Documentation file-embedded machine learning models

    公开(公告)号:US12008448B2

    公开(公告)日:2024-06-11

    申请号:US18183120

    申请日:2023-03-13

    Inventor: Changchuan Yin

    CPC classification number: G06N20/00 G06F40/103

    Abstract: A processing system including at least one processor may obtain a machine learning model, serialize the machine learning model into a serialized format, and embed a delimiter indicator into a documentation file comprising information regarding the use of the machine learning model, where the delimiter indicator is in a file position that is after an end-of-file indicator of the documentation file. The processing system may further embed the machine learning model in the serialized format into the documentation file in a file position that is after the delimiter indicator. The processing system may then store the documentation file with the delimiter indicator and the machine learning model in the serialized format that are embedded.

    INTELLIGENT ADAPTIVE SIGNALING AUTOMATION FOR METAVERSE STREAMING

    公开(公告)号:US20240181339A1

    公开(公告)日:2024-06-06

    申请号:US18060863

    申请日:2022-12-01

    Inventor: Rohit Abhishek

    CPC classification number: A63F13/355 A63F2300/538

    Abstract: Aspects of the subject disclosure may include, for example, a method that includes receiving, by a processing system in communication with a client device, a signal from the client device according to a session description protocol (SDP); the signal comprises a capability parameter regarding streaming of video frames to the client device, a first focus area parameter and a second focus area parameter indicating respectively a first resolution for a high focus area and a second resolution for a low focus area of the video frame, a bandwidth parameter indicating a bandwidth availability for a streaming video session, and a streaming method parameter regarding the streaming of the video frames. The system calculates the high focus area and the low focus area of the video frame, and delivers, subsequent to receiving the signal, a plurality of video frames to the client device. Other embodiments are disclosed.

    Facilitation of voice over new radio for 5G or other next generation network

    公开(公告)号:US12004023B2

    公开(公告)日:2024-06-04

    申请号:US17985737

    申请日:2022-11-11

    Inventor: Zhi Cui Hongyan Lei

    CPC classification number: H04W36/0055

    Abstract: Automatic neighbor relationships (ANR) and voice over new radio (VoNR) microservices can be implemented within a radio access network intelligent controller (RIC), which receives load information from network nodes and user equipment measurements from user equipment. The RIC can create and maintain real-time load information and leverage slicing knowledge of neighbor cells during mobility to direct the user equipment to various network slices. Thus, an enhanced radio access network (RAN) neighbor relationship that utilizes slicing information and VoNR preferences based on current and historical data can provide slice level RAN neighbor visibility on top of physical ANR relationships to enable enhancements for mobility management optimization for VoNR functionality.

    Facilitation of deep service path discovery for 5G or other next generation network

    公开(公告)号:US11997583B2

    公开(公告)日:2024-05-28

    申请号:US17212961

    申请日:2021-03-25

    CPC classification number: H04W40/248 H04L45/306 H04L45/64 H04W28/12 H04W40/34

    Abstract: A software defined network (SDN) can add network repository functions (NRF) into a configurations database to enable NF discovery. The SDN can subscribe to NRF notifications to receive new cloud native functions (CNF), registrations, or any other update to the CNF status in 5G system. In addition to listening to NRF notifications, the SDN can implement CNF pooling processes to periodically retrieve CNF from an NRF repository and stay in sync with 5G systems. Thus, a deep service path discovery can be developed from network service configurations and container call flows to enable an accurate alarm correlation and troubleshooting for the operations. This service path deep discovery can be designed and implemented as a standalone system or in an SDN framework with integration of a container management framework such as K8 kubernetes.

    V2X network communication
    147.
    发明授权

    公开(公告)号:US11997567B2

    公开(公告)日:2024-05-28

    申请号:US17490870

    申请日:2021-09-30

    Inventor: Sreejith Menon

    CPC classification number: H04W4/06 H04W4/40

    Abstract: An evolved multimedia broadcast multicast services (eMBMS) core may map sensor data or telemetry data from a central repository server (e.g., tag it with an area code from where the data is being generated) to a service area. The eMBMS core may inform the radio access network to create a service area for the specific area code, which may be part of or an entire city, county, or the like. Then the radio access network may broadcast the data to the vehicles or devices in the specific area code.

    System for contextual diminished reality for meta verse immersions

    公开(公告)号:US11995739B2

    公开(公告)日:2024-05-28

    申请号:US17668069

    申请日:2022-02-09

    CPC classification number: G06T11/00 G06V20/20 G06T2210/21

    Abstract: Aspects of the subject disclosure may include, for example, receiving information about an augmented reality (AR) environment which includes a plurality of objects and a user participating in the AR environment, the user viewing the AR environment through a user equipment, receiving context information for the AR environment, identifying a diminished reality (DR) object of the plurality of objects, wherein the identifying is responsive to the context information, modifying the AR environment to render the DR object invisible to the user viewing the AR environment through user equipment, monitoring movement of one or more objects of the AR environment including the DR object, and modifying visibility of the DR object to the user viewing the AR environment through user equipment as a notification of a possible collision between the user and the DR object, wherein the modifying is responsive to detecting a risk of collision. Other embodiments are disclosed.

Patent Agency Ranking