CN115699842A - Model management method, system, device, communication equipment and storage medium - Google Patents

Model management method, system, device, communication equipment and storage medium Download PDF

Info

Publication number
CN115699842A
CN115699842A CN202080101791.0A CN202080101791A CN115699842A CN 115699842 A CN115699842 A CN 115699842A CN 202080101791 A CN202080101791 A CN 202080101791A CN 115699842 A CN115699842 A CN 115699842A
Authority
CN
China
Prior art keywords
model
network element
identification
model management
type
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.)
Pending
Application number
CN202080101791.0A
Other languages
Chinese (zh)
Inventor
许阳
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.)
Guangdong Oppo Mobile Telecommunications Corp Ltd
Original Assignee
Guangdong Oppo Mobile Telecommunications Corp 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 Guangdong Oppo Mobile Telecommunications Corp Ltd filed Critical Guangdong Oppo Mobile Telecommunications Corp Ltd
Publication of CN115699842A publication Critical patent/CN115699842A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/60—Software deployment
    • G06F8/65—Updates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Security & Cryptography (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The application discloses a model management method, a system, a device, a communication device and a storage medium, relating to the field of mobile communication. The method is applied to terminal equipment, and the terminal equipment comprises the following steps: an AI model management module, the method comprising: the AI model management module and the AI model management network element execute a first interactive process; the AI model management module is a functional module used for AI model management at a terminal equipment end, the AI model management network element is used for AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload AI models for an application layer and/or a wireless protocol layer in the terminal equipment.

Description

Model management method, system, device, communication equipment and storage medium Technical Field
The present application relates to the field of mobile communications, and in particular, to a model management method, system, apparatus, communication device, and storage medium.
Background
With the development of Artificial Intelligence (AI) technology, big data analysis based on AI models has become a trend.
Under the condition that the terminal device uses the AI model to participate in big data analysis, a third party application provider of an application corresponding to the AI model needs to store and manage the AI model used by the terminal device.
Because the storage space corresponding to a single AI model is large, and the number of AI models used by the terminal device is large, the resource overhead is large when the third-party application provider manages the AI models.
Disclosure of Invention
The embodiment of the application provides a model management method, a model management system, a model management device, communication equipment and a storage medium, so that AI model management can be performed without occupying machine room resources of a third-party application business local place, and resource use of an enterprise is guaranteed. The technical scheme is as follows.
According to an aspect of the present application, there is provided a model management method applied in a terminal device, where the terminal device includes: an AI model management module, the method comprising:
the AI model management module and the AI model management network element execute a first interaction process;
the AI model management module is a functional module used for AI model management at a terminal equipment end, the AI model management network element is used for AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload AI models for an application layer and/or a wireless protocol layer in the terminal equipment.
According to an aspect of the present application, there is provided a model management method applied in a data network, the data network including: an AI model managing network element, the method comprising:
the AI model management network element and the AI model management module execute a first interaction process;
the AI model management module is a functional module used for terminal equipment end AI model management, the AI model management network element is a network element used for data network end AI model management, and the AI model management module supports interaction with the AI model management network element so as to download or upload AI models for an application layer and/or a wireless protocol layer in the terminal equipment.
According to an aspect of the present application, there is provided a model management system including: an AI model management module and an AI model management network element;
the AI model management module is used for executing a first interactive process with the AI model management network element;
the AI model management module is a functional module used for AI model management at a terminal equipment end, the AI model management network element is used for AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload an AI model for an application layer and/or a radio protocol layer in the terminal equipment.
According to an aspect of the present application, there is provided a model management apparatus applied in a terminal device, the apparatus including: an AI model management module;
the AI model management module is used for executing a first interactive process with the AI model management network element;
the AI model management module is a functional module used for AI model management at a terminal equipment end, the AI model management network element is used for AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload AI models for an application layer and/or a wireless protocol layer in the terminal equipment.
According to an aspect of the present application, there is provided a model management apparatus for use in a data network, the apparatus comprising:
an AI model management network element module;
the AI model management network element module is used for executing a first interactive process with the AI model management module;
the AI model management module is a functional module used for terminal equipment end AI model management, the AI model management network element module is a network element module used for data network end AI model management, and the AI model management module supports interaction with the AI model management network element module so as to download or upload AI models for application layer and/or wireless protocol layer in the terminal equipment.
According to an aspect of the present application, there is provided a terminal device, including: a processor; a transceiver coupled to the processor; a memory for storing executable instructions of the processor; wherein the processor is configured to load and execute the executable instructions to implement the model management method as described in the above aspect.
According to an aspect of the present application, there is provided a data network comprising: a processor; a transceiver coupled to the processor; a memory for storing executable instructions of the processor; wherein the processor is configured to load and execute the executable instructions to implement the model management method as described in the above aspect.
According to one aspect of the present application, there is provided a computer-readable storage medium having stored therein executable instructions that are loaded and executed by a processor to implement the model management method according to the above aspect.
According to an aspect of the present application, there is provided a computer program product or a computer program comprising computer instructions stored in a computer-readable storage medium, the computer instructions being read by a processor of a computer device from the computer-readable storage medium, the computer instructions being executed by the processor to cause the computer device to perform the model management method of the above aspect.
The technical scheme provided by the embodiment of the application at least comprises the following beneficial effects:
the task of AI model management is handed over to be executed by the AI model management module and the AI model management network element, because the AI model management module is a functional module used for carrying out AI model management at the terminal equipment end, and the AI model management network element is used for carrying out AI model management at the data network end, the AI model management can be carried out without occupying the local machine room resources of a third party application business, thereby ensuring the resource use of enterprises.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic view of a scenario for big data analysis based on an AI model according to an exemplary embodiment of the present application;
FIG. 2 is a schematic diagram of a communication system provided by an exemplary embodiment of the present application;
FIG. 3 is a flow chart of a model management method provided by an exemplary embodiment of the present application;
FIG. 4 is a schematic illustration of a model management method provided in an exemplary embodiment of the present application;
FIG. 5 is a schematic diagram of a model management system provided by an exemplary embodiment of the present application;
FIG. 6 is a block diagram of a model management apparatus according to an exemplary embodiment of the present application;
FIG. 7 is a block diagram illustrating a model management apparatus according to an exemplary embodiment of the present application;
fig. 8 is a schematic structural diagram of a communication device according to an exemplary embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
First, terms referred to in the embodiments of the present application are briefly described:
artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.
The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence base technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
The AI model is a model corresponding to an artificial intelligence technique.
In a scenario of performing big data analysis based on the AI model, in order to improve the effect of big data analysis and user experience, a multi-level AI model may be adopted, that is, the network element and the terminal device on the network side perform big data analysis in a labor-sharing manner.
A) in fig. 1 is a centralized scenario, that is, after all terminal devices report required data, the big data analysis is performed by all servers on the network side. B) in fig. 1 is a completely distributed scenario, i.e. different terminal devices analyze the acquired data locally. Fig. 1 c) is a hybrid scenario, that is, after the terminal device locally performs a part of analysis on the acquired data, the result is sent to a server on the network side for further computational analysis. In addition, in the modes of b) and c), data interaction between the terminal equipment and the terminal equipment can be introduced to complete big data analysis or result sharing.
In the above scenario, the AI model has the following characteristics:
1) The AI models are of a very diverse nature and vary from third party vendor to third party vendor and under different conditions.
2) Since many AI services have the particularity of being divided by regions and time, the terminal device may need to use different AI models in different regions and times.
3) For services such as machine learning and distributed AI computation, the AI model needs to be updated frequently.
4) An AI model may require hundreds of megabytes or more of storage space, and a terminal device may not store all models locally, and therefore, it is necessary to be able to update the AI model accurately in real time.
As can be seen from the above description of the characteristics of the AI model, the AI model has many types, large quantity, and frequent update, and if an enterprise manages the AI model by using local machine room resources, the resource overhead is high.
Fig. 2 shows a block diagram of a communication system provided by an exemplary embodiment of the present application, which may include: a terminal device 12, a (radio) access Network ((R) AN) 14, a core Network 16, and a Data Network (DN) 18.
The terminal devices 12 may include various handheld devices, vehicle mounted devices, wearable devices, computing devices or other processing devices connected to wireless modems, as well as various forms of user equipment, mobile Stations (MSs), terminals (terminal devices), and the like, having wireless communication capabilities. For convenience of description, the above-mentioned devices are collectively referred to as terminal devices.
Access network 14 includes a number of network devices. The network device may be a base station, which is a means deployed in an access network to provide wireless communication functionality for the terminal. The base stations may include various forms of macro base stations, micro base stations, relay stations, access points, and the like. In systems using different radio access technologies, the names of devices with base station functionality may differ, for example in LTE systems, referred to as eNodeB or eNB; in a 5G NR-U system, it is called gNodeB or gNB. The description of "base station" may change as communication technology evolves. In this embodiment, the apparatus for providing a wireless communication function for a terminal device is referred to as a network device.
The core network 16 may include: user Plane Function (UPF) and control Plane Function. The control plane functions may include: access and Mobility Management Function (AMF), session Management Function (SMF), policy Control Function (PCF), and Unified Data Manager (UDM), application Function (AF), network Slice Selection Function (NSSF), authentication service Function (AUSF).
The wireless network may include an access network 14 and a core network 16, and different network elements in the wireless network may specifically correspond to base stations of the access network or various functional network elements of the core network (e.g., SMF, PCF, AMF, etc.).
The terminal device 12 and the network device in the access network 14 communicate with each other through some air interface technology, for example, a Uu interface. The core network 16 may communicate data with the external data network 18 via an N6 interface and with the access network 14 via an N3 interface.
The technical scheme of the embodiment of the application can be applied to various communication systems, for example: global System for Mobile communications (GSM) System, code Division Multiple Access (CDMA) System, wideband Code Division Multiple Access (WCDMA) System, general Packet Radio Service (GPRS), long Term Evolution (Long Term Evolution, LTE) System, LTE Frequency Division Duplex (FDD) System, LTE Time Division Duplex (TDD) System, advanced Long Term Evolution (Advanced Long Term Evolution), an LTE-a) System, a New Radio (NR) System, an Evolution System of the NR System, an LTE (LTE-based Access to Unlicensed spectrum) System on an Unlicensed Frequency band, an NR-U System, a Universal Mobile Telecommunications System (UMTS), a Worldwide Interoperability for Microwave Access (WiMAX) Communication System, a Wireless Local Area Network (WLAN), a Wireless Fidelity (WiFi), a next-generation Communication System, or other Communication systems.
Generally, the conventional Communication system supports a limited number of connections and is easy to implement, however, with the development of Communication technology, the mobile Communication system will support not only conventional Communication but also, for example, device-to-Device (D2D) Communication, machine-to-Machine (M2M) Communication, machine Type Communication (MTC), vehicle-to-Vehicle (V2V) Communication, and Vehicle networking (V2X) system, etc. The embodiments of the present application can also be applied to these communication systems.
FIG. 3 illustrates a flow chart of a model management method provided by an exemplary embodiment of the present application. The present embodiment is exemplarily described by applying the method to the communication system shown in fig. 2, where the method includes:
in step 310, the AI model management module and the AI model management network element execute a first interactive process.
The terminal equipment comprises a plurality of functional modules for realizing different functions. The AI model management module is a functional module used for terminal equipment end to manage AI models.
The terminal device is installed with several applications that need to perform corresponding AI services using the AI model. The AI model management module supports interaction with an AI model management network element so as to download or upload AI models for an application layer and/or a wireless protocol layer in the terminal equipment. The radio protocol layer is a protocol layer supporting processing of a 3GPP protocol, and the embodiment of the present application does not limit a specific implementation form of the radio protocol layer.
The AI model management network element is a network element used for AI model management at a data network end. Optionally, a specific implementation form of the AI model management network element is not limited in this embodiment of the application, and the AI model management network element may be an existing network element or a newly added network element and is used for AI model management.
The first interactive process is an interactive process between the AI model management module and the AI model management network element. The first interactive process is used for realizing downloading or uploading of the AI model.
Optionally, the AI model management module and the AI model management network element are both function modules or network elements corresponding to the same operator, and the AI model management module and the AI model management network element are part of the operator network.
Optionally, the AI model management module and the AI model management network element perform the first interaction process to realize downloading or uploading of the AI model, and also support compression and decompression of the AI model for transmission or storage of AI model content.
In summary, in the method provided in this embodiment, the task of managing the AI model is handed over to the AI model management module and the AI model management network element for execution, and since the AI model management module is a functional module used by the terminal device for performing AI model management and the AI model management network element is a network element used by the data network terminal for performing AI model management, it is not necessary to occupy machine room resources of the third party application local to perform AI model management, thereby ensuring resource usage of the enterprise.
In all embodiments of the present application, the model management method for the AI model may be used in combination with any one of the embodiments of the present application, or may be used alone, and the embodiments of the present application do not limit this. However, in order to make the technical solution easier to understand, the following embodiments of the present application are exemplified in combination with the foregoing embodiments.
In an alternative embodiment based on fig. 3, the AI model in the AI model management module and AI model management network element may include, but is not limited to: a first type AI model and a second type AI model.
Wherein the first type AI model corresponds to an operator and the second type AI model corresponds to a third party application (e.g., from an application or service provider). An operator is an operator who establishes and operates a network for the purpose of providing a land mobile communication service to the public. The third party application is an application developed for the terminal device. That is, the first type AI model is an operator's AI model and the second type AI model is a third party application AI model.
The AI model corresponds to an AI model identification. Optionally, the AI model identifications of the above two types of AI models correspond to the same or different model identification formats. The model identification format of the first type AI model is a first type model identification format, and the model identification format of the second type AI model is a second type model identification format.
Optionally, the first type model identification format and the second type model identification format are different, which may mean that the identification types included in the two types of model identification formats are different; or, the different identifier types have different positions in the format; or, the number of bits may be different for different identification types.
Illustratively, the first type of model identification format includes: operator identification and model identification.
The operator ID may be denoted as PLMN ID, and is used to distinguish different operators. The Model identification may be denoted as Model Id, which is used to identify the Model.
Illustratively, the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes at least one of: operator identification, application identification, and model identification.
In one implementation, the second type model identification format includes: operator identification, application identification and model identification; in another implementation, the second type model identification format includes: application identification and model identification.
The operator ID may be denoted as PLMN ID, and is used to distinguish different operators. The Application identification may be denoted as Application Id for identifying the third party Application. The Model identification may be denoted as Model Id, which is used to identify the Model.
For the operator identifier and the application identifier in the above two types of model identifier formats, the operator identifier may be determined by an operator or a third party application, and the application identifier may be determined by the operator or the third party application.
For the model identification in the above two types of model identification formats, the model identification may include: at least one of a model Type identifier (Type Id), a model Structure parameter identifier (Structure Id), and a model Weight parameter identifier (Weight Id).
Wherein, the model type identifier is used to identify the type of the AI model, and the type of the AI model includes but is not limited to: at least one of Deep Neural Network (DNN) model, convolutional Neural Network (CNN) model, and Recurrent Neural Network (RNN) model.
The model structure parameters identify the structure used to identify the model. Illustratively, for the DNN model, the internal neural network layers can be divided into three categories: an input layer, a hidden layer and an output layer. Typically the first layer is the input layer, the last layer is the output layer, and the number of layers in between are all hidden layers.
The model weight parameter identifies a weight value used to identify the model usage. For example, for the DNN model, each layer needs to use a specific algorithm, weight value, constant, etc. to calculate the sample parameter of the previous layer to obtain an output result, and the output result is used as an input of the next layer to calculate the output of the next layer, and finally obtain the result of the output layer (output layer).
For the model identifiers in the two types of model identifier formats, the values or partial values of the model identifiers may include: a normalized model identification value and a non-normalized model identification value.
The standardized model identification value is a value which is agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
By normalization, one or more values or partial values of the model identification can be agreed to serve as the values of the normalized model identification. Exemplarily, for a model identifier with a value of 000010, it means that the AI model is an AI model for processing a target service, i.e., a video service; for a model identification value of 000011, this means that the AI model is an AI model corresponding to the target type CNN; for a model identification value of 000012, this means that the AI model is the target AI model for the target traffic of autonomous driving.
Wherein the non-standardized model identity value is a model identity value determined by an operator or a third party application. The same non-standardized model value may be defined for different AI models, in different operators or different third party applications. Such as: 100001 is a non-standardized model identification value, in operator a, the AI model whose mode identification value is 100001 is the CNN model, and in operator B, the AI model whose mode identification value is 100001 is the RNN model. Or, the 100001-valued AI model of the operator a is used for an automatic driving service, and the 100001-valued AI model of the operator B is suitable for an Augmented Reality (AR)/Virtual Reality (VR) service.
Meanwhile, in an alternative embodiment based on fig. 3, there is also an interface for implementing AI model management. It is understood that the interaction is performed by the AI model identification described in the above embodiments.
1. A first interface.
The first interface is an interface between the AI model management module and the AI model management network element. The first interface may use an application layer Protocol, such as HyperText Transfer Protocol (HTTP).
Optionally, step 310 in the above embodiment may alternatively be implemented as: the AI model management module and the AI model management network element execute a first interactive process through a first interface.
Wherein, the first interactive process includes but is not limited to at least one of the following:
1) AI model request flow.
The AI model request flow is used for the AI model management module to request to download the target AI model from the AI model management network element or to upload the target AI model to the AI model management network element.
2) And the AI model informs the process.
The AI model notification process is used for the AI model management network element to notify the AI model management module of a target AI model to be uploaded or a target AI model to be sent to the AI model management network element.
3) And (5) capability negotiation flow.
The capability negotiation process is used for negotiating and supporting the target AI model to be downloaded or uploaded between the AI model management module and the AI model management network element.
4) And (5) an AI model downloading process.
The AI model downloading process is used for downloading the target AI model from the AI model management network element by the AI model management module.
5) And uploading the AI model.
And the AI model uploading process is used for realizing that the AI model management module uploads the target AI model to the AI model management network element.
Optionally, the first interface comprises a first control plane and a first user plane; the first control plane is used for executing an AI model request process, an AI model notification process and a capability negotiation process, and the first user plane is used for executing an AI model downloading process and an AI model uploading process.
That is, the processes that can be performed by the first interface may be separated, the control-related processes (such as the capability negotiation process) may be performed by the first control plane, and the bearer-related functions (such as the AI model download process) may be performed by the first user plane.
It is to be understood that, as shown in fig. 4, the AI model management module and the AI model management network element may be separated into: the control plane and the user plane separate the bearer and the control.
2. And a second interface.
The second interface is an interface between the AI model management module and a third party application in the terminal device.
Optionally, the AI model management module and the third party application execute the second interaction procedure through the second interface. Optionally, the second interaction flow includes: the AI model management module sends an AI model to a third party application, which sends the AI model to the AI model management module.
Illustratively, the AI model management module downloads an AI model from an AI model management network element, and if an Application identifier Application ID =1 in the AI model identifier of the AI model, the AI model management module sends the downloaded AI model to Application-1.
Illustratively, the second interface is used for the application to send a specific AI model to the AI model management module, so that the AI model management module saves or uploads the specific AI model to the AI model management network element for unified saving/management.
3. And a third interface.
The third interface is an interface between the AI model management network element and other servers/network elements in the data network.
Optionally, the AI model management network element and the application server execute a third interaction procedure through a third interface; or, the AI model management network element and a third party network element execute the fourth interaction process through the third interface, and the third party network element is another network element except the AI model management network element.
Wherein the application server may correspond to a third party network element.
Optionally, the AI model management network element may also interact with other network elements through the third interface, and the other network elements and the AI model management network element may correspond to the same operator or different operators.
Optionally, the third interaction flow includes: storing, downloading and uploading the AI model between the AI model management network element and the application server; the fourth interactive process comprises the following steps: and storing, downloading and uploading the AI model between the AI model management network element and the third-party network element.
Illustratively, the application server sends the AI model of the third-party application to an AI model management network element for safekeeping.
Illustratively, the third-party network element is an AI model selecting network element, and the AI model selecting network element is configured to determine an AI model that needs to be used by the terminal device, and trigger the AI model managing network element through the interface 3 to send the corresponding AI model to the terminal device for use.
4. And a fourth interface.
The fourth interface is an interface between the AI model management network element and the wireless network.
And the AI model management network element and the network element in the wireless network execute a fifth interactive process through the fourth interface. The network element in the wireless network may have the function of an AI model management module, and interactively execute a corresponding function with the AI model management network element.
Optionally, the purpose of the fifth interaction flow may be any one of the following:
1) The fifth interaction procedure is used to trigger the wireless network to establish a proprietary bearer or Quality of Service (QoS) data flow.
The security and the efficiency when the first interactive process is executed through the first interface are guaranteed by triggering the wireless network to establish the special bearer or the QoS data stream.
2) And the fifth interactive process is used for the AI model management network element to send the AI model to the wireless network.
The wireless network may use the AI model obtained from the AI model management network element.
3) And the fifth interactive process is used for the AI model management network element to obtain the AI model from the wireless network and/or store the AI model.
The AI model management network element may also obtain and store an AI model from the wireless network.
4) And the fifth interactive process is used for the AI model management network element to receive a request from the network element in the wireless network, and the request is used for requesting the AI model management network element to send the AI model.
The AI model management network element may correspondingly send an AI model to the terminal device according to a request of a network element in the wireless network.
Optionally, in addition to the above functions, the network element in the wireless network may also perform capability negotiation with the AI model management network element, and negotiate an AI model that can be downloaded or uploaded.
Optionally, the AI model management network element interfaces with a capability opening network element in the wireless network or directly interfaces with a certain network element in the wireless network. Wherein, the capability opening network element is a network element providing a capability opening service. That is, the AI model management network element may interact with the capability openness network element in the wireless network through the fourth interface. Capability-opening network elements include, but are not limited to: at least one of a Network Exposure Function (NEF) Network element, a Service Capability Exposure Function (SCEF) Network element, and a Policy Control Function (PCF) Network element. The NEF network element and the SCEF network element can open the service capability to a third party service provider through the API interface, and the PCF network element is responsible for strategy control.
In summary, in the method provided in this embodiment, the first type AI model and the second type AI model are defined, and the model identification formats of the different types of AI models are provided, so that the AI models are identified more accurately, and the management of the AI models is facilitated.
According to the method provided by the embodiment, by defining different interfaces, an operator can open the relevant functions of the AI management to other objects (such as third-party applications), so that the transmission of the content of the AI model is facilitated.
It should be noted that, the method embodiments described above may be implemented individually or in combination, and the present application is not limited to this.
In the above embodiments, the steps performed by the terminal device may be implemented solely as the model management method on the terminal device side, and the steps performed by the network device may be implemented solely as the model management method on the network device side.
Corresponding to the above method embodiment, the present application provides a model management system, including: an artificial intelligence AI model management module and an AI model management network element;
the AI model management module is used for managing the network element with the AI model and executing a first interactive process;
the AI model management module is a functional module used for carrying out AI model management at a terminal equipment end, the AI model management network element is a network element used for carrying out AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload an AI model for an application layer and/or a wireless protocol layer in the terminal equipment.
Optionally, the model management system further comprises: a first interface; the AI model management module is used for executing a first interaction process with the AI model management network element through a first interface; wherein, the first interactive process comprises at least one of the following steps:
an AI model request flow, which is used for the AI model management module to request to download the target AI model from the AI model management network element or upload the target AI model to the AI model management network element;
an AI model notification process, wherein the AI model notification process is used for informing an AI model management network element of a target AI model to be uploaded by an AI model management module or a target AI model to be sent to the AI model management network element;
the capability negotiation process is used for negotiating a target AI model supporting downloading or uploading between the AI model management module and the AI model management network element;
an AI model downloading process;
and uploading the AI model.
Optionally, the first interface comprises a first control plane and a first user plane; the first control plane is used for executing an AI model request process, an AI model notification process and a capability negotiation process, and the first user plane is used for executing an AI model downloading process and an AI model uploading process.
Optionally, the AI model management system further comprises: a second interface; the AI model management module is used for executing a second interaction process with a third party application through a second interface.
Optionally, the AI model management system further comprises: a third interface; the AI model management network element is used for executing a third interactive process with the application server through a third interface; or, the AI model managing network element is configured to execute a fourth interaction procedure with a third party network element through a third interface, where the third party network element is another network element except the AI model managing network element.
Optionally, the AI model management system further comprises: a fourth interface; the AI model management network element is used for opening the network element with the capability in the wireless network and executing a fifth interaction process through a fourth interface; the fifth interaction flow is used for triggering the wireless network to establish a special bearer or QoS data flow;
or, the fifth interactive process is used for the AI model management network element to send an AI model to the network element in the wireless network;
or, the fifth interactive process is used for the AI model management network element to obtain an AI model from the wireless network and/or store the AI model;
or, the fifth interaction flow is used for the AI model management network element to receive a request from a network element in the wireless network, and the request is used for requesting the AI model management network element to send an AI model.
Optionally, the network element in the wireless network includes: the network element with the capability openness comprises: at least one of a NEF network element, a SCEF network element, and a PCF network element.
Optionally, the model identification format of the first type AI model is a first type model identification format, and the first type model identification format includes: operator identification and model identification;
the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes: operator identification, model identification, and application identification.
Optionally, the value of the model identifier includes: the standardized model identification value is a value which is well agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
Optionally, the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
Optionally, the operator identity is determined by the operator or a third party application; the application identity is determined by the operator or a third party application.
Referring collectively to fig. 5, there is shown a block diagram of an AI model management system provided in an embodiment of the present application.
As shown in fig. 5 (a), the AI model management system includes: an AI model management module 501, and an AI model management network element 502.
An AI model management module 501 is introduced at the terminal device side, an AI model management network element 502 is introduced at the network side, and the AI model management module 501 and the AI model management network element 502 may be part of an operator network, and provide an AI model management service for a third party Application through an Application Program Interface (API).
The wireless network (including the base station and the core network) of the operator is between the terminal device and the Data network, the AI model management module 501 and the AI model management network element 502 may communicate through an interface 1 (i.e., a first interface) in a user plane of the wireless network, and Data transmitted by the interface 1 may be transmitted through a Protocol Data Unit (PDU) session of the wireless network.
Optionally, the AI model management system further includes an interface 2 (i.e., a second interface) and an interface 3 (i.e., a third interface). The AI model management module 501 may open relevant functions of AI management (including AI model storage, downloading, uploading, etc.) to third party applications in the terminal device through the interface 2. The AI model management network element 502 can open relevant functions of AI management (including AI model storage, downloading, uploading, etc.) to third party applications or other network elements through the interface 3.
Optionally, the AI model management system further comprises an interface 4 (i.e. a fourth interface). The AI model management network element 502 can interact with the capability openness network element of the operator network through the interface 4. Capability-opening network elements include, but are not limited to: NEF network element, SCEF network element and PCF network element.
Alternatively, as shown in (b) of fig. 5, the AI model management module 501 interfaces with a wireless protocol layer 503, and the wireless protocol layer 503 may process a 3GPP protocol. Illustratively, the AI model management module 501 sends the AI model used by the operator to the radio protocol layer 503 for use. Such as for radio channel quality optimization, mobility management optimization, session management optimization, UE policy optimization, etc. for a wireless network.
In summary, the embodiment of the present application provides a model management system, which can implement an AI model management service by using cloud computing resources of an operator, and a third party application provider does not need to manage an AI model by using local machine room resources.
Fig. 6 is a block diagram of a model management apparatus according to an exemplary embodiment of the present application. The apparatus is applied in a terminal device, or the apparatus is implemented as a terminal device or a part of a terminal device. The device comprises: an AI model management module 601;
an AI model management module 601, configured to execute a first interaction procedure with an AI model management network element;
the AI model management module 601 is a functional module used by a terminal device end to perform AI model management, the AI model management network element is a network element used by a data network end to perform AI model management, and the AI model management module 601 supports interaction with the AI model management network element so as to download or upload an AI model for an application layer and/or a wireless protocol layer in the terminal device.
In an optional embodiment, the AI model management module 601 is configured to execute a first interaction procedure with an AI model management network element through a first interface; wherein, the first interactive process comprises at least one of the following steps:
an AI model request process for the AI model management module 601 to request downloading of a target AI model from or uploading of the target AI model to an AI model management network element;
an AI model notification process, where the AI model notification process is used for an AI model management network element to notify an AI model management module 601 of a target AI model to be uploaded or a target AI model to be sent to the AI model management network element;
an ability negotiation process, which is used for negotiating a target AI model supporting downloading or uploading between the AI model management module 601 and the AI model management network element;
an AI model downloading process;
and uploading the AI model.
In an alternative embodiment, the first interface comprises a first control plane and a first user plane; the first control plane is used for executing an AI model request process, an AI model notification process and a capability negotiation process, and the first user plane is used for executing an AI model downloading process and an AI model uploading process.
In an alternative embodiment, the AI model management module 601 is configured to perform a second interactive process with a third party application through a second interface.
In an alternative embodiment, the AI model in the AI model management module 601 includes at least one of: a first type AI model and a second type AI model; wherein the first type of AI model corresponds to an operator and the second type of AI model corresponds to a third party application.
In an optional embodiment, the model identification format of the first type AI model is a first type model identification format, and the first type model identification format includes: operator identification and model identification; the model identification format of the second type AI model is a second type model identification format, and the second type model identification format comprises at least one of the following formats: operator identification, application identification, and model identification.
In an optional embodiment, the values of the model identifier include: the standardized model identification value is a value well agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
In an alternative embodiment, the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
In an alternative embodiment, the operator identity is determined by the operator or a third party application; the application identity is determined by the operator or a third party application.
Fig. 7 is a block diagram of a model management apparatus according to an exemplary embodiment of the present application. The apparatus is applied in a network device, or is implemented as a network device or a part of a network device. The device comprises: AI model management network element module 701;
an AI model management network element module 701, configured to execute a first interaction procedure with the AI model management module;
the AI model management module is a functional module used by a terminal device end for performing AI model management, the AI model management network element module 701 is a network element module used by a data network end for performing AI model management, and the AI model management module supports interaction with the AI model management network element module 701 so as to download or upload an AI model for an application layer and/or a wireless protocol layer in the terminal device.
In an optional embodiment, the AI model management network element module 701 is configured to execute a first interaction procedure with the AI model management module through a first interface; wherein, the first interactive process comprises at least one of the following steps:
an AI model request process, which is used for the AI model management module to request downloading of a target AI model from the AI model management network element module 701 or uploading of the target AI model to the AI model management network element;
an AI model notification process, where the AI model notification process is used for the AI model management network element module 701 to notify the AI model management module of a target AI model that needs to be uploaded or a target AI model that needs to be sent to the AI model management network element;
a capability negotiation process, which is used for negotiating a target AI model supporting downloading or uploading between the AI model management module and the AI model management network element module 701;
an AI model downloading process;
and uploading the AI model.
In an alternative embodiment, the first interface comprises a first control plane and a first user plane; the first control plane is used for executing an AI model request process, an AI model notification process and a capability negotiation process, and the first user plane is used for executing an AI model downloading process and an AI model uploading process.
In an optional embodiment, the AI model managing network element module 701 is configured to execute a third interactive process with the application server through a third interface; or, the AI model managing network element module 701 is configured to execute a fourth interaction procedure with a third party network element through a third interface, where the third party network element is another network element except the AI model managing network element.
In an optional embodiment, the AI model management network element module 701 is configured to execute a fifth interaction procedure with a network element in the wireless network through a fourth interface; the fifth interactive process is used for triggering the wireless network to establish a special bearer or a QoS data flow; or, the fifth interaction flow is used for the AI model management network element module 701 to send an AI model to a network element in the wireless network; or, the fifth interaction procedure is used for the AI model management network element module 701 to obtain an AI model from the wireless network and/or store the AI model; or, the fifth interaction flow is used for the AI model management network element module 701 to receive a request from a network element in the wireless network, where the request is used to request the AI model management network element module 701 to send an AI model.
In an optional embodiment, a network element in a wireless network includes: the network element with the capability being opened comprises: at least one of a network open function NEF network element, a service capability open function SCEF network element and a policy control function PCF network element.
In an optional embodiment, the AI model in the AI model management network element module 701 includes at least one of: a first type AI model and a second type AI model; wherein the first type AI model corresponds to an operator and the second type AI model corresponds to a third party application.
In an optional embodiment, the model identification format of the first type AI model is a first type model identification format, and the first type model identification format includes: operator identification and model identification;
the model identification format of the second type AI model is a second type model identification format, and the second type model identification format comprises at least one of the following formats: operator identification, model identification, and application identification.
In an optional embodiment, the values of the model identifier include: the standardized model identification value is a model identification value which is agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
In an alternative embodiment, the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
In an alternative embodiment, the operator identity is determined by the operator or a third party application; the application identity is determined by the operator or a third party application.
Fig. 8 is a schematic structural diagram of a communication device according to an exemplary embodiment of the present application, where the communication device includes: a processor 101, a receiver 102, a transmitter 103, a memory 104, and a bus 105.
The processor 101 includes one or more processing cores, and the processor 101 executes various functional applications and information processing by running software programs and modules.
The receiver 102 and the transmitter 103 may be implemented as one communication component, which may be one communication chip.
The memory 104 is connected to the processor 101 by a bus 105.
The memory 104 may be used to store at least one instruction that the processor 101 is configured to execute to implement the various steps in the above-described method embodiments.
Further, the memory 104 may be implemented by any type or combination of volatile or non-volatile storage devices, including but not limited to: magnetic or optical disk, electrically Erasable Programmable Read-Only Memory (EEPROM), erasable Programmable Read-Only Memory (EPROM), static Random Access Memory (SRAM), read-Only Memory (ROM), magnetic Memory, flash Memory, programmable Read-Only Memory (PROM).
In an exemplary embodiment, a computer-readable storage medium is further provided, and at least one instruction, at least one program, a code set, or a set of instructions is stored in the computer-readable storage medium, and the at least one instruction, the at least one program, the code set, or the set of instructions is loaded and executed by a processor to implement the model management method executed by a terminal device or the model management method executed by a data network provided in the above-described method embodiments.
In an exemplary embodiment, there is also provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium, the computer instructions being read by a processor of a computer device from the computer readable storage medium, the processor executing the computer instructions to cause the computer device to perform the model management method of the above aspect.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is intended only to illustrate the alternative embodiments of the present application, and should not be construed as limiting the present application, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims (55)

  1. A model management method is applied to a terminal device, and the terminal device comprises: an Artificial Intelligence (AI) model management module, the method comprising:
    the AI model management module and the AI model management network element execute a first interactive process;
    the AI model management module is a functional module used for AI model management at a terminal equipment end, the AI model management network element is used for AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload AI models for an application layer and/or a wireless protocol layer in the terminal equipment.
  2. The method of claim 1, wherein the AI model management module and the AI model management network element perform a first interactive process comprising:
    the AI model management module and the AI model management network element execute the first interaction process through a first interface;
    wherein the first interaction flow comprises at least one of:
    an AI model request process for the AI model management module to request downloading of a target AI model from the AI model management network element or uploading of the target AI model to the AI model management network element;
    an AI model notification process, configured to notify, by the AI model management network element, the AI model management module of the target AI model that needs to be uploaded or the target AI model that needs to be sent to the AI model management network element;
    an ability negotiation process, configured to allow the AI model management module and the AI model management network element to negotiate the target AI model supporting downloading or uploading;
    an AI model downloading process;
    and uploading the AI model.
  3. The method of claim 2, wherein the first interface comprises a first control plane and a first user plane;
    the first control plane is configured to execute the AI model request process, the AI model notification process, and the capability negotiation process, and the first user plane is configured to execute the AI model download process and the AI model upload process.
  4. The method of claim 1, further comprising:
    and the AI model management module and the third-party application execute a second interaction process through a second interface.
  5. The method according to any one of claims 1 to 4,
    the AI model in the AI model management module comprises at least one of the following: a first type AI model and a second type AI model;
    wherein the first type AI model corresponds to an operator and the second type AI model corresponds to a third party application.
  6. The method of claim 5,
    the model identification format of the first type AI model is a first type model identification format, and the first type AI model identification format includes: operator identification and model identification;
    the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes at least one of the following: the operator identity, the application identity and the model identity.
  7. The method of claim 6,
    the values of the model identification include: the standardized model identification value is a value which is well agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
  8. The method of claim 6,
    the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
  9. The method of claim 6,
    the operator identification is determined by the operator or the third party application;
    the application identification is determined by the operator or the third party application.
  10. A model management method, applied to a data network, the data network comprising: an Artificial Intelligence (AI) model management network element, the method comprising:
    the AI model management network element and the AI model management module execute a first interaction process;
    the AI model management module is a functional module used for carrying out AI model management at a terminal equipment end, the AI model management network element is a network element used for carrying out AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload an AI model for an application layer and/or a wireless protocol layer in the terminal equipment.
  11. The method of claim 10, wherein the AI model management network element and AI model management module perform a first interactive process comprising:
    the AI model management network element and the AI model management module execute a first interaction process through a first interface;
    wherein the first interaction flow comprises at least one of:
    an AI model request process for the AI model management module to request downloading of a target AI model from the AI model management network element or uploading of the target AI model to the AI model management network element;
    an AI model notification process, where the AI model notification process is used for the AI model management network element to notify the AI model management module of the target AI model that needs to be uploaded or the target AI model that needs to be sent to the AI model management network element;
    an ability negotiation process, configured to negotiate, between the AI model management module and the AI model management network element, the target AI model that supports downloading or uploading;
    an AI model downloading process;
    and uploading the AI model.
  12. The method of claim 11, wherein the first interface comprises a first control plane and a first user plane;
    the first control plane is configured to execute the AI model request procedure, the AI model notification procedure, and the capability negotiation procedure, and the first user plane is configured to execute the AI model download procedure and the AI model upload procedure.
  13. The method of claim 10, further comprising:
    the AI model management network element and the application server execute a third interactive process through a third interface;
    or the like, or a combination thereof,
    and the AI model management network element and a third-party network element execute a fourth interaction process through the third interface, wherein the third-party network element is other than the AI model management network element.
  14. The method of claim 10, further comprising:
    the AI model management network element and a network element in the wireless network execute a fifth interaction process through a fourth interface;
    wherein, the fifth interactive process is used for triggering the wireless network to establish a proprietary bearer or a quality of service (QoS) data flow;
    or, the fifth interaction flow is used for the AI model management network element to send an AI model to a network element in the wireless network;
    or, the fifth interaction procedure is used for the AI model management network element to obtain an AI model from the wireless network and/or store the AI model;
    or, the fifth interaction procedure is used for the AI model management network element to receive a request from a network element in the wireless network, where the request is used to request the AI model management network element to send an AI model.
  15. The method of claim 14,
    the network element in the wireless network comprises: a capability openness network element, the capability openness network element comprising: at least one of a network open function NEF network element, a service capability open function SCEF network element and a policy control function PCF network element.
  16. The method according to any one of claims 10 to 15,
    the AI model in the AI model management network element comprises at least one of the following: a first type AI model and a second type AI model;
    wherein the first type AI model corresponds to the operator and the second type AI model corresponds to a third party application.
  17. The method of claim 16,
    the model identification format of the first type AI model is a first type model identification format, and the first type model identification format includes: operator identification and model identification;
    the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes at least one of the following: the operator identification, the model identification, and the application identification.
  18. The method of claim 17,
    the values of the model identification include: the standardized model identification value is a model identification value which is well agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
  19. The method of claim 17,
    the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
  20. The method of claim 17,
    the operator identification is determined by the operator or the third party application;
    the application identification is determined by the operator or the third party application.
  21. A model management system, characterized in that the model management system comprises: an artificial intelligence AI model management module and an AI model management network element;
    the AI model management module is used for executing a first interactive process with the AI model management network element;
    the AI model management module is a functional module used for carrying out AI model management at a terminal equipment end, the AI model management network element is a network element used for carrying out AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload an AI model for an application layer and/or a wireless protocol layer in the terminal equipment.
  22. The system of claim 21, wherein the model management system further comprises: a first interface;
    the AI model management module is configured to execute a first interaction procedure with the AI model management network element, where the first interaction procedure includes:
    the AI model management module is configured to execute the first interaction procedure with the AI model management network element through the first interface;
    wherein the first interaction flow comprises at least one of:
    an AI model request process for the AI model management module to request downloading of a target AI model from the AI model management network element or uploading of the target AI model to the AI model management network element;
    an AI model notification process, where the AI model notification process is used for the AI model management network element to notify the AI model management module of the target AI model that needs to be uploaded or the target AI model that needs to be sent to the AI model management network element;
    an ability negotiation process, configured to allow the AI model management module and the AI model management network element to negotiate the target AI model supporting downloading or uploading;
    an AI model downloading process;
    and uploading the AI model.
  23. The system of claim 22, the first interface comprising a first control plane and a first user plane;
    the first control plane is configured to execute the AI model request process, the AI model notification process, and the capability negotiation process, and the first user plane is configured to execute the AI model download process and the AI model upload process.
  24. The system of claim 21, wherein the AI model management system further comprises: a second interface;
    and the AI model management module is used for executing a second interaction process with a third-party application through the second interface.
  25. The system of claim 21, wherein the AI model management system further comprises: a third interface;
    the AI model management network element is configured to execute a third interaction procedure with the application server through the third interface;
    or the like, or, alternatively,
    the AI model management network element is configured to execute a fourth interaction procedure with a third-party network element through the third interface, where the third-party network element is another network element except the AI model management network element.
  26. The system of claim 21, wherein the AI model management system further comprises: a fourth interface;
    the AI model management network element is configured to execute a fifth interaction procedure with a network element in a wireless network through the fourth interface;
    the fifth interaction flow is used for triggering the wireless network to establish a special bearer or a quality of service (QoS) data flow;
    or, the fifth interaction flow is used for the AI model management network element to send an AI model to a network element in the wireless network;
    or, the fifth interaction flow is used for the AI model management network element to obtain an AI model from the wireless network and/or store the AI model;
    or, the fifth interaction flow is used for the AI model management network element to receive a request from a network element in the wireless network, where the request is used to request the AI model management network element to send an AI model.
  27. The system of claim 26,
    the network element in the wireless network comprises: a capability openness network element, the capability openness network element comprising: at least one of a network open function NEF network element, a service capability open function SCEF network element and a policy control function PCF network element.
  28. The system of any one of claims 21 to 27,
    the AI model in the AI model management system includes at least one of: a first type AI model and a second type AI model;
    wherein the first type AI model corresponds to the operator and the second type AI model corresponds to a third party application.
  29. The system of claim 28,
    the model identification format of the first type AI model is a first type model identification format, and the first type model identification format includes: operator identification and model identification;
    the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes: the operator identification, the model identification and the application identification.
  30. The system of claim 29,
    the values of the model identification include: the standardized model identification value is a value which is well agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
  31. The system of claim 29,
    the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
  32. The system of claim 29,
    the operator identification is determined by the operator or the third party application;
    the application identification is determined by the operator or the third party application.
  33. A model management device is applied to a terminal device, and the device comprises: an artificial intelligence AI model management module;
    the AI model management module is used for executing a first interactive process with the AI model management network element;
    the AI model management module is a functional module used for carrying out AI model management at a terminal equipment end, the AI model management network element is a network element used for carrying out AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element so as to download or upload an AI model for an application layer and/or a wireless protocol layer in the terminal equipment.
  34. The apparatus of claim 33,
    the AI model management module is configured to execute the first interaction process with the AI model management network element through a first interface;
    wherein the first interaction flow comprises at least one of:
    an AI model request process for the AI model management module to request downloading of a target AI model from the AI model management network element or uploading of the target AI model to the AI model management network element;
    an AI model notification process, where the AI model notification process is used for the AI model management network element to notify the AI model management module of the target AI model that needs to be uploaded or the target AI model that needs to be sent to the AI model management network element;
    an ability negotiation process, configured to negotiate, between the AI model management module and the AI model management network element, the target AI model that supports downloading or uploading;
    an AI model downloading process;
    and uploading the AI model.
  35. The apparatus of claim 34, wherein the first interface comprises a first control plane and a first user plane;
    the first control plane is configured to execute the AI model request process, the AI model notification process, and the capability negotiation process, and the first user plane is configured to execute the AI model download process and the AI model upload process.
  36. The apparatus of claim 33,
    and the AI model management module is used for executing a second interaction process with a third-party application through a second interface.
  37. The apparatus of any one of claims 33 to 36,
    the AI model in the AI model management module comprises at least one of the following: a first type AI model and a second type AI model;
    wherein the first type AI model corresponds to an operator and the second type AI model corresponds to a third party application.
  38. The apparatus of claim 37,
    the model identification format of the first type AI model is a first type model identification format, and the first type AI model identification format includes: operator identification and model identification;
    the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes at least one of the following: the operator identification, the application identification, and the model identification.
  39. The apparatus of claim 38,
    the values of the model identification include: the standardized model identification value is a value well agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
  40. The apparatus of claim 38,
    the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
  41. The apparatus of claim 38,
    the operator identification is determined by the operator or the third party application;
    the application identification is determined by the operator or the third party application.
  42. A model management apparatus for use in a data network, the apparatus comprising: artificial intelligence AI model management network element module;
    the AI model management network element module is used for executing a first interactive process with the AI model management module;
    the AI model management module is a functional module used for AI model management at a terminal equipment end, the AI model management network element module is a network element module used for AI model management at a data network end, and the AI model management module supports interaction with the AI model management network element module so as to download or upload AI models for an application layer and/or a radio protocol layer in the terminal equipment.
  43. The apparatus of claim 42,
    the AI model management network element module is used for executing a first interaction process with the AI model management module through a first interface;
    wherein the first interaction flow comprises at least one of:
    an AI model request process for the AI model management module to request downloading of a target AI model from the AI model management network element module or uploading of the target AI model to the AI model management network element;
    an AI model notification process, configured to provide the AI model management network element module with notification of the target AI model that the AI model management module needs to upload or the target AI model that needs to be sent to the AI model management network element;
    an ability negotiation process, configured to allow the AI model management module and the AI model management network element module to negotiate the target AI model supporting downloading or uploading;
    an AI model downloading process;
    and uploading the AI model.
  44. The apparatus of claim 43, wherein the first interface comprises a first control plane and a first user plane;
    the first control plane is configured to execute the AI model request procedure, the AI model notification procedure, and the capability negotiation procedure, and the first user plane is configured to execute the AI model download procedure and the AI model upload procedure.
  45. The apparatus of claim 42,
    the AI model management network element module is used for executing a third interactive process with the application server through a third interface;
    or the like, or, alternatively,
    and the AI model management network element module is configured to execute a fourth interaction procedure with a third-party network element through the third interface, where the third-party network element is another network element except the AI model management network element.
  46. The apparatus of claim 42,
    the AI model management network element module is configured to execute a fifth interaction procedure with a network element in the wireless network through a fourth interface;
    the fifth interaction flow is used for triggering the wireless network to establish a special bearer or a quality of service (QoS) data flow;
    or, the fifth interaction flow is used for the AI model management network element module to send an AI model to a network element in the wireless network;
    or, the fifth interaction procedure is used for the AI model management network element module to obtain an AI model from the wireless network and/or store the AI model;
    or, the fifth interaction flow is used for the AI model management network element module to receive a request from a network element in the wireless network, where the request is used to request the AI model management network element module to send an AI model.
  47. The apparatus of claim 46,
    the network element in the wireless network comprises: a capability openness network element, the capability openness network element comprising: at least one of a network open function NEF network element, a service capability open function SCEF network element and a policy control function PCF network element.
  48. The apparatus of any one of claims 42 to 47,
    the AI model in the AI model management network element module comprises at least one of the following: a first type AI model and a second type AI model;
    wherein the first type AI model corresponds to the operator and the second type AI model corresponds to a third party application.
  49. The apparatus of claim 48,
    the model identification format of the first type AI model is a first type model identification format, and the first type model identification format includes: operator identification and model identification;
    the model identification format of the second type AI model is a second type model identification format, and the second type model identification format includes at least one of the following: the operator identification, the model identification and the application identification.
  50. The apparatus of claim 49,
    the values of the model identification include: the standardized model identification value is a model identification value which is agreed by standardization and corresponds to an AI model of a target type or an AI model of a target service or a target AI model.
  51. The apparatus of claim 49,
    the model identification comprises: at least one of a model type identification, a model structure parameter identification, and a model weight parameter identification.
  52. The apparatus of claim 49,
    the operator identification is determined by the operator or the third party application;
    the application identification is determined by the operator or the third party application.
  53. A terminal device, characterized in that the terminal device comprises:
    a processor;
    a transceiver coupled to the processor;
    a memory for storing executable instructions of the processor;
    wherein the processor is configured to load and execute the executable instructions to implement the model management method of any of claims 1 to 9.
  54. A data network, characterized in that the data network comprises:
    a processor;
    a transceiver coupled to the processor;
    a memory for storing executable instructions of the processor;
    wherein the processor is configured to load and execute the executable instructions to implement the model management method of any of claims 10 to 20.
  55. A computer-readable storage medium having stored thereon executable instructions that are loaded and executed by a processor to implement the model management method of any of claims 1 to 20.
CN202080101791.0A 2020-07-31 2020-07-31 Model management method, system, device, communication equipment and storage medium Pending CN115699842A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/106415 WO2022021421A1 (en) 2020-07-31 2020-07-31 Model management method, system and apparatus, communication device, and storage medium

Publications (1)

Publication Number Publication Date
CN115699842A true CN115699842A (en) 2023-02-03

Family

ID=80037376

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202080101791.0A Pending CN115699842A (en) 2020-07-31 2020-07-31 Model management method, system, device, communication equipment and storage medium

Country Status (2)

Country Link
CN (1) CN115699842A (en)
WO (1) WO2022021421A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024103544A1 (en) * 2023-02-10 2024-05-23 Zte Corporation Transfer of artificial intelligence network management models in wireless communication systems

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023039905A1 (en) * 2021-09-18 2023-03-23 Oppo广东移动通信有限公司 Ai data transmission method and apparatus, device, and storage medium
WO2024082261A1 (en) * 2022-10-21 2024-04-25 北京小米移动软件有限公司 Model management method and apparatus, and device and medium
WO2024092852A1 (en) * 2022-11-06 2024-05-10 北京小米移动软件有限公司 Communication method and apparatus, and storage medium

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190311287A1 (en) * 2016-12-30 2019-10-10 Google Llc Assessing accuracy of a machine learning model
US10743147B2 (en) * 2017-10-30 2020-08-11 Apple Inc. Expanded implementation of enhanced broadcast multicast services for broadcast multicast content selection and service
US20190286989A1 (en) * 2018-03-15 2019-09-19 Polarr, Inc. Distributed neural network model utilization system
CN109698822A (en) * 2018-11-28 2019-04-30 众安信息技术服务有限公司 Combination learning method and system based on publicly-owned block chain and encryption neural network
CN111435926B (en) * 2019-01-11 2023-05-05 中国移动通信有限公司研究院 MIMO system channel prediction method, device, medium and equipment
CN110866588B (en) * 2019-11-08 2023-05-02 中国科学院软件研究所 Training learning method and system for realizing individuation of learning ability model of intelligent virtual digital animal

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024103544A1 (en) * 2023-02-10 2024-05-23 Zte Corporation Transfer of artificial intelligence network management models in wireless communication systems

Also Published As

Publication number Publication date
WO2022021421A1 (en) 2022-02-03

Similar Documents

Publication Publication Date Title
US11089515B2 (en) Adaptable radio access network
US20230209390A1 (en) Intelligent Radio Access Network
EP4099635A1 (en) Method and device for selecting service in wireless communication system
CN115699842A (en) Model management method, system, device, communication equipment and storage medium
US11223976B2 (en) Multiple-slice application delivery based on network slice associations
CN115119331A (en) Reinforcement learning for multi-access traffic management
EP3855843A1 (en) Method and apparatus for transmitting packet messages based on priority in a wireless communication system
WO2020020775A1 (en) Transmission of network slicing constraints in wireless networks
CN114787793A (en) Management method of network model and method and device for establishing or modifying session
WO2022206513A1 (en) Model processing method, communication device, and system
CN114258151A (en) Calculation data transmission method and device
Nardini et al. Simulating cellular communications in vehicular networks: making SimuLTE interoperable with Veins
CN116097904A (en) Network edge computing method and communication device
US11778031B2 (en) Service model management request
CN117716674A (en) Network resource model-based solution for AI-ML model training
CN113473544B (en) Network slice configuration
CN115190452B (en) Method, device, terminal and medium for managing side uplink bearer
KR20230065806A (en) Method and apparatus for providing split computing in wireless communications systems
CN114143832B (en) Service processing method, device and storage medium
WO2023246267A1 (en) Communication method, communication device, and system
WO2022247739A1 (en) Data transmission method and related device
US20230247688A1 (en) Method in a Telecommunications System, a Telecommunications System, a User Equipment and a Base Station and Methods therein
EP4369253A1 (en) Dataset sharing transmission instructions for separated two-sided ai/ml based model training
US20240114332A1 (en) Method and apparatus for managing user consent for roaming ue in wireless communication system
US20240056888A1 (en) Quality of Service Management for Protocol Data Unit Sets

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination