11 results about "Smart client" patented technology

Apparatuses, methods, and systems for client steering in mesh networks are disclosed. A first access point (AP) of a network sends a quality of service (QoS) data packet to a client device. A received signal strength indication (RSSI) of the client device is captured using the QoS data packet. A distance between the AP and client device is determined using a Wi-Fi round trip time. A time and day of week is determined. Using machine learning, a second AP is identified for steering the client device to, based on the RSSI, the distance, and the time and day of week. A machine learning model is trained to steer each client device to a respective AP for increasing a throughput of the network based on features extracted from client behavior of the client devices. The second AP is connected to the client device.

The invention discloses a safe and efficient sensor data prediction and search method, and belongs to the field of Internet of Things. The method comprises the steps of region division, data acquisition, model construction, joint reasoning, search initiation and search response. Wherein the region division comprises the step of dividing the terminal sensor equipment into a fixed management range according to the geographic position of the terminal sensor equipment, and uniformly managing the terminal sensor equipment by a belonging edge server. The data collection comprises the steps that the sensor collects periodic data or non-periodic data according to the function type of the sensor and uploads the periodic data or the non-periodic data to the edge server. And model construction: constructing a federated learning framework based on edge cloud collaboration, and initializing to construct a sensor data prediction model based on T-LSTM. The joint reasoning comprises the steps that the edge servers and the cloud server cooperatively train prediction models, and the trained prediction models are deployed to the edge servers. Search initiation comprises the step that a user initiates a search request to the edge server through the intelligent client. The search response comprises the steps that the edge server recognizes a user request and executes search operation, and required sensor data information is fed back to the user. The method can effectively improve the search efficiency of the sensor data and protect the data privacy at the same time.

Apparatuses, methods, and systems for client steering in mesh networks are disclosed. A first access point (AP) of a network sends a quality of service (QoS) data packet to a client device. A received signal strength indication (RSSI) of the client device is captured using the QoS data packet. A distance between the AP and client device is determined using a Wi-Fi round trip time. A time and day of week is determined. Using machine learning, a second AP is identified for steering the client device to, based on the RSSI, the distance, and the time and day of week. A machine learning model is trained to steer each client device to a respective AP for increasing a throughput of the network based on features extracted from client behavior of the client devices. The second AP is connected to the client device.

The invention discloses an environmental protection industry science and technology consultation service platform, and relates to the technical field of big data resource service. The platform comprises a big data cloud server, an intelligent client, an edge computing server and an Internet of Things terminal, and integrates artificial intelligence, Internet of Things and block chain technologies. The big data cloud server comprises an AI intelligent analysis module, a block chain evidence storage module and an environment data middle platform; the intelligent client is integrated with VR training, a data visualization billboard and other units. The edge computing server processes the equipment data in real time; and the Internet of Things terminal realizes environmental protection facility state sensing. The AI module predicts the project declaration success rate through deep learning, the error rate is smaller than or equal to 5%, and the block chain module ensures that data cannot be tampered based on an alliance chain. The platform can accurately match policies and enterprise requirements, integrate real-time environment data, provide services such as VR training and intelligent early warning, solve the problems of insufficient intelligence, weak data security and the like of an existing platform, and improve the accuracy and efficiency of environment-friendly consultation services.

The application relates to a business data security access system, method and equipment applying a third-party large model, and relates to the field of traffic management safety. The business data security access system and method applying the third-party large model provided by the application realize technical isolation between business data and a third-party public cloud large model based on an intelligent client control carrier, solve the problem of leakage of business data and personal information in the process of developing intelligent services, and realize intelligent application of the third-party public cloud large model in the field of public security government service.

Aspects of the present disclosure provide a method and apparatus (electronic device) for media processing. In some examples, the electronic device includes a processing circuit for performing a process of a smart client, which is a client interface of the electronic device to an immersive media streaming network. The method for media processing includes transmitting, by the client interface of the electronic device, information of capabilities and availability of the electronic device for playing scene-based immersive media to a server device in the immersive media streaming network. The method further includes receiving, by the client interface, a media stream carrying adapted media content for the scene-based immersive media. The adapted media content is generated from the scene-based immersive media by the server device based on the capabilities and availability information. Then, the method includes playing the scene-based immersive media according to the adapted media content.

Improving read performance using pre-fetches by encoding an extent map provided by an application executed on a client, where the extent map identifies valid data of the file as extents that may be interspersed with holes constituting invalid data. A server-based filesystem interprets the extent map as a read receipt to bring the right extents in memory ahead of a read request. The extents of the extent map are prefetched at a prefetch read horizon. Because the client is aware of the application-resident extent map, the prefetch operations can intelligently skip over the holes. Multi-stream restores using multiple streams implemented through corresponding TCP connections between client and server are used for the read and pre-fetches.

The invention provides a semi-asynchronous federal learning training method based on intelligent client selection and adaptive weighting, and the method comprises the steps: enabling a client to carry out the model initialization according to the broadcast of a central server, and enabling the central server to calculate the local data distribution according to a training set, clustering the clients according to a clustering algorithm on the basis of local data distribution of the clients; the central server randomly selects clients from the clusters of each cluster in a fixed period and broadcasts the latest global model weight; all the selected clients upload local model weights and version numbers immediately, and the unselected clients continue to carry out local training; the central server performs synchronous self-adaptive weighted aggregation on each selected client according to the uploaded local model weight and the local model version to update the global model weight; and S4, repeating the steps S2 to S4 to realize global training until the global model reaches the accuracy of the target model, and ending the training. The convergence speed and accuracy of the model are improved.

Embodiments of the present invention provide computer-implemented methods, computer program products and computer systems for enabling modifying a behavior of a consuming component. The method comprises requesting a smart client by a consuming component. The smart client is adapted to modify a control logic of the consuming component and to establish a network connection. Next, the consuming component executes the smart client and receives a modification for the smart client of the consuming component. Further, a computer-implemented method for enabling modifying a behavior of a consuming component by a serving component is disclosed. This method comprises, upon receiving a request for a smart client by the serving component, providing a smart client, transmitting the smart client, receiving service requests comprising procedure calls to be executed by the serving component, and, upon detecting a performance change by the serving component, sending a modification for the smart client.

Arrangements for dynamic selection of delta client copy strategies are provided. An intelligent client copy computing platform may determine whether a database table of a source client is empty. If not, the copy computing platform may determine whether the database table has changed. If not, the computing platform may determine whether the database table has been copied from a same source client of a last copy operation. If so, the computing platform may determine whether a record count of the source client is the same as the record count of a target client. If not, the computing platform may apply a delta copy algorithm. The algorithm may select records of the source client, deduct records of the target client from those records, insert records that are in the source but not in the target client, and delete records that are in the target but not in the source client.

The application provides a smart parental control lock based on display content, a control method and equipment, and belongs to the technical field of parental control locks. A controller module of a parental control lock client is respectively connected with an intelligent client module, a screenshot module and a timer module. The intelligent client module sends content acquired by the screenshot module to an intelligent server module and receives a content level message fed back by the intelligent server module. The content level message is forwarded to an equipment processing module through the controller module. The controller module acquires first content level information of a parental control lock server and forwards the first content level information to the equipment processing module. The equipment processing module receives the first content level information sent by the controller module and forwards the first content level information to a system of a screen-equipped equipment. Real-time control of children's application electronic equipment by the parental control lock and comprehensive control of application content watched by the children can be realized, development cost is saved, and the comprehensiveness of the function of the parental control lock is improved.