WO2021218302A1 - 机器学习模型参数传递方法及装置 - Google Patents

机器学习模型参数传递方法及装置 Download PDF

Info

Publication number
WO2021218302A1
WO2021218302A1 PCT/CN2021/076843 CN2021076843W WO2021218302A1 WO 2021218302 A1 WO2021218302 A1 WO 2021218302A1 CN 2021076843 W CN2021076843 W CN 2021076843W WO 2021218302 A1 WO2021218302 A1 WO 2021218302A1
Authority
WO
WIPO (PCT)
Prior art keywords
endogenous
management unit
service
intelligent
machine learning
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/076843
Other languages
English (en)
French (fr)
Inventor
索士强
王映民
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Datang Mobile Communications Equipment Co Ltd
Original Assignee
Datang Mobile Communications Equipment Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Datang Mobile Communications Equipment Co Ltd filed Critical Datang Mobile Communications Equipment Co Ltd
Priority to EP21795817.2A priority Critical patent/EP4145359B1/en
Priority to KR1020227040107A priority patent/KR20230002801A/ko
Priority to JP2022565894A priority patent/JP7418610B2/ja
Priority to US17/921,642 priority patent/US20230169398A1/en
Publication of WO2021218302A1 publication Critical patent/WO2021218302A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0806—Configuration setting for initial configuration or provisioning, e.g. plug-and-play
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/082—Configuration setting characterised by the conditions triggering a change of settings the condition being updates or upgrades of network functionality
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W80/00—Wireless network protocols or protocol adaptations to wireless operation
    • H04W80/08—Upper layer protocols
    • H04W80/12—Application layer protocols, e.g. WAP [Wireless Application Protocol]

Definitions

  • This application relates to the field of communication technology, and in particular to a method and device for transmitting machine learning model parameters.
  • Neural Network Neural Network
  • ANN Artificial Neural Network
  • a simple neural network includes an input layer, an output layer, and a hidden layer (if necessary), and each layer includes multiple neurons (Neurons).
  • DNN Deep Neural Network
  • RNN Recurrent Neural Network
  • CNN Convolutional Neural Network
  • the embodiments of the present application provide a method and device for transferring machine learning model parameters, which are used to implement the deployment/update of the machine learning model when the machine learning deduction model is inside the wireless mobile communication system.
  • a method for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous business slave management unit in the first device receives the machine learning model file and target functional unit information sent by the intelligent endogenous business master management unit in the second device;
  • the intelligent endogenous service slave management unit distributes the machine learning model file to the target function unit in the first device according to the target function unit information.
  • the intelligent endogenous service slave management unit in the first device receives the machine learning model file and target functional unit information sent by the intelligent endogenous service master management unit in the second device; the intelligent endogenous service slave management Unit, according to the target function unit information, distribute the machine learning model file to the target function unit in the first device, so that when the machine learning deduction model is inside the wireless mobile communication system, the deployment of the machine learning model is realized /renew.
  • the intelligent endogenous service slave management unit is located in the application layer of the first device; the intelligent endogenous service master management unit is located in the application layer of the second device.
  • the method further includes:
  • the target functional unit installs the machine learning model file.
  • the method further includes:
  • the target functional unit notifies the endogenous business slave management unit that the machine learning model file has been installed
  • the endogenous service slave management unit applies to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the intelligent endogenous business uses the transmission resource obtained by the application from the management unit to apply for service registration to the intelligent endogenous business main management unit;
  • the intelligent endogenous business obtains the service registration permission of the intelligent endogenous business main management unit from the management unit, the service included in the latest machine learning model installed in the target functional unit is started.
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • the method for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous business main management unit in the second device determines the target functional unit of the machine learning model to be deployed in the first device and its corresponding machine learning model file;
  • the intelligent endogenous service main management unit in the second device applies to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the intelligent endogenous service master management unit in the second device uses the transmission resource obtained by the application to send the target functional unit information and the machine learning model file to the intelligent endogenous service slave management unit in the first device.
  • the intelligent endogenous service slave management unit is located in the application layer of the first device; the intelligent endogenous service master management unit is located in the application layer of the second device.
  • the method further includes:
  • the smart endogenous service master management unit in the second device receives the transmission resource application sent by the smart endogenous service slave management unit in the first device, and allocates the first device to the smart endogenous service slave management unit Transmission resources with the second device.
  • the method further includes:
  • the smart endogenous service main management unit in the second device receives the service registration application sent by the smart endogenous service from the management unit in the first device, and sends a service registration license for the smart endogenous service from the management unit .
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • an apparatus for transmitting machine learning model parameters in a mobile communication system includes:
  • Memory used to store program instructions
  • the processor is configured to call the program instructions stored in the memory, and execute according to the obtained program:
  • Control the intelligent endogenous service slave management unit and distribute the machine learning model file to the target function unit in the first device according to the target function unit information.
  • the intelligent endogenous service slave management unit is located in the application layer of the first device; the intelligent endogenous service master management unit is located in the application layer of the second device.
  • the processor is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • the processor is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • an apparatus for transmitting machine learning model parameters in a mobile communication system includes:
  • Memory used to store program instructions
  • the processor is configured to call the program instructions stored in the memory, and execute according to the obtained program:
  • the intelligent endogenous service slave management unit is located in the application layer of the first device; the intelligent endogenous service master management unit is located in the application layer of the second device.
  • the processor is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • the processor is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • Control the smart endogenous service main management unit in the second device to receive the service registration application sent by the smart endogenous service from the management unit in the first device, and send service registration for the smart endogenous service from the management unit license.
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • another device for transmitting machine learning model parameters in a mobile communication system includes:
  • the receiving unit is configured to receive the machine learning model file and target functional unit information sent by the intelligent endogenous business main management unit in the second device;
  • the sending unit is configured to distribute the machine learning model file to the target function unit in the first device according to the target function unit information.
  • another device for transmitting machine learning model parameters in a mobile communication system includes:
  • the determining unit is used to determine the target functional unit of the machine learning model to be deployed in the first device and its corresponding machine learning model file;
  • An application unit configured to apply to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the sending unit is configured to send the target functional unit information and the machine learning model file to the intelligent endogenous service slave management unit in the first device using the transmission resource obtained by the application.
  • the third device for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous service slave management unit is used to receive the machine learning model file and the target functional unit information sent by the intelligent endogenous service main management unit in the second device; according to the target functional unit information, the machine learning model file Distribute to the target functional unit in the first device;
  • the functional unit is used to install the machine learning model file.
  • the intelligent endogenous service slave management unit is located in the application layer of the first device.
  • the functional unit is further configured to notify the endogenous business slave management unit that the machine learning model file has been installed;
  • the endogenous service slave management unit is further configured to: apply to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the intelligent endogenous business main management unit applies for service registration; after obtaining the service registration permission of the intelligent endogenous business main management unit, the service included in the latest machine learning model installed in the target functional unit is started.
  • the third device for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous business main management unit is used to determine the target functional unit of the machine learning model to be deployed in the first device and its corresponding machine learning model file; apply to the transmission resource management unit in the second device for the first device and the second device Transmission resources between devices; use the transmission resources obtained by the application to send the target functional unit information and the machine learning model file to the intelligent endogenous service slave management unit in the first device;
  • the transmission resource management unit is used to provide transmission resources between the first device and the second device.
  • the intelligent endogenous service main management unit is located in the application layer of the second device.
  • the device further includes:
  • the machine learning model library is used to store machine learning model files.
  • the smart endogenous service master management unit is further configured to: receive a transmission resource application sent by the smart endogenous service slave management unit in the first device, and allocate the smart endogenous service slave management unit Transmission resources between the first device and the second device.
  • the smart endogenous service master management unit is further configured to: receive a service registration application sent by the smart endogenous service slave management unit in the first device, and send the service registration application for the smart endogenous service slave management unit Service registration license.
  • Another embodiment of the present application provides a computing device, which includes a memory and a processor, wherein the memory is used to store program instructions, and the processor is used to call the program instructions stored in the memory, according to the obtained program Perform any of the above methods.
  • Another embodiment of the present application provides a computer storage medium, the computer storage medium stores computer-executable instructions, and the computer-executable instructions are used to make the computer execute any of the foregoing methods.
  • FIG. 1 is a schematic diagram of the user plane provided by an embodiment of the application, where the start point and the end point of the endogenous service are respectively constructed on the RAN side;
  • FIG. 2 is a schematic diagram of the control plane provided by an embodiment of the application, where the start point and the end point of the endogenous service are respectively constructed on the RAN side;
  • FIG. 3 is a schematic diagram of the main process of updating a machine learning model provided by an embodiment of the application
  • FIG. 4 is a schematic flowchart of a method for transmitting machine learning model parameters on the first device side according to an embodiment of the application;
  • FIG. 5 is a schematic flowchart of a method for transmitting machine learning model parameters on the second device side according to an embodiment of the application
  • FIG. 6 is a schematic structural diagram of a machine learning model parameter transmission device on the first device side according to an embodiment of the application;
  • FIG. 7 is a schematic structural diagram of a machine learning model parameter transmission device on the second device side according to an embodiment of the application.
  • FIG. 8 is a schematic structural diagram of another machine learning model parameter transmission device on the first device side according to an embodiment of the application.
  • FIG. 9 is a schematic structural diagram of another machine learning model parameter transmission device on the second device side according to an embodiment of the application.
  • FIG. 10 is a schematic structural diagram of a third machine learning model parameter transmission device on the first device side according to an embodiment of the application.
  • FIG. 11 is a schematic structural diagram of a third type of machine learning model parameter transmission device on the second device side according to an embodiment of the application.
  • the introduction of machine learning into the wireless mobile communication system can solve the complex problems of the mobile communication system or improve the performance.
  • how to deploy/update the machine learning model trained on the network side to the terminal side has become a systematic problem and requires a unified solution.
  • the embodiments of the present application provide a method and device for transferring machine learning model parameters, so as to realize the transfer of the machine learning model trained on the network side to the terminal side.
  • the method and the device are based on the same application concept. Since the method and the device have similar principles for solving the problem, the implementation of the device and the method can be referred to each other, and the repetition will not be repeated.
  • the applicable system can be the global system of mobile communication (GSM) system, code division multiple access (CDMA) system, and wideband code division multiple access (Wideband Code Division Multiple Access, WCDMA) general packet Wireless service (general packet radio service, GPRS) system, long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD), general Mobile system (universal mobile telecommunication system, UMTS), worldwide interoperability for microwave access (WiMAX) system, 5G system, 5G NR system, etc.
  • GSM global system of mobile communication
  • CDMA code division multiple access
  • WCDMA wideband Code Division Multiple Access
  • general packet Wireless service general packet radio service
  • GPRS general packet Radio service
  • LTE long term evolution
  • FDD frequency division duplex
  • TDD LTE time division duplex
  • UMTS general Mobile system
  • WiMAX worldwide interoperability for microwave access
  • the terminal device involved in the embodiments of the present application may be a device that provides voice and/or data connectivity to a user, a handheld device with a wireless connection function, or other processing devices connected to a wireless modem.
  • the name of the terminal device may be different.
  • the terminal device may be called a user equipment (UE).
  • the wireless terminal device can communicate with one or more core networks via the RAN.
  • the wireless terminal device can be a mobile terminal device, such as a mobile phone (or called a "cellular" phone) and a computer with a mobile terminal device, for example, it can be a portable , Pocket, handheld, computer built-in or vehicle-mounted mobile devices that exchange language and/or data with the wireless access network.
  • Wireless terminal equipment can also be called system, subscriber unit, subscriber station, mobile station, mobile station, remote station, and access point , Remote terminal equipment (remote terminal), access terminal equipment (access terminal), user terminal equipment (user terminal), user agent (user agent), user device (user device), which are not limited in the embodiments of the present application.
  • the network device involved in the embodiment of the present application may be a base station, and the base station may include multiple cells.
  • a base station may also be called an access point, or may refer to a device in an access network that communicates with a wireless terminal device through one or more sectors on an air interface, or other names.
  • the network device can be used to convert the received air frame and the Internet protocol (IP) packet to each other, as a router between the wireless terminal device and the rest of the access network, where the rest of the access network can include the Internet Protocol (IP) communication network.
  • IP Internet Protocol
  • the network equipment can also coordinate the attribute management of the air interface.
  • the network equipment involved in the embodiment of this application may be a network equipment (base transmitter station, BTS) in the global system for mobile communications (GSM) or code division multiple access (CDMA). ), it can also be a network device (NodeB) in wide-band code division multiple access (WCDMA), or an evolved network device in a long-term evolution (LTE) system (evolutional node B, eNB or e-NodeB), 5G base station in the 5G network architecture (next generation system), or home evolved node B (HeNB), relay node (relay node), home base station ( Femto), pico base station (pico), etc., are not limited in the embodiment of the present application.
  • BTS network equipment
  • GSM global system for mobile communications
  • CDMA code division multiple access
  • NodeB wide-band code division multiple access
  • LTE long-term evolution
  • 5G base station in the 5G network architecture next generation system
  • HeNB home evolved node B
  • relay node relay node
  • Femto home
  • the embodiment of the present application proposes a solution to deploy/update a machine learning model trained on the system side to the terminal side in a wireless mobile communication system using endogenous services.
  • the endogenous services described in the embodiments of the present application may refer to the types of services built between the base station and the terminal in the mobile communication system in order to solve the complex problems of the mobile communication system or improve the performance, which is not directly required by the user.
  • Functional service the object of its service is the network itself, or the operator, so it is called endogenous business.
  • the machine learning model to be deployed (its corresponding function) can be regarded as an endogenous business, so that the deployment or update of the machine learning model can be performed in a business deployment or update manner.
  • the machine learning model can be regarded as an application (APP), and the deployment/update of the model is the download and installation process of the APP.
  • APP application
  • the endogenous business approach can deploy more complex models, and it is also more conducive to the know-how protection of the model itself.
  • the control plane mainly executes the process of service registration of endogenous services on the RAN side, radio resource application and other processes.
  • the service registration process initiated by the terminal side is taken as an example for description, that is, a specific endogenous service (such as the "terminal-side channel estimation" service executed on the terminal side) needs to be reported to the processor where it is located before being executed.
  • a specific endogenous service such as the "terminal-side channel estimation" service executed on the terminal side
  • the registration process is generally initiated by the application layer of the subject performing the service (for example, the subject of the "terminal-side channel estimation" service is the terminal), and passes through the wireless network layer below the application layer (refer to the 5G wireless network layer, which may pass through the RRC , PDCP, RLC, MAC, and PHY layers) to form an air interface signal and send it to the base station; after the base station receives the air interface signal, it passes through the wireless network layer (refer to the 5G wireless network layer, which may pass through PHY, MAC, RLC, PDCP) , RRC layer) to obtain the service registration application initiated by the terminal. After the base station examines the application, it will confirm the registration information, and then send it to the terminal after being processed by the wireless network layer below the application layer; after receiving the air interface signal, the terminal will be authorized by the base station after being processed by the wireless network layer information.
  • the 5G wireless network layer which may pass through the RRC , PDCP, RLC, MAC, and PHY layers
  • the user plane mainly deals with user data in endogenous services on the RAN side.
  • This data may be generated by the application layer (such as the test data used by the terminal to perform measurement services), or it may be generated in the wireless network layer under the control of the control plane (such as the reference signal used in the "terminal-side channel estimation" service, It is generated by the PHY layer in the wireless network layer).
  • the test data used by the terminal to perform the measurement service is taken as an example.
  • the data is generated by the application layer on the base station side and passes through the wireless network layer (refer to the 5G wireless network layer, which may pass through the SDAP, PDCP, RLC, MAC, and PHY layers). ) To form an air interface signal and send it to the terminal; after receiving the air interface signal, the terminal side calls the measurement service to process the air interface signal.
  • Step 1 When the base station decides to deploy/update the machine learning model for the target functional unit on the terminal side, the intelligent endogenous service main management unit in the base station selects the corresponding machine learning model file from the machine learning model library;
  • the target functional unit includes channel estimation, signal detection, channel decoding, channel quality indicator (Channel Quality Indicator, CQI) measurement, synchronization detection, transmission antenna selection, etc.
  • the intelligent endogenous service main management unit is located at the application layer of the base station side, and is mainly used to provide the service registration function for the endogenous service and the function of applying for transmission resources between the terminal and the base station.
  • the machine learning model file is generally understandable executable program file, such as .exe. Generally include multiple files, and packaged into a compressed file package, such as .zip.
  • Step 2 The intelligent endogenous service main management unit in the base station applies to the transmission resource management unit on the base station side for transmission resources between the terminal and the base station;
  • Step 3 The intelligent endogenous service main management unit in the base station uses the transmission resources obtained by the application to send the machine learning model file from the machine learning model library to the intelligent endogenous service slave management unit on the terminal side, and sends the corresponding target function along with it Unit information
  • the intelligent endogenous service slave management unit is located at the application layer on the terminal side.
  • the target functional unit information includes, for example, the name of the target functional unit, as shown in the second column of Table 1.
  • Step 4 The intelligent endogenous business slave management unit on the terminal side distributes the machine learning model file to the target functional unit on the terminal side according to the received target function unit information;
  • Step 5 After the target functional unit on the terminal side receives the machine learning model file distributed by the intelligent endogenous service from the management unit, install it and notify the endogenous service from the management unit;
  • Step 6 The endogenous service management unit on the terminal side applies to the transmission resource management unit on the base station side for transmission resources between the terminal and the base station;
  • Step 7 The intelligent endogenous service on the terminal side uses the transmission resource obtained by the application from the management unit to apply for service registration to the main management unit of the intelligent endogenous service in the base station;
  • Step 8 After the intelligent endogenous service on the terminal side obtains the service registration permission of the intelligent endogenous service main management unit on the base station side from the management unit, the service included in the newly deployed/updated machine learning model in the target functional unit is started.
  • Table 1 For example, a list of machine learning model libraries is shown in Table 1 below, in which the machine learning model file compression package uniquely corresponding to it can be found according to the machine learning model number. Table 1
  • the embodiment of the present application provides a system for deploying/updating a machine learning model using endogenous services. It includes at least a base station side and a terminal side.
  • the base station side includes an intelligent endogenous service main management unit and a transmission resource management unit.
  • the machine learning model library includes the intelligent endogenous business slave management unit and the target functional unit of the applicable machine learning model.
  • CU Centralized Unit
  • DU Distributed Unit
  • a method for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous service slave management unit in the first device receives the machine learning model file and target functional unit information sent by the intelligent endogenous service master management unit in the second device;
  • the smart endogenous service slave management unit is located in the application layer of the first device; the endogenous service master management unit is located in the application layer of the second device.
  • the intelligent endogenous service slave management unit distributes the machine learning model file to the target function unit in the first device according to the target function unit information.
  • the method further includes:
  • the target functional unit installs the machine learning model file.
  • the method further includes:
  • the target functional unit notifies the endogenous business slave management unit that the machine learning model file has been installed
  • the endogenous service slave management unit applies to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the intelligent endogenous business uses the transmission resource obtained by the application from the management unit to apply for service registration to the intelligent endogenous business main management unit;
  • the intelligent endogenous business obtains the service registration permission of the intelligent endogenous business main management unit from the management unit, the service included in the latest machine learning model installed in the target functional unit is started.
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • a method for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous service main management unit in the second device determines the target functional unit of the machine learning model to be deployed in the first device and its corresponding machine learning model file;
  • the endogenous service main management unit is located in the application layer of the second device.
  • S202 The intelligent endogenous service main management unit in the second device applies to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the intelligent endogenous service master management unit in the second device uses the transmission resource obtained by the application to send the target functional unit information and the machine learning model file to the intelligent endogenous service slave management unit in the first device. unit.
  • the intelligent endogenous service slave management unit is located in the application layer of the first device.
  • the method further includes:
  • the smart endogenous service master management unit in the second device receives the transmission resource application sent by the smart endogenous service slave management unit in the first device, and allocates the first device to the smart endogenous service slave management unit Transmission resources with the second device.
  • the method further includes:
  • the smart endogenous service main management unit in the second device receives the service registration application sent by the smart endogenous service from the management unit in the first device, and sends a service registration license for the smart endogenous service from the management unit .
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • an apparatus for transmitting machine learning model parameters in a mobile communication system includes:
  • the memory 620 is used to store program instructions
  • the processor 600 is configured to call the program instructions stored in the memory, and execute according to the obtained program:
  • the smart endogenous service slave management unit is located in the application layer of the first device; the endogenous service master management unit is located in the application layer of the second device.
  • Control the intelligent endogenous service slave management unit and distribute the machine learning model file to the target function unit in the first device according to the target function unit information.
  • processor 600 is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • processor 600 is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • the transceiver 610 is configured to receive and send data under the control of the processor 600.
  • the bus architecture may include any number of interconnected buses and bridges. Specifically, one or more processors represented by the processor 600 and various circuits of the memory represented by the memory 620 are linked together.
  • the bus architecture can also link various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., which are all known in the art, and therefore, will not be further described herein.
  • the bus interface provides the interface.
  • the transceiver 610 may be a plurality of elements, that is, including a transmitter and a receiver, and provide a unit for communicating with various other devices on a transmission medium.
  • the user interface 630 may also be an interface capable of connecting externally and internally with the required equipment.
  • the connected equipment includes but not limited to a keypad, a display, a speaker, a microphone, a joystick, and the like.
  • the processor 600 is responsible for managing the bus architecture and general processing, and the memory 620 can store data used by the processor 600 when performing operations.
  • the processor 600 may be a CPU (central embedded device), ASIC (Application Specific Integrated Circuit, application-specific integrated circuit), FPGA (Field-Programmable Gate Array, field programmable gate array) or CPLD (Complex Programmable Logic Device) , Complex Programmable Logic Devices).
  • CPU central embedded device
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array, field programmable gate array
  • CPLD Complex Programmable Logic Device
  • Complex Programmable Logic Devices Complex Programmable Logic Devices
  • an apparatus for transmitting machine learning model parameters in a mobile communication system includes:
  • the memory 520 is used to store program instructions
  • the processor 500 is configured to call the program instructions stored in the memory, and execute according to the obtained program:
  • the endogenous service main management unit is located in the application layer of the second device.
  • the intelligent endogenous service slave management unit is located in the application layer of the first device.
  • the processor 500 is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • the processor 500 is further configured to call program instructions stored in the memory, and execute according to the obtained program:
  • Control the smart endogenous service main management unit in the second device to receive the service registration application sent by the smart endogenous service from the management unit in the first device, and send service registration for the smart endogenous service from the management unit license.
  • the first device is a terminal device in a mobile communication system
  • the second device is a base station device in a mobile communication system
  • the transmission resource between the first device and the second device is a terminal device and Wireless transmission resources between base station equipment.
  • the first device is a base station distributed unit in a mobile communication system
  • the second device is a base station centralized unit in a mobile communication system
  • the transmission resources between the first device and the second device are It is the wired or wireless transmission resource between the distributed unit of the base station and the centralized unit of the base station.
  • the transceiver 510 is configured to receive and send data under the control of the processor 500.
  • the bus architecture may include any number of interconnected buses and bridges. Specifically, one or more processors represented by the processor 500 and various circuits of the memory represented by the memory 520 are linked together.
  • the bus architecture can also link various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., which are all known in the art, and therefore, will not be further described herein.
  • the bus interface provides the interface.
  • the transceiver 510 may be a plurality of elements, that is, include a transmitter and a receiver, and provide a unit for communicating with various other devices on a transmission medium.
  • the processor 500 is responsible for managing the bus architecture and general processing, and the memory 520 can store data used by the processor 500 when performing operations.
  • the processor 500 may be a central processor (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or a complex programmable logic device (Complex Programmable Logic Device). , CPLD).
  • CPU central processor
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • FPGA field programmable gate array
  • CPLD complex programmable logic device
  • another device for transmitting machine learning model parameters in a mobile communication system includes:
  • the receiving unit 11 is configured to receive the machine learning model file and target functional unit information sent by the intelligent endogenous service main management unit in the second device;
  • the sending unit 12 is configured to distribute the machine learning model file to the target function unit in the first device according to the target function unit information.
  • the device on the first device side may be the said intelligent endogenous service slave management unit.
  • another device for transmitting machine learning model parameters in a mobile communication system includes:
  • the determining unit 21 is configured to determine the target functional unit of the machine learning model to be deployed in the first device and its corresponding machine learning model file;
  • the application unit 22 is configured to apply to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the sending unit 23 is configured to send the target function unit information and the machine learning model file to the intelligent endogenous service slave management unit in the first device using the transmission resource obtained by the application.
  • the device on the second device side may be the said intelligent endogenous service main management unit.
  • the third device for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous service slave management unit 31 is configured to receive the machine learning model file and target functional unit information sent by the intelligent endogenous service main management unit in the second device; according to the target functional unit information, the machine learning model The file is distributed to the target functional unit in the first device;
  • the smart endogenous service slave management unit 31 is located in the application layer of the first device.
  • the functional unit 32 is used to install the machine learning model file.
  • the functional unit 32 is further configured to notify the endogenous business slave management unit that the machine learning model file has been installed;
  • the endogenous service slave management unit 31 is further configured to: apply to the transmission resource management unit in the second device for transmission resources between the first device and the second device;
  • the intelligent endogenous business main management unit applies for service registration; after obtaining the service registration permission of the intelligent endogenous business main management unit, the service included in the latest machine learning model installed in the target functional unit is started.
  • the third device for transmitting machine learning model parameters in a mobile communication system includes:
  • the intelligent endogenous service main management unit 41 is used to determine the target functional unit of the machine learning model to be deployed in the first device and its corresponding machine learning model file; apply to the transmission resource management unit in the second device for the first device and the second device Second, the transmission resources between the devices; using the transmission resources obtained by the application, send the target functional unit information and the machine learning model file to the intelligent endogenous service slave management unit in the first device;
  • the intelligent endogenous service main management unit 41 is located in the application layer of the second device.
  • the transmission resource management unit 42 is configured to provide transmission resources between the first device and the second device.
  • the device further includes:
  • the machine learning model library 43 is used to store machine learning model files.
  • the smart endogenous service master management unit 41 is further configured to: receive a transmission resource request sent by the smart endogenous service slave management unit in the first device, and provide the smart endogenous service slave management unit for the Allocate transmission resources between the first device and the second device.
  • the smart endogenous service master management unit 41 is further configured to: receive a service registration application sent by the smart endogenous service slave management unit in the first device, and provide the smart endogenous service slave management unit for the service registration application. Send service registration permission.
  • the division of units in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including a number of instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .
  • the embodiments of the present application provide a computing device, and the computing device may specifically be a desktop computer, a portable computer, a smart phone, a tablet computer, a personal digital assistant (Personal Digital Assistant, PDA), etc.
  • the computing device may include a central processing unit (CPU), a memory, an input/output device, etc.
  • the input device may include a keyboard, a mouse, a touch screen, etc.
  • an output device may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), Cathode Ray Tube (CRT), etc.
  • the memory may include read only memory (ROM) and random access memory (RAM), and provides the processor with program instructions and data stored in the memory.
  • ROM read only memory
  • RAM random access memory
  • the memory may be used to store the program of any of the methods provided in the embodiments of the present application.
  • the processor calls the program instructions stored in the memory, and the processor is configured to execute any of the methods provided in the embodiments of the present application according to the obtained program instructions.
  • the embodiment of the present application provides a computer storage medium for storing computer program instructions used by the device provided in the foregoing embodiment of the present application, which includes a program for executing any method provided in the foregoing embodiment of the present application.
  • the computer storage medium may be any available medium or data storage device that the computer can access, including but not limited to magnetic storage (such as floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor memory (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state drive (SSD)), etc.
  • magnetic storage such as floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.
  • optical storage such as CD, DVD, BD, HVD, etc.
  • semiconductor memory such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state drive (SSD)
  • the method provided in the embodiments of the present application can be applied to terminal equipment, and can also be applied to network equipment.
  • the terminal equipment can also be called User Equipment (User Equipment, referred to as "UE"), Mobile Station (Mobile Station, referred to as “MS”), Mobile Terminal (Mobile Terminal), etc.
  • UE User Equipment
  • MS Mobile Station
  • Mobile Terminal Mobile Terminal
  • the terminal can be It has the ability to communicate with one or more core networks via a Radio Access Network (RAN).
  • RAN Radio Access Network
  • the terminal can be a mobile phone (or called a "cellular" phone), or a mobile computer, etc.
  • the terminal may also be a portable, pocket-sized, handheld, built-in computer or vehicle-mounted mobile device.
  • the network device may be a base station (for example, an access point), which refers to a device that communicates with a wireless terminal through one or more sectors on an air interface in an access network.
  • the base station can be used to convert received air frames and IP packets into each other, and act as a router between the wireless terminal and the rest of the access network, where the rest of the access network can include an Internet Protocol (IP) network.
  • IP Internet Protocol
  • the base station can also coordinate the attribute management of the air interface.
  • the base station may be a base station (BTS, Base Transceiver Station) in GSM or CDMA, a base station (NodeB) in WCDMA, or an evolved base station (NodeB or eNB or e-NodeB, evolutional NodeB) in LTE. B), or it can also be gNB in the 5G system.
  • BTS Base Transceiver Station
  • NodeB base station
  • eNB evolved base station
  • e-NodeB evolutional NodeB
  • the processing flow of the above method can be implemented by a software program, which can be stored in a storage medium, and when the stored software program is called, the steps of the above method are executed.
  • the embodiment of the present application establishes endogenous services within the mobile communication system to support the operation of the machine learning model, and deploy and update the machine learning model through the management of endogenous services.
  • the embodiment of the application provides a unified method for deploying/updating machine learning models, which is convenient for operators to solve complex problems of wireless mobile communication systems by deploying/updating machine learning models, and improve network performance, including controlling the performance of the terminal side.
  • the method of applying endogenous services to build machine learning models can support complex machine learning models, and is more suitable for scenarios that are relatively insensitive to processing delays.
  • this application can be provided as methods, systems, or computer program products. Therefore, this application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program codes.
  • a computer-usable storage media including but not limited to disk storage, optical storage, etc.
  • These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device.
  • the device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.
  • These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment.
  • the instructions provide steps for implementing the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

一种在移动通信系统中进行机器学习模型参数传递方法及装置，用以当机器学习推演模型在无线移动通信系统内部时，实现机器学习模型的部署/更新。所述一种在移动通信系统中进行机器学习模型参数传递方法包括：第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息（S101）；所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中（S102）。

Description

机器学习模型参数传递方法及装置
相关申请的交叉引用
本申请要求在2020年04月28日提交中国专利局、申请号为202010349503.5、申请名称为“机器学习模型参数传递方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信技术领域，尤其涉及机器学习模型参数传递方法及装置。
背景技术
作为人工智能的关键方法之一，机器学习在1950年代被提出。随着机器学习技术的发展，神经网络(NN：Neural Network)或人工神经网络(ANN：Artificial Neural Network)被提出，它是受生物神经网络启发，便于在机器学习中构建模型而引入的通用的模型。一个简单的神经网络包括输入层、输出层以及隐藏层(如果需要的话)，每层包括多个神经元(Neurons)。
为了解决复杂的非线性问题，所设计出的神经网络中的隐藏层逐渐增多，形成深度神经网络(DNN：Deep Neural Network)，其对应的学习方法即深度机器学习，或深度学习。深度神经网络模型目前已经发展出多种类型，包括DNN、递归神经网络(Recurrent Neural Network，RNN)、卷积神经网络(Convolutional Neural Network，CNN)等。从2010年代开始，深度学习作为机器学习的一个重要分支，获得了广泛的关注，获得了爆发式的增长。
机器学习的发展主要体现在语音识别、图像识别等领域，在其中沉淀了大量经典的模型与算法。将机器学习引入到无线移动通信系统中，用来解决无线移动通信系统的问题，在最近几年才凸显出来。
然而，当机器学习推演模型在无线移动通信系统内部时，如何进行机器学习模型的部署/更新是一个待研究的问题。特别的，当用户终端侧的机器学 习模型需要更新时，其还影响了空口的传输。
发明内容
本申请实施例提供了机器学习模型参数传递的方法及装置，用以当机器学习推演模型在无线移动通信系统内部时，实现机器学习模型的部署/更新。
在第一设备侧，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递方法，包括：
第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
通过该方法，第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中，从而当机器学习推演模型在无线移动通信系统内部时，实现了机器学习模型的部署/更新。
可选地，所述智能内生业务从管理单元位于所述第一设备中的应用层；所述智能内生业务主管理单元位于所述第二设备中的应用层。
可选地，该方法还包括：
所述目标功能单元安装所述机器学习模型文件。
可选地，该方法还包括：
所述目标功能单元，向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
所述内生业务从管理单元，向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
所述智能内生业务从管理单元使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；
所述智能内生业务从管理单元获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
在第二设备侧，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递方法，包括：
第二设备中的智能内生业务主管理单元确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
所述第二设备中的智能内生业务主管理单元向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
所述第二设备中的智能内生业务主管理单元使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
可选地，所述智能内生业务从管理单元位于所述第一设备中的应用层；所述智能内生业务主管理单元位于所述第二设备中的应用层。
可选地，该方法还包括：
所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
可选地，该方法还包括：
所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
在第一设备侧，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递装置，包括：
存储器，用于存储程序指令；
处理器，用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
控制所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
可选地，所述智能内生业务从管理单元位于所述第一设备中的应用层；所述智能内生业务主管理单元位于所述第二设备中的应用层。
可选地，所述处理器还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述目标功能单元安装所述机器学习模型文件。
可选地，所述处理器还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述目标功能单元，向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
控制所述内生业务从管理单元，向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
控制所述智能内生业务从管理单元使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；
控制所述智能内生业务从管理单元获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
在第二设备侧，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递装置，包括：
存储器，用于存储程序指令；
处理器，用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制第二设备中的智能内生业务主管理单元确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
控制所述第二设备中的智能内生业务主管理单元向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
控制所述第二设备中的智能内生业务主管理单元使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
可选地，所述智能内生业务从管理单元位于所述第一设备中的应用层；所述智能内生业务主管理单元位于所述第二设备中的应用层。
可选地，所述处理器，还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
可选地，所述处理器，还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
在第一设备侧，本申请实施例提供的另一种在移动通信系统中进行机器学习模型参数传递装置，包括：
接收单元，用于接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
发送单元，用于根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
在第二设备侧，本申请实施例提供的另一种在移动通信系统中进行机器学习模型参数传递装置，包括：
确定单元，用于确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
申请单元，用于向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
发送单元，用于使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
在第一设备侧，本申请实施例提供的第三种在移动通信系统中进行机器学习模型参数传递装置，包括：
智能内生业务从管理单元，用于接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中；
功能单元，用于安装所述机器学习模型文件。
可选地，所述智能内生业务从管理单元位于所述第一设备中的应用层。
可选地，所述功能单元，还用于向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
所述内生业务从管理单元，还用于：向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
在第二设备侧，本申请实施例提供的第三种在移动通信系统中进行机器学习模型参数传递装置，包括：
智能内生业务主管理单元，用于确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元；
传输资源管理单元，用于提供第一设备与第二设备之间的传输资源。
可选地，所述智能内生业务主管理单元位于所述第二设备中的应用层。
可选地，该装置还包括：
机器学习模型库，用于存储机器学习模型文件。
可选地，所述智能内生业务主管理单元还用于：接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
可选地，所述智能内生业务主管理单元还用于：接收所述第一设备中的 智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
本申请另一实施例提供了一种计算设备，其包括存储器和处理器，其中，所述存储器用于存储程序指令，所述处理器用于调用所述存储器中存储的程序指令，按照获得的程序执行上述任一种方法。
本申请另一实施例提供了一种计算机存储介质，所述计算机存储介质存储有计算机可执行指令，所述计算机可执行指令用于使所述计算机执行上述任一种方法。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简要介绍，显而易见地，下面描述中的附图仅是本申请的一些实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的用户面的将内生业务的起点和终点分别构建在RAN侧的示意图；
图2为本申请实施例提供的控制面的将内生业务的起点和终点分别构建在RAN侧的示意图；
图3为本申请实施例提供的机器学习模型更新的主要流程示意图；
图4为本申请实施例提供的第一设备侧的一种机器学习模型参数传递方法的流程示意图；
图5为本申请实施例提供的第二设备侧的一种机器学习模型参数传递方法的流程示意图；
图6为本申请实施例提供的第一设备侧的一种机器学习模型参数传递装置的结构示意图；
图7为本申请实施例提供的第二设备侧的一种机器学习模型参数传递装置的结构示意图；
图8为本申请实施例提供的第一设备侧的另一种机器学习模型参数传递装置的结构示意图；
图9为本申请实施例提供的第二设备侧的另一种机器学习模型参数传递装置的结构示意图；
图10为本申请实施例提供的第一设备侧的第三种机器学习模型参数传递装置的结构示意图；
图11为本申请实施例提供的第二设备侧的第三种机器学习模型参数传递装置的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，并不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
在无线移动通信系统中引入机器学习，可以解决移动通信系统的复杂问题或提升性能。但是，由于无线移动通信系统的构成复杂、厂商众多，如何将网络侧训练好的机器学习模型部署/更新到终端侧成为一个系统性的问题，需要有统一的解决方案。
因此，本申请实施例提供了机器学习模型参数传递的方法及装置，用以实现将网络侧训练好的机器学习模型传递到终端侧。
其中，方法和装置是基于同一申请构思的，由于方法和装置解决问题的原理相似，因此装置和方法的实施可以相互参见，重复之处不再赘述。
本申请实施例提供的技术方案可以适用于多种系统，尤其是5G系统或6G系统。例如适用的系统可以是全球移动通讯(global system of mobile communication，GSM)系统、码分多址(code division multiple access，CDMA)系统、宽带码分多址(Wideband Code Division Multiple Access，WCDMA)通用分组无线业务(general packet radio service，GPRS)系统、长期演进(long  term evolution，LTE)系统、LTE频分双工(frequency division duplex，FDD)系统、LTE时分双工(time division duplex，TDD)、通用移动系统(universal mobile telecommunication system，UMTS)、全球互联微波接入(worldwide interoperability for microwave access，WiMAX)系统、5G系统以及5G NR系统等。这多种系统中均包括终端设备和网络设备。
本申请实施例涉及的终端设备，可以是指向用户提供语音和/或数据连通性的设备，具有无线连接功能的手持式设备、或连接到无线调制解调器的其他处理设备。在不同的系统中，终端设备的名称可能也不相同，例如在5G系统中，终端设备可以称为用户设备(user equipment，UE)。无线终端设备可以经RAN与一个或多个核心网进行通信，无线终端设备可以是移动终端设备，如移动电话(或称为“蜂窝”电话)和具有移动终端设备的计算机，例如，可以是便携式、袖珍式、手持式、计算机内置的或者车载的移动装置，它们与无线接入网交换语言和/或数据。例如，个人通信业务(personal communication service，PCS)电话、无绳电话、会话发起协议(session initiated protocol，SIP)话机、无线本地环路(wireless local loop，WLL)站、个人数字助理(personal digital assistant，PDA)等设备。无线终端设备也可以称为系统、订户单元(subscriber unit)、订户站(subscriber station)，移动站(mobile station)、移动台(mobile)、远程站(remote station)、接入点(access point)、远程终端设备(remote terminal)、接入终端设备(access terminal)、用户终端设备(user terminal)、用户代理(user agent)、用户装置(user device)，本申请实施例中并不限定。
本申请实施例涉及的网络设备，可以是基站，该基站可以包括多个小区。根据具体应用场合不同，基站又可以称为接入点，或者可以是指接入网中在空中接口上通过一个或多个扇区与无线终端设备通信的设备，或者其它名称。网络设备可用于将收到的空中帧与网际协议(internet protocol，IP)分组进行相互转换，作为无线终端设备与接入网的其余部分之间的路由器，其中接入网的其余部分可包括网际协议(IP)通信网络。网络设备还可协调对空中接口 的属性管理。例如，本申请实施例涉及的网络设备可以是全球移动通信系统(global system for mobile communications，GSM)或码分多址接入(code division multiple access，CDMA)中的网络设备(base transceiver station，BTS)，也可以是带宽码分多址接入(wide-band code division multiple access，WCDMA)中的网络设备(NodeB)，还可以是长期演进(long term evolution，LTE)系统中的演进型网络设备(evolutional node B，eNB或e-NodeB)、5G网络架构(next generation system)中的5G基站，也可是家庭演进基站(home evolved node B，HeNB)、中继节点(relay node)、家庭基站(femto)、微微基站(pico)等，本申请实施例中并不限定。
下面结合说明书附图对本申请各个实施例进行详细描述。需要说明的是，本申请实施例的展示顺序仅代表实施例的先后顺序，并不代表实施例所提供的技术方案的优劣。
本申请实施例提出了利用内生业务，在无线移动通信系统中将系统侧训练好的机器学习模型部署/更新到终端侧的方案。
本申请实施例所述的内生业务，例如可以是指为了解决移动通信系统复杂问题或提升性能，在移动通信系统中基站与终端之间构建的业务类型，其并不直接为用户所需要的功能服务，其服务的对象是网络本身，或者说运营商，所以称为内生业务。可以将待部署的机器学习模型(其所对应的功能)当成一种内生业务，从而机器学习模型的部署或更新可以采用业务部署或更新的方式来进行。简单的说，可以把机器学习模型看作一种应用(APP)，其模型的部署/更新即APP的下载、安装过程。采用内生业务的方式可以部署更为复杂的模型，并且也更有利于模型本身的know-how保护。
一种将内生业务的起点和终点分别构建在RAN侧(包括基站和终端)的方式，如图1和图2所示，分别包括用户面和控制面。
具体的，控制面主要执行RAN侧的内生业务的服务注册过程、无线资源申请等过程。这里以终端侧发起的服务注册过程为例进行说明，即某一具体的内生业务(比如在终端侧执行的“终端侧的信道估计”服务)，其在执行之 前需要向其所在处理器上的操作系统进行注册，注册时需要获得基站侧的授权，以避免终端肆意执行非授权的内生业务。该注册过程一般由执行该业务的主体的应用层发起(比如“终端侧的信道估计”服务的主体为终端)，经过应用层之下的无线网络层(参考5G无线网络层，其可能经过RRC、PDCP、RLC、MAC、PHY层)的处理，形成空口信号，发送给基站；基站接收到该空口信号之后，经过无线网络层(参考5G无线网络层，其可能经过PHY、MAC、RLC、PDCP、RRC层)的处理，获得来自终端发起的服务注册申请。基站审核该申请后，将确认注册的信息，经过应用层之下的无线网络层处理后，再发送给终端；终端接收到该空口信号后，经过无线网络层的处理后，获得基站侧的授权信息。
用户面主要处理RAN侧的内生业务中的用户数据。该数据可能由应用层生成(比如终端执行测量业务所使用的测试数据)，也有可能受控制面的控制在无线网络层中生成(比如“终端侧的信道估计”业务中所使用的参考信号，它在无线网络层中由PHY层生成)。这里以终端执行测量业务所使用的测试数据为例进行说明，该数据由基站侧的应用层生成，经过无线网络层(参考5G无线网络层，其可能经过SDAP、PDCP、RLC、MAC、PHY层)的处理，形成空口信号，发送给终端；终端侧接收到该空口信号后，调用测量业务对该空口信号进行处理。
机器学习模型更新的主要流程如图3所示，基于上述所构建的RAN系统内生业务，一种具体的机器学习模型部署/更新方法如下：
步骤1、当基站决策对终端侧目标功能单元部署/更新机器学习模型时，基站中的智能内生业务主管理单元，从机器学习模型库中选取对应的机器学习模型文件；
其中，所述目标功能单元比如信道估计、信号检测、信道译码、信道质量指示(Channel Quality Indicator，CQI)测量、同步检测、发送天线选择等。所述智能内生业务主管理单元位于基站侧的应用层，主要用于为内生业务提供服务注册功能、申请终端与基站之间的传输资源的功能。所述机器学习模 型文件一般可理解的可执行程序文件，比如.exe。一般包括多个文件，并打包为一个压缩文件包，比如.zip。
步骤2、基站中的智能内生业务主管理单元向基站侧的传输资源管理单元申请终端与基站之间的传输资源；
步骤3、基站中的智能内生业务主管理单元使用申请获得的传输资源，将机器学习模型文件从机器学习模型库发送给终端侧的智能内生业务从管理单元，并伴随发送对应的目标功能单元信息；
所述智能内生业务从管理单元位于终端侧的应用层。
所述目标功能单元信息例如包括目标功能单元的名称，如表1中第二列所示。
步骤4、终端侧的智能内生业务从管理单元根据接收到的目标功能单元信息，将机器学习模型文件分发到终端侧的目标功能单元中；
步骤5、终端侧的目标功能单元接收到智能内生业务从管理单元分发的机器学习模型文件后，进行安装，并通知内生业务从管理单元；
步骤6、终端侧的内生业务从管理单元，向基站侧的传输资源管理单元申请终端与基站之间的传输资源；
步骤7、终端侧的智能内生业务从管理单元使用申请获得的传输资源，向基站中的智能内生业务主管理单元申请服务注册；
步骤8、终端侧的智能内生业务从管理单元获得基站侧的智能内生业务主管理单元的服务注册许可后，启动目标功能单元中新部署/更新的机器学习模型所包含的服务。
例如，一种机器学习模型库的列表如下表1所示，其中根据机器学习模型编号可以找到与之唯一对应的机器学习模型文件压缩包。 表1
Figure PCTCN2021076843-appb-000001
对应的，本申请实施例给出了一种利用内生业务部署/更新机器学习模型的系统，它至少包括基站侧和终端侧，其中基站侧包括智能内生业务主管理单元、传输资源管理单元、以及机器学习模型库；终端侧包括智能内生业务从管理单元和可应用机器学习模型的目标功能单元。
另外需要说明的是，除了可以将内生业务部署在基站和终端上之外，在系统侧存在多个单元时，比如中心单元(Centralized Unit，CU)和分布式单元(Distributed Unit，DU)，也可以将内生业务部署在CU与DU上，从而利用类似的方式进行机器学习模型的部署与更新。
综上所述，参见图4，在第一设备侧，例如终端或者基站分布式单元，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递方法，包括：
S101、第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
可选地，所述智能内生业务从管理单元位于第一设备中的应用层；所述内生业务主管理单元位于第二设备中的应用层。
S102、所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
可选地，该方法还包括：
所述目标功能单元安装所述机器学习模型文件。
可选地，该方法还包括：
所述目标功能单元，向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
所述内生业务从管理单元，向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
所述智能内生业务从管理单元使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；
所述智能内生业务从管理单元获得所述智能内生业务主管理单元的服务 注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
参见图5，在第二设备侧，例如基站或基站的集中式单元，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递方法，包括：
S201、第二设备中的智能内生业务主管理单元确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
可选地，所述内生业务主管理单元位于第二设备中的应用层。
S202、所述第二设备中的智能内生业务主管理单元向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
S203、所述第二设备中的智能内生业务主管理单元使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
可选地，所述智能内生业务从管理单元位于第一设备中的应用层。
可选地，该方法还包括：
所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
可选地，该方法还包括：
所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是 移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
参见图6，在第一设备侧，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递装置，包括：
存储器620，用于存储程序指令；
处理器600，用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
可选地，所述智能内生业务从管理单元位于第一设备中的应用层；所述内生业务主管理单元位于第二设备中的应用层。
控制所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
可选地，所述处理器600还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述目标功能单元安装所述机器学习模型文件。
可选地，所述处理器600还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述目标功能单元，向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
控制所述内生业务从管理单元，向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
控制所述智能内生业务从管理单元使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；
控制所述智能内生业务从管理单元获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
收发机610，用于在处理器600的控制下接收和发送数据。
其中，在图6中，总线架构可以包括任意数量的互联的总线和桥，具体由处理器600代表的一个或多个处理器和存储器620代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起，这些都是本领域所公知的，因此，本文不再对其进行进一步描述。总线接口提供接口。收发机610可以是多个元件，即包括发送机和接收机，提供用于在传输介质上与各种其他装置通信的单元。针对不同的用户设备，用户接口630还可以是能够外接内接需要设备的接口，连接的设备包括但不限于小键盘、显示器、扬声器、麦克风、操纵杆等。
处理器600负责管理总线架构和通常的处理，存储器620可以存储处理器600在执行操作时所使用的数据。
可选的，处理器600可以是CPU(中央处埋器)、ASIC(Application Specific Integrated Circuit，专用集成电路)、FPGA(Field－Programmable Gate Array，现场可编程门阵列)或CPLD(Complex Programmable Logic Device，复杂可编程逻辑器件)。
参见图7，在第二设备侧，本申请实施例提供的一种在移动通信系统中进行机器学习模型参数传递装置，包括：
存储器520，用于存储程序指令；
处理器500，用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制第二设备中的智能内生业务主管理单元确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
可选地，所述内生业务主管理单元位于第二设备中的应用层。
控制所述第二设备中的智能内生业务主管理单元向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
控制所述第二设备中的智能内生业务主管理单元使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
可选地，所述智能内生业务从管理单元位于第一设备中的应用层。
可选地，所述处理器500，还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
可选地，所述处理器500，还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
控制所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
可选地，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
可选地，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
收发机510，用于在处理器500的控制下接收和发送数据。
其中，在图7中，总线架构可以包括任意数量的互联的总线和桥，具体由处理器500代表的一个或多个处理器和存储器520代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起，这些都是本领域所公知的，因此，本文不再对其进行进一步描述。总线接口提供接口。收发机510可以是多个元件，即包括发送机和接收机，提供用于在传输介质上与各种其他装置通信的单元。处理器500负责管理总线架构和通常的处理，存储器520可以存储处理器500在执行操作时所使用的数据。
处理器500可以是中央处埋器(CPU)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field－Programmable Gate Array，FPGA)或复杂可编程逻辑器件(Complex Programmable Logic Device，CPLD)。
参见图8，在第一设备侧，本申请实施例提供的另一种在移动通信系统中进行机器学习模型参数传递装置，包括：
接收单元11，用于接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
发送单元12，用于根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
例如，该第一设备侧的装置可以是所述的智能内生业务从管理单元。
参见图9，在第二设备侧，本申请实施例提供的另一种在移动通信系统中进行机器学习模型参数传递装置，包括：
确定单元21，用于确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
申请单元22，用于向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
发送单元23，用于使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
例如，该第二设备侧的装置可以是所述的智能内生业务主管理单元。
参见图10，在第一设备侧，本申请实施例提供的第三种在移动通信系统中进行机器学习模型参数传递装置，包括：
智能内生业务从管理单元31，用于接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中；
可选地，所述智能内生业务从管理单元31位于第一设备中的应用层。
功能单元32，用于安装所述机器学习模型文件。
可选地，所述功能单元32，还用于向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
所述内生业务从管理单元31，还用于：向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
参见图11，在第二设备侧，本申请实施例提供的第三种在移动通信系统中进行机器学习模型参数传递装置，包括：
智能内生业务主管理单元41，用于确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元；
可选地，所述智能内生业务主管理单元41位于第二设备中的应用层。
传输资源管理单元42，用于提供第一设备与第二设备之间的传输资源。
可选地，该装置还包括：
机器学习模型库43，用于存储机器学习模型文件。
可选地，所述智能内生业务主管理单元41还用于：接收所述第一设备中 的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
可选地，所述智能内生业务主管理单元41还用于：接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
需要说明的是，本申请实施例中对单元的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
本申请实施例提供了一种计算设备，该计算设备具体可以为桌面计算机、便携式计算机、智能手机、平板电脑、个人数字助理(Personal Digital Assistant，PDA)等。该计算设备可以包括中央处理器(Center Processing Unit，CPU)、存储器、输入/输出设备等，输入设备可以包括键盘、鼠标、触摸屏等，输出设备可以包括显示设备，如液晶显示器(Liquid Crystal Display，LCD)、阴极射线管(Cathode Ray Tube，CRT)等。
存储器可以包括只读存储器(ROM)和随机存取存储器(RAM)，并向处理器提供存储器中存储的程序指令和数据。在本申请实施例中，存储器可 以用于存储本申请实施例提供的任一所述方法的程序。
处理器通过调用存储器存储的程序指令，处理器用于按照获得的程序指令执行本申请实施例提供的任一所述方法。
本申请实施例提供了一种计算机存储介质，用于储存为上述本申请实施例提供的装置所用的计算机程序指令，其包含用于执行上述本申请实施例提供的任一方法的程序。
所述计算机存储介质可以是计算机能够存取的任何可用介质或数据存储设备，包括但不限于磁性存储器(例如软盘、硬盘、磁带、磁光盘(MO)等)、光学存储器(例如CD、DVD、BD、HVD等)、以及半导体存储器(例如ROM、EPROM、EEPROM、非易失性存储器(NAND FLASH)、固态硬盘(SSD))等。
本申请实施例提供的方法可以应用于终端设备，也可以应用于网络设备。
其中，终端设备也可称之为用户设备(User Equipment，简称为“UE”)、移动台(Mobile Station，简称为“MS”)、移动终端(Mobile Terminal)等，可选的，该终端可以具备经无线接入网(Radio Access Network，RAN)与一个或多个核心网进行通信的能力，例如，终端可以是移动电话(或称为“蜂窝”电话)、或具有移动性质的计算机等，例如，终端还可以是便携式、袖珍式、手持式、计算机内置的或者车载的移动装置。
网络设备可以为基站(例如，接入点)，指接入网中在空中接口上通过一个或多个扇区与无线终端通信的设备。基站可用于将收到的空中帧与IP分组进行相互转换，作为无线终端与接入网的其余部分之间的路由器，其中接入网的其余部分可包括网际协议(IP)网络。基站还可协调对空中接口的属性管理。例如，基站可以是GSM或CDMA中的基站(BTS，Base Transceiver Station)，也可以是WCDMA中的基站(NodeB)，还可以是LTE中的演进型基站(NodeB或eNB或e-NodeB，evolutional Node B)，或者也可以是5G系统中的gNB等。本申请实施例中不做限定。
上述方法处理流程可以用软件程序实现，该软件程序可以存储在存储介 质中，当存储的软件程序被调用时，执行上述方法步骤。
综上所述，本申请实施例在移动通信系统内部建立内生业务，用来支撑机器学习模型的运行，以及通过对内生业务的管理，进行机器学习模型的部署与更新。本申请实施例给出了一种机器学习模型部署/更新的统一方法，便于运营商通过部署/更新机器学习模型解决无线移动通信系统的复杂问题，提升网络性能，包括控制终端侧的性能。应用内生业务构建机器学习模型的方法可以支撑复杂的机器学习模型，更适用于对处理时延相对不敏感的场景。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图 一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (35)

  1. 一种在移动通信系统中进行机器学习模型参数传递方法，其特征在于，该方法包括：
    第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
    所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
  2. 根据权利要求1所述的方法，其特征在于，所述智能内生业务从管理单元位于所述第一设备中的应用层；所述智能内生业务主管理单元位于所述第二设备中的应用层。
  3. 根据权利要求1所述的方法，其特征在于，该方法还包括：
    所述目标功能单元安装所述机器学习模型文件。
  4. 根据权利要求3所述的方法，其特征在于，该方法还包括：
    所述目标功能单元，向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
    所述内生业务从管理单元，向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
    所述智能内生业务从管理单元使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；
    所述智能内生业务从管理单元获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
  5. 根据权利要求4所述的方法，其特征在于，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
  6. 根据权利要求4所述的方法，其特征在于，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单 元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
  7. 一种在移动通信系统中进行机器学习模型参数传递方法，其特征在于，该方法包括：
    第二设备中的智能内生业务主管理单元确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
    所述第二设备中的智能内生业务主管理单元向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
    所述第二设备中的智能内生业务主管理单元使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
  8. 根据权利要求7所述的方法，其特征在于，所述智能内生业务从管理单元位于第一设备中的应用层；所述智能内生业务主管理单元位于第二设备中的应用层。
  9. 根据权利要求7所述的方法，其特征在于，该方法还包括：
    所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
  10. 根据权利要求9所述的方法，其特征在于，该方法还包括：
    所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
  11. 根据权利要求7～10任一权项所述的方法，其特征在于，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
  12. 根据权利要求7～10任一权项所述的方法，其特征在于，所述第一设 备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
  13. 一种在移动通信系统中进行机器学习模型参数传递装置，其特征在于，包括：
    存储器，用于存储程序指令；
    处理器，用于调用所述存储器中存储的程序指令，按照获得的程序执行：
    控制第一设备中的智能内生业务从管理单元，接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
    控制所述智能内生业务从管理单元，根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
  14. 根据权利要求13所述的装置，其特征在于，所述智能内生业务从管理单元位于第一设备中的应用层；所述智能内生业务主管理单元位于第二设备中的应用层。
  15. 根据权利要求13所述的装置，其特征在于，所述处理器还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
    控制所述目标功能单元安装所述机器学习模型文件。
  16. 根据权利要求15所述的装置，其特征在于，所述处理器还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
    控制所述目标功能单元，向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
    控制所述内生业务从管理单元，向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
    控制所述智能内生业务从管理单元使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；
    控制所述智能内生业务从管理单元获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的 服务。
  17. 根据权利要求16所述的装置，其特征在于，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
  18. 根据权利要求16所述的装置，其特征在于，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
  19. 一种在移动通信系统中进行机器学习模型参数传递装置，其特征在于，该装置包括：
    存储器，用于存储程序指令；
    处理器，用于调用所述存储器中存储的程序指令，按照获得的程序执行：
    控制第二设备中的智能内生业务主管理单元确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
    控制所述第二设备中的智能内生业务主管理单元向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
    控制所述第二设备中的智能内生业务主管理单元使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
  20. 根据权利要求19所述的装置，其特征在于，所述智能内生业务从管理单元位于第一设备中的应用层；所述智能内生业务主管理单元位于第二设备中的应用层。
  21. 根据权利要求20所述的装置，其特征在于，所述处理器，还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
    控制所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管 理单元分配第一设备与第二设备之间的传输资源。
  22. 根据权利要求21所述的装置，其特征在于，所述处理器，还用于调用所述存储器中存储的程序指令，按照获得的程序执行：
    控制所述第二设备中的智能内生业务主管理单元接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
  23. 根据权利要求20～22任一权项所述的装置，其特征在于，所述第一设备是移动通信系统中的终端设备，所述第二设备是移动通信系统中的基站设备，所述第一设备与第二设备之间的传输资源是终端设备与基站设备之间的无线传输资源。
  24. 根据权利要求20～22任一权项所述的装置，其特征在于，所述第一设备是移动通信系统中的基站分布式单元，所述第二设备是移动通信系统中的基站集中式单元，所述第一设备与第二设备之间的传输资源是基站分布式单元与基站集中式单元之间的有线或无线传输资源。
  25. 一种在移动通信系统中进行机器学习模型参数传递装置，其特征在于，该装置包括：
    接收单元，用于接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；
    发送单元，用于根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中。
  26. 一种在移动通信系统中进行机器学习模型参数传递装置，其特征在于，该装置包括：
    确定单元，用于确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；
    申请单元，用于向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；
    发送单元，用于使用申请获得的所述传输资源，将目标功能单元信息以 及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元。
  27. 一种在移动通信系统中进行机器学习模型参数传递装置，其特征在于，该装置包括：
    智能内生业务从管理单元，用于接收第二设备中的智能内生业务主管理单元发送的机器学习模型文件以及目标功能单元信息；根据所述目标功能单元信息，将所述机器学习模型文件分发到第一设备中的目标功能单元中；
    功能单元，用于安装所述机器学习模型文件。
  28. 根据权利要求27所述的装置，其特征在于，所述智能内生业务从管理单元位于第一设备中的应用层。
  29. 根据权利要求28所述的装置，其特征在于，所述功能单元，还用于向所述内生业务从管理单元通知所述机器学习模型文件已完成安装；
    所述内生业务从管理单元，还用于：向所述第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；使用申请获得的所述传输资源，向所述智能内生业务主管理单元申请服务注册；获得所述智能内生业务主管理单元的服务注册许可后，启动目标功能单元中已安装的最新机器学习模型所包含的服务。
  30. 一种在移动通信系统中进行机器学习模型参数传递装置，其特征在于，该装置包括：
    智能内生业务主管理单元，用于确定第一设备中待部署机器学习模型的目标功能单元及其对应的机器学习模型文件；向第二设备中的传输资源管理单元申请第一设备与第二设备之间的传输资源；使用申请获得的所述传输资源，将目标功能单元信息以及所述机器学习模型文件发送给第一设备中的智能内生业务从管理单元；
    传输资源管理单元，用于提供第一设备与第二设备之间的传输资源。
  31. 根据权利要求30所述的装置，其特征在于，所述智能内生业务主管理单元位于第二设备中的应用层。
  32. 根据权利要求30所述的装置，其特征在于，该装置还包括：
    机器学习模型库，用于存储机器学习模型文件。
  33. 根据权利要求30所述的装置，其特征在于，所述智能内生业务主管理单元还用于：接收所述第一设备中的智能内生业务从管理单元发送的传输资源申请，并为所述智能内生业务从管理单元分配第一设备与第二设备之间的传输资源。
  34. 根据权利要求33所述的装置，其特征在于，所述智能内生业务主管理单元还用于：接收所述第一设备中的智能内生业务从管理单元发送的服务注册申请，并为所述智能内生业务从管理单元发送服务注册许可。
  35. 一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机可执行指令，所述计算机可执行指令用于使所述计算机执行权利要求1至12任一项所述的方法。
PCT/CN2021/076843 2020-04-28 2021-02-19 机器学习模型参数传递方法及装置 Ceased WO2021218302A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP21795817.2A EP4145359B1 (en) 2020-04-28 2021-02-19 Method and apparatus for transferring machine learning model parameter
KR1020227040107A KR20230002801A (ko) 2020-04-28 2021-02-19 기계 학습 모델 매개변수를 전달하기 위한 방법 및 장치
JP2022565894A JP7418610B2 (ja) 2020-04-28 2021-02-19 機械学習モデルパラメータを転送するための方法および装置
US17/921,642 US20230169398A1 (en) 2020-04-28 2021-02-19 Method and apparatus for transferring machine learning model parameter

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010349503.5A CN113570063B (zh) 2020-04-28 2020-04-28 机器学习模型参数传递方法及装置
CN202010349503.5 2020-04-28

Publications (1)

Publication Number Publication Date
WO2021218302A1 true WO2021218302A1 (zh) 2021-11-04

Family

ID=78157909

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/076843 Ceased WO2021218302A1 (zh) 2020-04-28 2021-02-19 机器学习模型参数传递方法及装置

Country Status (6)

Country Link
US (1) US20230169398A1 (zh)
EP (1) EP4145359B1 (zh)
JP (1) JP7418610B2 (zh)
KR (1) KR20230002801A (zh)
CN (1) CN113570063B (zh)
WO (1) WO2021218302A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114302506A (zh) * 2021-12-24 2022-04-08 中国联合网络通信集团有限公司 一种基于人工智能的协议栈、数据处理方法和装置
CN116347426A (zh) * 2021-12-22 2023-06-27 中国电信股份有限公司 无线接入网及其节点、数据收集和处理方法和系统
WO2024244612A1 (zh) * 2023-06-01 2024-12-05 大唐移动通信设备有限公司 通信系统、通信方法及装置
EP4422332A4 (en) * 2021-10-22 2025-02-19 Datang Mobile Communications Equipment Co., Ltd. ENDOGENOUS SERVICE TRANSMISSION METHOD AND DEVICE AS WELL AS STORAGE MEDIUM

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023097522A1 (zh) * 2021-11-30 2023-06-08 Oppo广东移动通信有限公司 用于无线信道处理的模型更新方法、装置、设备及介质
CN118414842A (zh) * 2022-01-17 2024-07-30 深圳Tcl新技术有限公司 用户设备、基站和基于机器学习/人工智能的无线通信方法
WO2024263820A2 (en) * 2023-06-20 2024-12-26 Xeba Technologies, LLC Object messaging and intelligent objects (omio)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108076471A (zh) * 2016-11-14 2018-05-25 华为技术有限公司 一种管理ran切片的方法和装置
CN109495485A (zh) * 2018-11-29 2019-03-19 深圳市永达电子信息股份有限公司 支持强制访问控制的全双工防火墙防护方法
US20190132203A1 (en) * 2017-10-31 2019-05-02 Myndshft Technologies, Inc. System and method for configuring an adaptive computing cluster
CN110147915A (zh) * 2018-02-11 2019-08-20 陕西爱尚物联科技有限公司 一种资源配置的方法及其系统
CN110263935A (zh) * 2019-06-14 2019-09-20 安徽奔奔科技有限公司 基于机器学习算法的人工智能行业在线建模平台

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10512022B2 (en) * 2013-11-05 2019-12-17 Sharp Kabushiki Kaisha Terminal apparatus, relay terminal apparatus, and communication control method
WO2018053717A1 (en) 2016-09-21 2018-03-29 Accenture Global Solutions Limited Dynamic resource allocation for application containers
US11074599B2 (en) * 2016-12-08 2021-07-27 App Annie Inc. Determining usage data of mobile applications for a population
US10229312B2 (en) * 2016-12-30 2019-03-12 Facebook, Inc. Systems and methods for providing augmented reality overlays
CN106815644B (zh) * 2017-01-26 2019-05-03 北京航空航天大学 机器学习方法和系统
CN107885762B (zh) * 2017-09-19 2021-06-11 北京百度网讯科技有限公司 智能大数据系统、提供智能大数据服务的方法和设备
US12293260B2 (en) 2017-11-21 2025-05-06 Amazon Technologies, Inc. Generating and deploying packages for machine learning at edge devices
CN108280522B (zh) * 2018-01-03 2021-08-20 北京大学 一种插件式分布式机器学习计算框架及其数据处理方法
CN110119808A (zh) * 2018-02-06 2019-08-13 华为技术有限公司 一种基于机器学习的数据处理方法以及相关设备
US10496936B2 (en) * 2018-03-05 2019-12-03 Capital One Services, Llc Systems and methods for preventing machine learning models from negatively affecting mobile devices through intermittent throttling
CN109255234B (zh) * 2018-08-15 2023-03-24 腾讯科技（深圳）有限公司 机器学习模型的处理方法、装置、介质及电子设备
CN109445953A (zh) * 2018-08-30 2019-03-08 北京大学 一种面向大规模机器学习系统的机器学习模型训练方法
CN109409528A (zh) * 2018-09-10 2019-03-01 平安科技（深圳）有限公司 模型生成方法、装置、计算机设备及存储介质
WO2020080989A1 (en) 2018-10-19 2020-04-23 Telefonaktiebolaget Lm Ericsson (Publ) Handling of machine learning to improve performance of a wireless communications network
US11423254B2 (en) * 2019-03-28 2022-08-23 Intel Corporation Technologies for distributing iterative computations in heterogeneous computing environments
EP3948821B1 (en) * 2019-04-05 2024-09-25 NEC Corporation Method and system for supporting autonomous driving of an autonomous vehicle
US11170322B2 (en) * 2019-08-13 2021-11-09 Salesforce.Com, Inc. Using a client-side machine learning model dynamically in a mobile enterprise application on a mobile device
US20210081920A1 (en) * 2019-09-13 2021-03-18 Capital One Services, Llc Systems and methods for determining a mode of financial transaction based on identifiers associated with participants of the financial transaction
WO2021183955A1 (en) * 2020-03-13 2021-09-16 PAIGE.AI, Inc. Systems and methods of automatically processing electronic images across regions
US12235840B2 (en) * 2021-11-16 2025-02-25 Google Llc Database query optimization via parameter-sensitive plan selection

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108076471A (zh) * 2016-11-14 2018-05-25 华为技术有限公司 一种管理ran切片的方法和装置
US20190132203A1 (en) * 2017-10-31 2019-05-02 Myndshft Technologies, Inc. System and method for configuring an adaptive computing cluster
CN110147915A (zh) * 2018-02-11 2019-08-20 陕西爱尚物联科技有限公司 一种资源配置的方法及其系统
CN109495485A (zh) * 2018-11-29 2019-03-19 深圳市永达电子信息股份有限公司 支持强制访问控制的全双工防火墙防护方法
CN110263935A (zh) * 2019-06-14 2019-09-20 安徽奔奔科技有限公司 基于机器学习算法的人工智能行业在线建模平台

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
LAN CAO MEI: "Application of AI and 5G in wireless communication", 14 April 2020 (2020-04-14), pages 1 - 6, XP055861485, Retrieved from the Internet <URL:https://cloud.tencent.com/developer/article/1614274> *
LIU YUANYUAN: "5G is here, 6G is also beginning to "pathfinder"", 22 November 2019 (2019-11-22), pages 1 - 6, XP055861474, Retrieved from the Internet <URL:http://tc.people.com.cn/n1/2019/1122/c183008-31468468.html> *
See also references of EP4145359A4

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4422332A4 (en) * 2021-10-22 2025-02-19 Datang Mobile Communications Equipment Co., Ltd. ENDOGENOUS SERVICE TRANSMISSION METHOD AND DEVICE AS WELL AS STORAGE MEDIUM
US12317358B2 (en) 2021-10-22 2025-05-27 Datang Mobile Communications Equipment Co., Ltd. Transmission method and apparatus for ai-native service, and storage medium
CN116347426A (zh) * 2021-12-22 2023-06-27 中国电信股份有限公司 无线接入网及其节点、数据收集和处理方法和系统
CN114302506A (zh) * 2021-12-24 2022-04-08 中国联合网络通信集团有限公司 一种基于人工智能的协议栈、数据处理方法和装置
CN114302506B (zh) * 2021-12-24 2023-06-30 中国联合网络通信集团有限公司 一种基于人工智能ai的协议栈单元、数据处理方法和装置
WO2024244612A1 (zh) * 2023-06-01 2024-12-05 大唐移动通信设备有限公司 通信系统、通信方法及装置

Also Published As

Publication number Publication date
JP7418610B2 (ja) 2024-01-19
EP4145359C0 (en) 2025-10-29
CN113570063A (zh) 2021-10-29
JP2023523073A (ja) 2023-06-01
EP4145359A1 (en) 2023-03-08
EP4145359A4 (en) 2023-10-18
US20230169398A1 (en) 2023-06-01
KR20230002801A (ko) 2023-01-05
CN113570063B (zh) 2024-04-30
EP4145359B1 (en) 2025-10-29

Similar Documents

Publication Publication Date Title
CN113570063B (zh) 机器学习模型参数传递方法及装置
CN113570062B (zh) 机器学习模型参数传递方法及装置
CN115334599A (zh) 一种小区切换及其控制方法及装置
CN112752253B (zh) 消息传输方法及装置
CN112217618B (zh) 信息传输方法及装置
WO2021223637A1 (zh) 应用迁移方法及装置
CN111200841B (zh) 消息发送及其控制方法及装置
CN114342468A (zh) 信息更新方法、设备及系统
WO2021068830A1 (zh) 用于多播传输的方法和装置
US20250081148A1 (en) Prs resource determination method and apparatus
CN112055387A (zh) 一种小区切换方法及装置
CN110474742A (zh) 一种确定rrm测量配置的方法及设备
WO2024016970A1 (zh) L1/l2移动性配置信息的处理方法、装置及存储介质
US12588103B2 (en) Method for maintaining DRX and/or DTX state of sidelink communication interface, and sidelink communication terminals
US10588034B1 (en) Antenna farm intelligent software defined networking enabled dynamic resource controller in advanced networks
WO2020088204A1 (zh) 一种数据传输的方法及装置
US20190069125A1 (en) Determining location of a user device in a wireless communication network based on location information stored on the user device
US20240260122A1 (en) Inactive state ue context handling in service-based (r)adio access network
EP4340438A1 (en) Positioning method and apparatus
WO2020020057A1 (zh) 用于半静态授权上行传输的方法、终端设备和网络侧设备
US10805787B2 (en) Determining location of a femtocell using a mobile device transmitting over an unlicensed band
CN114501612A (zh) 一种资源分配方法及终端、网络设备、存储介质
CN112929896B (zh) 一种信息传输方法及装置
US20240171341A1 (en) Method and apparatus for determining prs configuration information
WO2026021098A1 (zh) 算力参数处理方法、装置及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21795817

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022565894

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20227040107

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021795817

Country of ref document: EP

Effective date: 20221128

WWG Wipo information: grant in national office

Ref document number: 2021795817

Country of ref document: EP