4225 results about "Edge computing" patented technology

The invention relates to a GIS device state intelligent monitoring system and method based on the edge computing technology. The system includes a state sensor terminal, a data node device and an intelligent monitoring center, wherein the state sensor terminal is used for gathering collected state data to the data node device to perform power IoT edge calculation and is connected with the intelligent monitoring center through the transmission network, the condition monitoring sensor terminal is used for acquiring related state parameters in real time and receiving and performing limited edge calculation tasks, the data node device is used for collecting the data and receiving and performing the edge calculation tasks, and the intelligent monitoring center is used for performing equipment IoT management, intelligent evaluation analysis and advanced application for a GIS device state and configuring an optimized edge calculation algorithm model and an SG-CIM data model to the data node device. The system is advantaged in that a circuit breaker GIS equipment Internet of Things system of a substation is constructed, and the automation, informationization and intelligence level of equipment state evaluation and diagnosis can be improved.

The invention provides a computing task unloading method based on edge computing and cloud computing collaboration. The computing task unloading method comprises the steps that variable parameters areset and initialized; Constructing respective time delay models and energy consumption models of the mobile terminal, the edge node and the remote cloud, obtaining a time delay expected value model and a total energy consumption model of the mobile terminal when the current task load is completely executed, and further obtaining a time delay expected value model and a total energy consumption model of all tasks in the total mobile terminal when the tasks are executed; Defining an optimal distribution problem and converting the optimal distribution problem into a convex optimization problem; And introducing a Lagrangian function to solve an optimal solution of task execution amounts of the terminal local machine, the edge node and the far-end cloud under the KKT constraint condition, so that each mobile terminal adjusts and executes according to the task execution amounts of the terminal local machine, the edge node and the far-end cloud, which are obtained by respectively and correspondingly solving the optimal solution. According to the implementation of the invention, the computing power and the power consumption limitation of the mobile terminal, the edge node and the remote cloud are comprehensively considered, and an optimal computing task unloading decision is realized.

An exemplary system includes a cloud server configured to provide a service to a client device by way of a network. The cloud server implements a batch layer of a data-processing architecture. The system further includes an edge computing device configured to provide a latency-sensitive task of the service to the client device by way of the network in response to a request from the client device to offload the latency-sensitive task. The edge computing device is separate from and communicatively coupled to the cloud server, is located at an edge of the network such that a latency of data communications between the edge computing device and the client device is lower than a latency of data communications between the cloud server and the client device, and implements a speed layer of the data-processing architecture at the edge of the network.

An edge computing platform with machine learning capability is provided between a local network with a plurality of sensors and a remote network. A machine learning model is created and trained in the remote network using aggregated sensor data and deployed to the edge platform. Before being deployed, the model is edge-converted (“edge-ified”) to run optimally with the constrained resources of the edge device and with the same or better level of accuracy. The “edge-ified” model is adapted to operate on continuous streams of sensor data in real-time and produce inferences. The inferences can be used to determine actions to take in the local network without communication to the remote network. A closed-loop arrangement between the edge platform and remote network provides for periodically evaluating and iteratively updating the edge-based model.

The invention discloses an industrial Internet of Things system based on edge computing, and the system comprises a data collection module, an edge gateway, and a cloud processing platform; the data collection module is used for collecting the original data of a field device; The edge gateway is used for converting the original data into standard data, carrying out real-time analysis on key data in the original data, carrying out real-time intelligent processing on the field equipment according to an analysis result, carrying out classified packaging on the standard data, carrying out protocolconversion on the standard data, and sending the standard data to the cloud processing platform; And the cloud processing platform is used for carrying out data analysis, system monitoring and energymanagement according to the data from the edge gateway, and sending a data analysis result to a corresponding Internet of Things application. Through the edge gateway, the data transmission cost canbe effectively reduced, and the real-time decision and response of the key service can be ensured; By utilizing the powerful management capability of the cloud platform, mass data access is supported,a safe and reliable message processing capability is provided, and enterprises are assisted to integrate value chains and quickly construct innovative Internet of Things services.

The invention discloses a deep learning computing system and method for cloud terminal edge computing fusion and relates to the fields of cloud computing, edge computing and artificial intelligence technology. According to the method, deep learning computing is distributed to a cloud terminal, a pipeline and an edge side, the cloud terminal is responsible for basic model training of a large historical data computing quantity, personalized model distribution is performed according to the demand of the edge side, nodes from the cloud terminal to the edge side deploy a deep learning model after learning, the deep learning model is used for completing inference, the edge side continuously feeds back an inference result, and then the inference result is uploaded to the cloud terminal to continuously optimize the model. Compared with the traditional mode that training and inference are both performed on the cloud terminal, in the whole deep learning computing process, the computing model iscontinuously optimized, personalized model distribution is performed according to the demand of the edge side, bandwidth is effectively utilized, network transmission efficiency is guaranteed, and real-time service execution efficiency is improved.

The invention discloses an industrial big data monitoring system based on edge computing. The system comprises a data source layer, an edge layer, a platform layer and an application layer. The data source layer collects heterogeneous data information through various Internet of Things technologies and transmits the data to the edge server through a short-distance communication technology. The edge layer specifically analyzes and processes the real-time data acquired by the data source layer on the edge side, extracts high-value data and transmits the high-value data to the platform layer. theplatform layer mainly utilizes a cloud computing platform to realize the functions of industrial data analysis, data storage and the like on the data transmitted by the edge layer. The application layer is used for providing related software services based on the platform layer and displaying the information to a user through user terminal equipment. According to the invention, the problems of large platform load, low data processing efficiency, high network time delay, low security and the like of a traditional data monitoring system based on a cloud computing platform in an industrial dataviolence environment are mainly solved.

The invention relates to a transformer substation electrical equipment monitoring, analyzing and processing system and particularly relates to a transformer substation electrical equipment comprehensive physical examination system based on cloud side end collaborative perception. The system is characterized in that multi-node detection and unified perception access are realized through an integrated communication network architecture on the edge side of a station end, an intelligent sensing terminal is comprehensively covered in real time. Meanwhile, local intelligent data processing is realized by utilizing an edge computing technology; an omnibearing intelligent diagnosis and analysis system of the transformer substation is constructed at a cloud end, and multiple functions such as abnormal real-time early warning and intelligent sensing prediction are realized, so that the real-time tracking of the operation state of the power equipment of the transformer substation is realized. Themethod has the advantages that the transformer substation can be covered, monitoring data can be comprehensively utilized, an intelligent processing scheme can be provided, construction of the intelligent transformer substation is promoted, and the intelligent evaluation and safety early warning level of the electrical equipment state of the transformer substation is improved.

The invention belongs to the technical field of privacy protection in cloud computing and discloses a privacy protection method in face authentication system based on edge calculation. The method comprises the steps that: a camera collects a face image and uploads the face image to a connected edge computing node, an authority assigning mechanism sends a permission vector Li of a user to the edgecomputing node after collecting user face information, an edge computing device carries out feature extraction on a face image of a user requesting registration to obtain a feature vector by using a method based on a deep convolutional neural network, a safe nearest neighbor algorithm is performed to encrypt the feature vector of a registered user, and a secret sharing homomorphic algorithm is performed and n sub-secrets are generated according to the feature vector and are distributively stored in n edge computing devices. The direct upload of sensitive data to a cloud server is avoided, theprivacy of the face data is protected, and the fault tolerance of a system is improved. Results prove that the accuracy of face recognition under a ciphertext of the present invention is completely equal to the accuracy of face recognition under a plaintext.

Techniques for embedded event processing are provided. In some examples, instructions for executing at least a subset of complex event processing features may be received. A stream of events received by the edge computing device may be identified. Additionally, the events of the stream utilizing the subset of complex event processing features may be processed. Further, in some aspects, the processed events may be provided to a complex event platform of a server.

The invention discloses a task unloading method based on user experience in an edge computing network. The method comprises the following steps: a scheduler receives a task unloading request submittedby a user; a task unloading strategy model is jointly optimized based on time and energy expenditure; the model is solved and calculated based on a branch and bound algorithm of linear reconstruction; the task unloading strategy model is optimized based on time and energy expenditure. Based on the results of the calculation, the decision is made whether the task is unloaded to the edge server, and the tasks unloaded to the edge server are communicated and the computational resources are allocated. The invention combines two key indexes of optimization time delay and energy consumption, and solves the problem by branch and bound algorithm based on linear reconstruction technology, which can reduce the energy consumption of mobile intelligent device under the condition of ensuring the completion of the task, reduce the task processing time delay, and achieve the purpose of maximizing the optimization of user experience.

The invention discloses a dynamic resource allocation method based on evolutionary game in a mobile edge computing system. The method comprises the following steps that (1) a network is divided into a plurality of areas according to the network coverage condition, accessible service points in the areas are different, and a centralized controller is arranged in the network; (2) terminals with the task unloading need in the same area form a population, and the terminals in the population establish task unloading cost functions; (3) all terminals in each population randomly select accessible SPs in an SP selection strategy set; the evolutionary game is established in each population in the network; (4) the terminals in each population compute task unloading costs and send the SP selection strategies and the cost information to the controller; (5) the population carries out SP selection strategy update according to dynamic copy; and (6) the dynamic copy reaches evolution equilibrium. The method fully utilizes the computing resources and the radio resources of the SPs, aims at the equal task unloading costs of all terminals in the populations and meets the task unloading need of each mobile terminal based on the evolutionary game.

The invention discloses a production line control system based on digital twinning, and the system comprises a data collection subsystem which collects the current operation state data of a productionline in real time; a virtual twinning control subsystem which is used for generating a digital twinning model representing physical equipment at each station of the production line according to the historical operation state data of the production line and the current operation state data of the production line; a data analysis subsystem which generates new operation state data according to a recommendation algorithm, the historical operation state data of the production line, the current operation state data and the target operation state data of the production line. The digital twinning model simulates the operation state of the production line according to the new operation state data, and the data analysis subsystem is further used for finally generating optimal operation state data according to a simulation result of the digital twin model. The system also comprises an edge computing gateway which is used for realizing data interaction between the virtual twinning control subsystem and the data acquisition subsystem.

The invention belongs to the technical field of the communication 5G and the edge computing, specifically an edge computing method for a 5G ultra-dense networking scene. The basic thought of the invention is to provide the edge computing service for a cellular user of a microcell by placing a MEC sever at a small base station; and meanwhile, the cellular user can select the cloud computing service connected with a macro base station according to the feature of the to-be-executed task. The edge computing method disclosed by the invention has the advantages that the user can select the most suitable computing service, and the delay performance is improved.

The invention relates to the field of mobile communication, and particularly relates to a mobile edge computing node selection method, device and system. the method comprises: the edge cloud gateway receiving a domain name request forwarded by the user plane function, wherein the domain name request is forwarded when a domain name in the domain name request sent by the terminal is analyzed by a user plane function to be a service acceleration domain name or a destination address is an IP address with global load balance; or forwarding the domain name request when the user plane function determines that the protocol port information in the domain name request sent by the terminal meets the protocol port requirement or the destination address is an IP address with global load balance; and obtaining a corresponding edge application VIP from the global load balance, sending the VIP to the terminal through the user plane function, receiving a service request sent by the terminal forwarded by the user plane function, and distributing the service request to the corresponding MEC processing server. An edge cloud gateway is newly added, DNS and HTTP DNS addressing scenes are compatible, edge computing nodes are selected nearby, and the purpose of edge acceleration is achieved.

The invention discloses a dynamic migration method of video tasks in an edge computing environment. Multiple ECNs (Edge Compute Nodes) are selected in an ECN cluster under the edge environment to runa task scheduling migration scheduler to be responsible for distribution and scheduling work of the tasks, and the other ECNs in the cluster undertake task migration work and provide computing resources. The computing task migration scheduler executes task migration decision in each scheduling cycle to determine whether the tasks are executed locally in the ECNs or migrated to a data center, and tasks needing to be migrated to the data center are directly sent to a task execution unit of the remote-end data center through an ECN transmission unit for analysis processing. According to the method, the tasks are dynamically scheduled and processed on the basis of meeting network bandwidth and calculation costs, goals of reducing transmission time delay of edge data and improving task analysisprocessing speed are achieved, thus low delay time and high response of a task migration process are effectively guaranteed, and video service experience of users is improved while higher video service quality is realized.

The invention discloses a machine learning training system and method based on edge cloud computing. A user terminal continuously collects new data expansion and optimal training samples, an output model is trained by a machine learning platform, the trained output model is quickly issued to the user terminal, the user terminal collects the application status of the model and feeds back to the machine learning platform to complete the retraining of the model, a virtuous circle is formed to continuously improve the training ability of the machine learning platform and model accuracy, computingresources are closer to users, and rapid deployment and unified management of the machine learning platform is achieved with the help of resource pooling and resource management capabilities of a cloud computing platform.

The invention discloses an edge computing privacy protection system and method based on joint learning. The system comprises a client and a server, the client is used for local training, adding disturbance to updated parameters, encrypting the updated parameters and sending the encrypted parameters to the server, and the server receives encrypted data sent by the client, decrypts the encrypted data and updates local parameters to update a deep learning model; the protection method comprises the following steps: step 1, adding disturbance to parameters at a client; step 2, encrypting the data at a client; and step 3, decrypting the data at the server. According to the method, each participant can safely submit data on the premise of not needing any trusted aggregator; noise disturbance is added to local updating in a distributed mode, updating of the disturbance is encrypted through a Paillier homomorphic cryptosystem, safety and performance analysis shows that the PPFL protocol can guarantee privacy of client data and learning accuracy at the same time, and the conflict problem of privacy protection and learning accuracy is solved.