. A novel convolutional interest network (CIN) improves the accuracy by modeling the long- and short-term interests and accelerates the prediction through parallelfriendly convolutions. To further improve the serving throughput and latency, a novel device-cloud collaboration strategy reduces workloads by pre-computing and caching long-term interests in the cloud offline and real-time computation of short-term interests in devices. Extensive experiments on real-world datasets show that CoRec significantly outperforms the state-of-the-art methods in accuracy and has achieved at least an order of magnitude improvement in latency and throughput compared to cloud-only RNN-based approaches for long behaviors. [ABSTRACT FROM AUTHOR]

Federated learning has a distributed collaborative training mode, widely used in IoT scenarios of edge computing intelligent services. However, federated learning is vulnerable to malicious attacks, mainly backdoor attacks. Once an edge node implements a backdoor attack, the embedded backdoor mode will rapidly expand to all relevant edge nodes, which poses a considerable challenge to security-sensitive edge computing intelligent services. In the traditional edge collaborative backdoor defense method, only the cloud server is trusted by default. However, edge computing intelligent services have limited bandwidth and unstable network connections, which make it impossible for edge devices to retrain their models or update the global model. Therefore, it is crucial to detect whether the data of edge nodes are polluted in time. This paper proposes a layered defense framework for edge-computing intelligent services. At the edge, we combine the gradient rising strategy and attention self-distillation mechanism to maximize the correlation between edge device data and edge object categories and train a clean model as much as possible. On the server side, we first implement a two-layer backdoor detection mechanism to eliminate backdoor updates and use the attention self-distillation mechanism to restore the model performance. Our results show that the two-stage defense mode is more suitable for the security protection of edge computing intelligent services. It can not only weaken the effectiveness of the backdoor at the edge end but also conduct this defense at the server end, making the model more secure. The precision of our model on the main task is almost the same as that of the clean model. [ABSTRACT FROM AUTHOR]

Retinopathy of prematurity is an ophthalmic disease with a very high blindness rate. With its increasing incidence year by year, its timely diagnosis and treatment are of great significance. Due to the lack of timely and effective fundus screening for premature infants in remote areas, leading to an aggravation of the disease and even blindness, in this paper, a deep learning-based collaborative edge-cloud telemedicine system is proposed to mitigate this issue. In the proposed system, deep learning algorithms are mainly used for classification of processed images. Our algorithm is based on ResNet101 and uses undersampling and resampling to improve the data imbalance problem in the field of medical image processing. Artificial intelligence algorithms are combined with a collaborative edge–cloud architecture to implement a comprehensive telemedicine system to realize timely screening and diagnosis of retinopathy of prematurity in remote areas with shortages or a complete lack of expert medical staff. Finally, the algorithm is successfully embedded in a mobile terminal device and deployed through the support of a core hospital of Guangdong Province. The results show that we achieved 75% ACC and 60% AUC. This research is of great significance for the development of telemedicine systems and aims to mitigate the lack of medical resources and their uneven distribution in rural areas. [ABSTRACT FROM AUTHOR]

Mine video surveillance has a key role in ensuring the production safety of intelligent mining. However, existing mine intelligent monitoring technology mainly processes the video data in the cloud, which has problems, such as network congestion, large memory consumption, and untimely response to regional emergencies. In this paper, we address these limitations by utilizing the edge-cloud collaborative optimization framework. First, we obtained a coarse model using the edge-cloud collaborative architecture and updated this to realize the continuous improvement of the detection model. Second, we further proposed a target detection model based on the Vision Swin Transformer-YOLOv5(ViST-YOLOv5) algorithm and improved the model for edge device deployment. The experimental results showed that the object detection model based on ViST-YOLOv5, with a model size of only 27.057 MB, improved the average detection accuracy is by 25% compared to the state-of-the-art model, which makes it suitable for edge-end deployment in mining workface. For the actual mine surveillance video, the edge-cloud collaborative architecture can achieve better performance and robustness in typical application scenarios, such as weak lighting and occlusion, which verifies the feasibility of the designed architecture. [ABSTRACT FROM AUTHOR]

Face recognition is widely used in daily life such as shopping,security check,travel,payment and work attendance.Face recognition systems need strong computing power and large storage space,so face images that need to be recognized are usually transmitted to the cloud platform through the network.Due to the problems of network coverage,congestion and delay,face re-cognition systems are difficult to meet the needs of actual application,and the user experience is poor.Aiming at the problems in face recognition,a face recognition method based on edge-cloud collaboration is proposed.This method combines the processing ability of cloud computing and the real-time performance of edge computing,so that face recognition systems are not constrained by the network status,and its application is more extensive and the user experience is better.In the cloud,the LResNet feature extraction method is proposed to improve the ResNet34 network structure,and the ArcFace face loss function is used to supervise the training process,so that the network can learn more face angle features.At the edge,due to the limited computing resources and storage resources,a SResNet feature extraction method is proposed.Deep separable convolution is used to lighten the LResNet network structure,and it has greatly reduced network parameters and computation.The face recognition experiment on edge-cloud collaboration shows that the system can recognize faces in real time with a high accuracy rate under any network status.

With the development of 5G and the enrichment of application functions, applications have put forward higher requirements on the computing capabilities of terminal devices.In order to improve the computing capabilities of terminal devices on applications and reduce the processing time of tasks, it is aimed at mobile edge computing environments, a task offloading method for edge-cloud collaboration was proposed，and an elite hierarchical evolutionary algorithm combined with reinforcement learning (RL-EHEA) was designed to perform offloading decisions, so that multiple tasks with dependencies and deadlines compete for computing resources.The simulation experiment results show that, compared with genetic algorithm (GA) and elite genetic algorithm (EGA), RL-EHEA can shorten task processing time and obtain better resource allocation strategy.

Resistance spot welding poses potential challenges for automotive manufacturing enterprises with regard to ensuring the real-time and accurate quality detection of each welding spot. Nowadays, many machine learning and deep learning methods have been proposed to utilize monitored sensor data to solve these challenges. However, poor detection results or process interpretations are still unaddressed key issues. To bridge the gap, this paper takes the automotive bodies as objects, and proposes a resistance spot welding quality online detection method with dynamic current and resistance data based on a combined convolutional neural network (CNN), long short-term memory network (LSTM), and an attention mechanism. First, an overall online detection framework using an edge–cloud collaboration was proposed. Second, an online quality detection model was established. In it, the combined CNN and LSTM network were used to extract local detail features and temporal correlation features of the data. The attention mechanism was introduced to improve the interpretability of the model. Moreover, the imbalanced data problem was also solved with a multiclass imbalance algorithm and weighted cross-entropy loss function. Finally, an experimental verification and analysis were conducted. The results show that the quality detection accuracy was 98.5%. The proposed method has good detection performance and real-time detection abilities for the in-site welding processes of automobile bodies.

Internet of Things (IoT) and artificial intelligence (AI) can realize the concept of “smart city.” Video surveillance in smart cities is, usually, based on a centralized framework in which large amounts of real-time media data are transmitted to and processed in the cloud. However, the cloud relies on network connectivity of the Internet that is sometimes limited or unavailable; thus, the centralized framework is not sufficient for real-time processing of media data needed for smart video surveillance. To tackle this problem, edge computing - a technique for accelerating the development of AIoT (AI across IoT) in smart cities - can be conducted. In this paper, a distributed real-time object detection framework based on edge-cloud collaboration for smart video surveillance is proposed. When collaborating with the cloud, edge computing can serve as converged computing through which media data from distributed edge devices of the network are consolidated by AI in the cloud. After AI discovers global knowledge in the cloud, it to be shared at the edge is deployed remotely on distributed edge devices for real-time smart video surveillance. First, the proposed framework and its preliminary implementation are described. Then, the performance evaluation is provided regarding potential benefits, real-time responsiveness and low-throughput media data transmission.

Federated Learning (FL) is one of the key technologies to solve privacy protection for cloud‐edge intelligent collaborative computing, and its security and privacy issues have attracted extensive attention from academia and industry. FL is a distributed privacy protection framework. Multiple edged nodes or servers jointly train a machine learning model by sharing model parameters without exchanging local data. However, there are still many security risks and privacy threats in FL in edge‐cloud collaborative computing. In this paper, we mainly discuss the security and privacy challenges on FL in collaborative computing at the edge. First, we introduce the principle, classification, and threat model of FL in edge‐cloud collaboration, which helps understand the challenges faced by edge‐cloud collaborative computing. Second, privacy leakage attacks and poisoning attacks launched by adversaries or honest but curious actors are summarized and compared. Then, the problems existing on the attack method are summarized and analyzed. Finally, the future development direction of FL in the field of edge‐cloud collaborative computing is further discussed. [ABSTRACT FROM AUTHOR]

With the continuous development of intelligent transportation system technology, vehicle users have higher and higher requirements for low latency and high service quality of task computing. The computing offloading technology of mobile edge computing (MEC) has received extensive attention in the Internet of Vehicles (IoV) architecture. However, due to the limited resources of the MEC server, it cannot meet the task requests from multiple vehicle users simultaneously. For this reason, making correct and fast offloading decisions to provide users with a service with low latency, low energy consumption, and low cost is still a considerable challenge. Regarding the issue above, in the IoV environment where vehicle users race, this paper designs a three-layer system task offloading overhead model based on the Edge-Cloud collaboration of multiple vehicle users and multiple MEC servers. To solve the problem of minimizing the total cost of the system performing tasks, an Edge-Cloud collaborative, dynamic computation offloading method (ECDDPG) based on a deep deterministic policy gradient is designed. This method is deployed at the edge service layer to make fast offloading decisions for tasks generated by vehicle users. The simulation results show that the performance is better than the Deep Q-network (DQN) method and the Actor-Critic method regarding reward value and convergence. In the face of the change in wireless channel bandwidth and the number of vehicle users, compared with the basic method strategy, the proposed method has better performance in reducing the total computational cost, computing delay, and energy consumption. At the same time, the computational complexity of the system execution tasks is significantly reduced. [ABSTRACT FROM AUTHOR]

Visually impaired people face many inconveniences in daily life, and there are problems such as high prices and single functions in the market of assistance tools for visually impaired people. In this work, we designed and implemented a low-cost intelligent assistance cane, particularly for visually impaired individuals, based on computer vision, sensors, and an edge-cloud collaboration scheme. Obstacle detection, fall detection, and traffic light detection functions have been designed and integrated for the convenience of moving for visually impaired people. We have also designed an image captioning function and object detection function with high-speed processing capability based on an edge-cloud collaboration scheme to improve the user experience. Experiments show that the performance metrics have an aerial obstacle detection accuracy of 92.5%, fall detection accuracy of 90%, and average image retrieval period of 1.124 s. It proves the characteristics of low power consumption, strong real-time performance, adaptability to multiple scenarios, and convenience, which can ensure the safety of visually impaired people when moving and can help them better perceive and understand the surrounding environment. [ABSTRACT FROM AUTHOR]

Abstract The convergence of Artificial Intelligence (AI) and the Internet of Things (IoT), or AIoT, has breathed a new life into IoT operations and human-machine interactions. Currently, resource-constrained IoT devices usually cannot provide sufficient capability for data storage and processing so as to support building modern AI models. An intuitive solution is to integrate cloud computing technology into AIoT and exploit the powerful and elastic computing as well as the storage capacity of the servers on the cloud end. Nevertheless, the network bandwidth and communication latency increasingly become serious bottlenecks. The emerging edge computing can complement the cloud-based AIoT in terms of communication latency, and hence attracts more and more attention from the AIoT area. In this paper, we present an industrial edge-cloud collaborative computing platform, namely Sophon Edge, that helps to build and deploy AIoT applications efficiently. As an enterprise-level solution for the AIoT computing paradigm, Sophon Edge adopts a pipeline-based computing model for streaming data from IoT devices. Besides, this platform supports an iterative way for model evolution and updating so as to enable the AIoT applications agile and data-driven. Through a real-world example, we demonstrate the effectiveness and efficiency of building an AIoT application based on the Sophon Edge platform.

Currently, the world is experiencing the rapid spread of Coronavirus Disease 2019 (COVID-19). Since the epidemic continues to take a devastating impact on the society, economy, and healthcare, the real-time detection of COVID-19 is essential for fast and cost-effective diagnosis services. Fortunately, deep learning (DL), as a promising technology, enables the COVID-19 diagnosis services on chest X-ray (CXR) images. The training task of DL model is generally implemented at the centralized cloud. However, due to the geo-distributed data sources and the transmission of large amounts of raw data to the centralized cloud, the transmission latency becomes a bottleneck of the COVID-19 diagnosis model training. In this paper, we propose a Distributed COVID-19 detection model training method on CXR images with edge-cloud collaboration, named DisCOV. Specifically, to improve the training efficiency and guarantee the model accuracy, a distributed lightweight model-based training algorithm is designed with the cooperation of edge computing and cloud computing. In addition, a resource allocation algorithm is developed during the training to jointly minimize the time cost and energy consumption. Extensive experiments based on real-world CXR image datasets demonstrate that DisCOV is better performed and more promising than the existing baselines. [ABSTRACT FROM AUTHOR]

With the prevalence of smart appliances, smart meters, and Internet of Things (IoT) devices in smart grids, artificial intelligence (AI) built on the rich IoT big data enables various energy data analysis applications and brings intelligent and personalized energy services for users. In conventional AI of Things (AIoT) paradigms, a wealth of individual energy data distributed across users’ IoT devices needs to be migrated to a central storage (e.g., cloud or edge device) for knowledge extraction, which may impose severe privacy violation and data misuse risks. Federated learning, as an appealing privacy-preserving AI paradigm, enables energy data owners (EDOs) to cooperatively train a shared AI model without revealing the local energy data. Nevertheless, potential security and efficiency concerns still impede the deployment of federated-learning-based AIoT services in smart grids due to the low-quality shared local models, non-independently and identically distributed (non-IID) data distributions, and unpredictable communication delays. In this article, we propose a secure and efficient federated-learning-enabled AIoT scheme for private energy data sharing in smart grids with edge-cloud collaboration. Specifically, we first introduce an edge-cloud-assisted federated learning framework for communication-efficient and privacy-preserving energy data sharing of users in smart grids. Then, by considering non-IID effects, we design a local data evaluation mechanism in federated learning and formulate two optimization problems for EDOs and energy service providers. Furthermore, due to the lack of knowledge of multidimensional user private information in practical scenarios, a two-layer deep reinforcement-learning-based incentive algorithm is developed to promote EDOs’ participation and high-quality model contribution. Extensive simulation results show that the proposed scheme can effectively stimulate EDOs to share high-quality local model updates and improve the communication efficiency. [ABSTRACT FROM AUTHOR]

Taking autonomous driving and driverless as the research object, we discuss and define intelligent high-precision map. Intelligent high-precision map is considered as a key link of future travel, a carrier of real-time perception of traffic resources in the entire space-time range, and the criterion for the operation and control of the whole process of the vehicle. As a new form of map, it has distinctive features in terms of cartography theory and application requirements compared with traditional navigation electronic maps. Thus, it is necessary to analyze and discuss its key features and problems to promote the development of research and application of intelligent high-precision map. Accordingly, we propose an information transmission model based on the cartography theory and combine the wheeled robot’s control flow in practical application. Next, we put forward the data logic structure of intelligent high-precision map, and analyze its application in autonomous driving. Then, we summarize the computing mode of “Crowdsourcing+Edge-Cloud Collaborative Computing”, and carry out key technical analysis on how to improve the quality of crowdsourced data. We also analyze the effective application scenarios of intelligent high-precision map in the future. Finally, we present some thoughts and suggestions for the future development of this field.

With the popularization of video surveillance systems and the diversification of their applications,traditional data processing methods using cloud computing to handle all the computations are more and more unsustainable.Edge computing pre-processes video data nearby,and performs well in terms of bandwidth,storage,and latency,but edge-cloud collaboration is still required to improve overall performance.The architecture of edge-cloud collaborative video surveillance system was proposed,then an open edge node design scheme and algorithm training and inferencing ideas were proposed.Finally,taking forest fire prevention and smart poles ＆amp; towers as examples,the application realization and value of edge computing in video surveillance services were illustrated.

The depth development and widespread application of edge intelligence technology based on the Internet of Things has led to edge-cloud collaboration and related research. In recent years, with the rapid development of the Internet of Things and the formation of super-city groups, the management characteristics of enterprises with multiple manufacturing plants served for headquarters have become increasingly obvious. The problem of order dynamic fluctuations caused by personalized customization requirements has become more prominent, which makes it impossible to do global long-period prediction or real-time short-period response relied solely on the cloud or edge. Therefore, this paper proposes a production system scheduling framework under the edge-cloud collaborative paradigm based on the dynamic fluctuation of orders under these background, and builds an edge-cloud collaborative scheduling model, which guarantees real-time distributed scheduling at the edge. It enabled the cloud to periodically predict the total completion time of production tasks at the headquarters based on the value-added data uploaded by the edge, and to support more accurate and efficient scheduling at the edge based on the prediction results. Finally, an example analysis proved the rationality of the scheduling mechanism and the effectiveness of the scheduling model. The proposed method can provide a certain reference for task scheduling in the edge-cloud collaborative production paradigm.

In the practice of cloud manufacturing, there still exist some major challenges, including: 1) cloud based big data analytics and decision-making cannot meet the requirements of many latency-sensitive applications on shop floors; 2) existing manufacturing systems lack enough reconfigurability, openness and evolvability to deal with shop-floor disturbances and market changes; and 3) big data from shop-floors and the Internet has not been effectively utilized to guide the optimization and upgrade of manufacturing systems. This paper proposes an open evolutionary architecture of the intelligent cloud manufacturing system with collaborative edge and cloud processing. Hierarchical gateways connecting and managing shop-floor things at the “edge” side are introduced to support latency-sensitive applications for real-time responses. Big data processed both at the gateways and in the cloud will be used to guide continuous improvement and evolution of edge-cloud systems for better performance. As software tools are becoming dominant as the “brain” of manufacturing control and decision-making, this paper also proposes a new mode - “AI-Mfg-Ops” (AI enabled Manufacturing Operations) with a supporting software defined framework, which can promote fast operation and upgrading of cloud manufacturing systems with smart monitoring-analysis-planning-execution in a closed loop. This research can contribute to the rapid response and efficient operation of cloud manufacturing systems.

As a key equipment of manufacturing, the intelligence of CNC machine tools affects the development and progress of intelligent manufacturing. At the same time, the rapid development of various technologies in recent years, such as cloud computing and edge computing, has brought new methods to improve the intelligence of CNC machine tools. This article proposes a new architecture of intelligent machine tools based on edge-cloud collaboration (IMT-ECC). The hierarchical structure of IMT-ECC is introduced and consists of three layers, data acquisition, network communication and edge-cloud collaboration. Combining the real-time characteristic of edge computing and complex problem processing ability of cloud computing, the edge-cloud collaboration is designed to improve the intelligence of machine tools through data collaboration, information collaboration and knowledge collaboration. Finally, the feasibility of the new intelligent machine tool architecture based on edge-cloud collaboration is verified by experiments with gantry heavy-duty CNC machine tools.

Federated learning has a distributed collaborative training mode, widely used in IoT scenarios of edge computing intelligent services. However, federated learning is vulnerable to malicious attacks, mainly backdoor attacks. Once an edge node implements a backdoor attack, the embedded backdoor mode will rapidly expand to all relevant edge nodes, which poses a considerable challenge to security-sensitive edge computing intelligent services. In the traditional edge collaborative backdoor defense method, only the cloud server is trusted by default. However, edge computing intelligent services have limited bandwidth and unstable network connections, which make it impossible for edge devices to retrain their models or update the global model. Therefore, it is crucial to detect whether the data of edge nodes are polluted in time. This paper proposes a layered defense framework for edge-computing intelligent services. At the edge, we combine the gradient rising strategy and attention self-distillation mechanism to maximize the correlation between edge device data and edge object categories and train a clean model as much as possible. On the server side, we first implement a two-layer backdoor detection mechanism to eliminate backdoor updates and use the attention self-distillation mechanism to restore the model performance. Our results show that the two-stage defense mode is more suitable for the security protection of edge computing intelligent services. It can not only weaken the effectiveness of the backdoor at the edge end but also conduct this defense at the server end, making the model more secure. The precision of our model on the main task is almost the same as that of the clean model.

Retinopathy of prematurity is an ophthalmic disease with a very high blindness rate. With its increasing incidence year by year, its timely diagnosis and treatment are of great significance. Due to the lack of timely and effective fundus screening for premature infants in remote areas, leading to an aggravation of the disease and even blindness, in this paper, a deep learning-based collaborative edge-cloud telemedicine system is proposed to mitigate this issue. In the proposed system, deep learning algorithms are mainly used for classification of processed images. Our algorithm is based on ResNet101 and uses undersampling and resampling to improve the data imbalance problem in the field of medical image processing. Artificial intelligence algorithms are combined with a collaborative edge–cloud architecture to implement a comprehensive telemedicine system to realize timely screening and diagnosis of retinopathy of prematurity in remote areas with shortages or a complete lack of expert medical staff. Finally, the algorithm is successfully embedded in a mobile terminal device and deployed through the support of a core hospital of Guangdong Province. The results show that we achieved 75% ACC and 60% AUC. This research is of great significance for the development of telemedicine systems and aims to mitigate the lack of medical resources and their uneven distribution in rural areas.

The application scenarios and market shares of industrial robots have been increasing in recent years, and with them comes a huge market and technical demand for industrial robot-monitoring system (IRMS). With the development of IoT and cloud computing technologies, industrial robot monitoring has entered the cloud computing era. However, the data of industrial robot-monitoring tasks have characteristics of large data volume and high information redundancy, and need to occupy a large amount of communication bandwidth in cloud computing architecture, so cloud-based IRMS has gradually become unable to meet its performance and cost requirements. Therefore, this work constructs edge–cloud architecture for the IRMS. The industrial robot-monitoring task will be executed in the form of workflow and the local monitor will allocate computing resources for the subtasks of the workflow by analyzing the current situation of the edge–cloud network. In this work, the allocation problem of industrial robot-monitoring workflow is modeled as a latency and cost bi-objective optimization problem, and its solution is based on the evolutionary algorithm of the heuristic improvement NSGA-II. The experimental results demonstrate that the proposed algorithm can find non-dominated solutions faster and be closer to the Pareto frontier of the problem. The monitor can select an effective solution in the Pareto frontier to meet the needs of the monitoring task. [ABSTRACT FROM AUTHOR]

The awareness of physical network topology in a large-scale Internet of Things (IoT) system is critical to enable location-based service provisioning and performance optimization. However, due to the dynamics and complexity of IoT networks, it is usually very difficult to discover and update the physical topology of the large-scale IoT systems in real-time. Considering the stringent latency requirements in IoT systems, while the initial processing time for topology discovery can be tolerated, latency due to real-time topology update constitutes an even higher level of challenge. In this paper, a novel fast hierarchical topology update scheme is proposed for the large-scale IoT systems enabled by using the edge-cloud collaborative architecture. Specifically, an event-driven neighbor update algorithm, termed as TriggerOn, is firstly developed to update the local neighbor table of the end devices when device association or disassociation occurs. Based on the updated neighbor tables, the physical topology update of the subnet is conducted at the coordinated edge device, where a hybrid multidimensional scaling (MDS) based 3D localization algorithm is developed to locate the newly associated devices. Simulation results have indicated that as compared to the benchmark methods, the neighbor discovery latency has been reduced dramatically, and the 3D localization accuracy has been improved. Furthermore, the overall latency incurred by the proposed hierarchical physical topology update scheme is significantly lower than the distributed consensus-based update scheme, especially for the large-scale IoT subnets. [ABSTRACT FROM AUTHOR]

Caching state data of real-world entities just in the cloud without any distinction will cause search performance degrading, due to the characteristics of uncountable number of entities and time-varying state of entities in cyber-physical systems (CPSs). Considering the diverse time-varying features of CPS entities, an edge-cloud collaborative entity state data caching strategy toward networking search application in CPSs is proposed in this article. Specifically, an entity state feature extraction method is presented to mine underlying changing rules of CPS entities via raw entity state observation sequence. Then, an edge and cloud collaborative entity state data caching strategy is devised to improve the search accuracy of CPSs search service and reduce the search delay and energy consumption, in which entities are clustered first according to the time-varying degree of their state and then these state information are discriminately cached based on their belonging clusters. Simulation results validate the effectiveness of the proposed strategy in terms of real-time and accuracy performances. [ABSTRACT FROM AUTHOR]