Model processing method, model execution method, and model registration method
By sending model files and receiving processing results and converted model files in the communication network, the problem of model incompatibility between different UE manufacturers is solved, and the correct execution and unified management of models on terminal devices are realized.
Patent Information
- Application Number
- PCT/CN2024/090734
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-06
- Filing Date
- 2024-04-30
- Publication Date
- 2024-11-14
AI Technical Summary
Different UE manufacturers have their own optimization methods for communication models, which makes it impossible for terminals to use the AI/ML models trained by network-side equipment normally.
The first device sends the model file to the second device and receives the processing result information and the converted model file, ensuring that the model is correctly executed in the second device and associated terminals.
This avoids the problem of the terminal being unable to use the model properly, and achieves unified management and correct execution of the model across different devices and terminals.
Smart Images

Figure CN2024090734_14112024_PF_FP_ABST
Abstract
Description
Model processing, execution, and registration methods
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims priority to Chinese patent application No. 202310506302.5, filed on May 6, 2023, entitled “Model Processing, Execution and Registration Method,” the entire contents of which are incorporated herein by reference. Technical Field
[0003] The present application belongs to the field of communication technology, and specifically relates to a model processing, execution and registration method. Background Art
[0004] In communication networks, artificial intelligence (AI) methods have been introduced to improve network communication performance. Network-side devices or terminals can first train AI or machine learning (ML) models based on training data to obtain a model suitable for a specific AI task. The trained AI / ML model is then sent to other terminal devices, allowing other terminal devices to infer the input data of a certain AI task based on the AI / ML model, thereby obtaining the inference result data corresponding to the specific AI task. Taking the terminal as a user equipment (UE) as an example, the UE predicts the signal strength based on the AI / ML model and sends the predicted signal strength to the base station device, so that the base station device can more accurately select the switching timing.
[0005] In related technologies, different UE vendors have their own optimization methods for communication models. When network-side equipment or a UE sends a trained AI / ML model to other UEs, the other UEs may not be able to use the model properly. Therefore, how to prevent terminals from being unable to use the model properly is a technical problem that needs to be solved urgently.
[0006] Summary of the Invention
[0007] The embodiment of the present application provides a model processing, execution and registration method, which can solve the problem of how to avoid the terminal being unable to use the model normally.
[0008] In a first aspect, a model processing method is provided, which is executed by a first device, and the method includes:
[0009] The first device sends a first message to at least one second device, where the first message includes a first model file corresponding to the first model;
[0010] The first device receives a second message returned by the second device;
[0011] The second message includes at least one of the following:
[0012] Processing result information, used to indicate the result obtained by processing the first model file;
[0013] A second model file is obtained by converting the first model file.
[0014] In a second aspect, a model processing method is provided, which is executed by a second device, and the method includes:
[0015] The second device receives a first message sent by the first device, where the first message includes a first model file corresponding to the first model;
[0016] The second device returns a second message to the first device;
[0017] The second message includes at least one of the following:
[0018] Processing result information, used to indicate the result obtained by processing the first model file;
[0019] A second model file is obtained by converting the first model file.
[0020] In a third aspect, a model execution method is provided, which is executed by a first terminal, and the method includes:
[0021] The first terminal obtains a fifth message, where the fifth message includes a first model file or a second model file corresponding to the first model, where the second model file is obtained by converting the first model file;
[0022] The first terminal executes the first model file or the second model file.
[0023] In a fourth aspect, a model registration method is provided, which is performed by a third device, and the method includes:
[0024] The third device sends a third message to the first device, where the third message includes a first model file corresponding to the first model, and the third message is used to register the first model;
[0025] The third device receives a fourth message returned by the first device;
[0026] The fourth message includes at least one of the following:
[0027] Processing result information, used to indicate the result obtained by processing the first model file;
[0028] A second model file is obtained by converting the first model file.
[0029] In a fifth aspect, a model processing device is provided, comprising:
[0030] A first sending module, configured to send a first message to at least one second device, where the first message includes a first model file corresponding to the first model;
[0031] A first receiving module, configured to receive a second message returned by the second device;
[0032] The second message includes at least one of the following:
[0033] Processing result information, used to indicate the result obtained by processing the first model file;
[0034] A second model file is obtained by converting the first model file.
[0035] In a sixth aspect, a model processing device is provided, comprising:
[0036] A second receiving module is configured to receive a first message sent by a first device, where the first message includes a first model file corresponding to the first model;
[0037] A returning module, configured to return a second message to the first device;
[0038] The second message includes at least one of the following:
[0039] Processing result information, used to indicate the result obtained by processing the first model file;
[0040] A second model file is obtained by converting the first model file.
[0041] In a seventh aspect, a model execution device is provided, comprising:
[0042] an acquisition module, configured to acquire a fifth message, wherein the fifth message includes a first model file or a second model file corresponding to the first model, wherein the second model file is obtained by converting the first model file;
[0043] An execution module is used to execute the first model file or the second model file.
[0044] In an eighth aspect, a model registration device is provided, comprising:
[0045] A second sending module is configured to send a third message to the first device, where the third message includes a first model file corresponding to the first model, and the third message is used to register the first model;
[0046] A third receiving module, configured to receive a fourth message returned by the first device;
[0047] The fourth message includes at least one of the following:
[0048] Processing result information, used to indicate the result obtained by processing the first model file;
[0049] A second model file is obtained by converting the first model file.
[0050] In a ninth aspect, a first terminal is provided, comprising a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the third aspect are implemented.
[0051] In the tenth aspect, a first terminal is provided, comprising a processor and a communication interface, wherein the communication interface is used to obtain a fifth message, the fifth message including a first model file or a second model file corresponding to the first model, and the second model file is obtained by converting the first model file; the processor is used to execute the first model file or the second model file.
[0052] In the eleventh aspect, a communication device is provided, comprising a processor and a memory, wherein the memory stores programs or instructions, and when the programs or instructions are executed by the processor, the steps of the model processing method as described in the first aspect are implemented, or the steps of the model processing method as described in the second aspect are implemented, or the steps of the model execution method as described in the third aspect are implemented, or the steps of the model registration method as described in the fourth aspect are implemented.
[0053] In the twelfth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the model processing method as described in the first aspect are implemented, or the steps of the model processing method as described in the second aspect are implemented, or the steps of the model execution method as described in the third aspect are implemented, or the steps of the model registration method as described in the fourth aspect are implemented.
[0054] In the thirteenth aspect, a wireless communication system is provided, including: a first device, a second device, a first terminal and a third device, wherein the first device can be used to execute the steps of the model processing method as described in the first aspect, the second device can be used to execute the steps of the model processing method as described in the second aspect, the first terminal can be used to execute the steps of the model execution method as described in the third aspect, and the third device can be used to execute the steps of the model registration method as described in the fourth aspect.
[0055] In the fourteenth aspect, a chip is provided, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the model processing method as described in the first aspect, or the model processing method as described in the second aspect, or the model execution method as described in the third aspect, or the model registration method as described in the fourth aspect.
[0056] In the fifteenth aspect, a computer program / program product is provided, which is stored in a storage medium, and is executed by at least one processor to implement the steps of the model processing method as described in the first aspect, or the steps of the model processing method as described in the second aspect, or the steps of the model execution method as described in the third aspect, or the steps of the model registration method as described in the fourth aspect.
[0057] In an embodiment of the present application, a first message is sent to at least one second device via a first device, the first message including a first model file corresponding to the first model, and then the first device receives a second message returned by the second device, the second message including processing result information and at least one item in the second model file, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. The first device can know whether the first model file can be correctly executed in the second device and the terminal associated with the second device based on the processing result information, and the first device can determine whether the second model file can be correctly executed in the second device and the terminal associated with the second device based on the second message including the second model file. Therefore, the first device can determine whether the first model can be correctly executed in the terminal associated with the second device based on the received processing result information or the second model file, thereby avoiding the problem that the terminal cannot use the model normally. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] FIG1 is a block diagram of a wireless communication system to which embodiments of the present application may be applied;
[0059] FIG2 is a flow chart of a model processing method according to an embodiment of the present invention;
[0060] FIG3 is a second flow chart of the model processing method provided in an embodiment of the present application;
[0061] FIG4 is a flow chart of a model execution method provided in an embodiment of the present application;
[0062] FIG5 is a flow chart of a model registration method according to an embodiment of the present application;
[0063] FIG6 is a schematic diagram of signaling interaction of a model processing, registration, and execution method provided in an embodiment of the present application;
[0064] FIG7 is a second schematic diagram of signaling interaction of the model processing, registration and execution method provided in an embodiment of the present application;
[0065] FIG8 is a third schematic diagram of signaling interaction of the model processing, registration and execution method provided in an embodiment of the present application;
[0066] FIG9 is a fourth schematic diagram of signaling interaction of the model processing, registration and execution method provided in an embodiment of the present application;
[0067] FIG10 is a fifth schematic diagram of signaling interaction of the model processing, registration and execution method provided in an embodiment of the present application;
[0068] FIG11 is a sixth schematic diagram of signaling interaction of the model processing, registration and execution method provided in an embodiment of the present application;
[0069] FIG12 is a seventh schematic diagram of signaling interaction of the model processing, registration and execution method provided in an embodiment of the present application;
[0070] FIG13 is a schematic diagram of a structure of a model processing device according to an embodiment of the present application;
[0071] FIG14 is a second structural diagram of the model processing device provided in an embodiment of the present application;
[0072] FIG15 is a schematic diagram of the structure of a model execution device provided in an embodiment of the present application;
[0073] FIG16 is a schematic diagram of the structure of a model registration device provided in an embodiment of the present application;
[0074] FIG17 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application;
[0075] FIG18 is a schematic structural diagram of a first terminal provided in an embodiment of the present application;
[0076] FIG19 is a schematic diagram of the structure of a third device provided in an embodiment of the present application;
[0077] FIG20 is a second structural diagram of the third device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0078] The following will be combined with the accompanying drawings in the embodiments of this application to clearly describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of this application.
[0079] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and do not limit the number of objects, for example, the first object can be one or more. In addition, "or" in this application represents at least one of the connected objects. For example, "A or B" covers three options, namely, Option 1: including A but not including B; Option 2: including B but not including A; Option 3: including both A and B. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.
[0080] The term "indication" in this application can be either a direct indication (or explicit indication) or an indirect indication (or implicit indication). A direct indication can be understood as the sender explicitly informing the receiver of specific information, the operation to be performed, or the requested result, etc. in the instruction sent; an indirect indication can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the operation to be performed or the requested result, etc. based on the judgment result.
[0081] It is worth noting that the technology described in the embodiments of the present application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency Division Multiple Access (SC-FDMA) or other systems. The terms "system" and "network" in the embodiments of the present application are often used interchangeably, and the described technology can be used for the systems and radio technologies mentioned above, as well as for other systems and radio technologies. The following description describes a New Radio (NR) system for example purposes, and NR terminology is used in most of the following description, but these technologies can also be applied to systems other than NR systems, such as 6th Generation (6G) communication systems.
[0082] FIG1 is a block diagram of a wireless communication system applicable to an embodiment of the present application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR), a virtual reality (VR) device, a robot, a wearable device (Wearable Device), an aircraft (Flight Vehicle), a vehicle-mounted device (VUE), a ship-mounted device, a pedestrian user equipment (PUE), a smart home (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), a game console, a personal computer (PC), a teller machine, or a self-service machine, or other terminal-side devices. Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle-mounted device can also be called a vehicle-mounted terminal, a vehicle-mounted controller, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit, etc. In addition to the above-mentioned terminal devices, it can also be a chip in the terminal, such as a modem chip, a system-on-chip (SoC). It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network side device 12 may include an access network device or a core network device, wherein the access network device may also be called a radio access network (RAN) device, a radio access network function or a radio access network unit. The access network device may include a base station, a wireless local area network (WLAN) access point (AS) or a wireless fidelity (WiFi) node, etc.Among them, the base station can be referred to as Node B (NB), Evolved Node B (eNB), the next generation Node B (gNB), New Radio Node B (NR Node B), access point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B (HNB), Home Evolved Node B (home evolved Node B), Transmission Reception Point (TRP) or other appropriate terms in the field. As long as the same technical effect is achieved, the base station is not limited to specific technical vocabulary. It should be noted that in the embodiment of the present application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited.
[0083] The core network equipment may include but is not limited to at least one of the following: core network node, core network function, mobility management entity (MME), access mobility management function (AMF), session management function (SMF), user plane function (UPF), policy control function (PCF), policy and charging rules function unit (PCRF), edge application service discovery function (EASDF), unified data management (UDM), unified data repository (UDR), home user server (HSS), centralized network configuration (CNC), network storage function (NRF), network exposure function (NEF), local NEF (L-NEF), binding support function (BSF), application function ( It should be noted that in the embodiments of the present application, only the core network device in the NR system is introduced as an example, and the specific type of the core network device is not limited.
[0084] The following, in combination with the accompanying drawings, describes in detail the model processing, execution and registration methods provided in the embodiments of the present application through some embodiments and their application scenarios.
[0085] First, the model registration method provided by the embodiment of the present application is described. In the related art, it is impossible to uniformly manage the models generated or trained by different network-side devices or terminals, resulting in inconsistent understanding of the model by the terminal and the network-side device when the terminal uses the model, thereby preventing the terminal from using the model normally. To solve this technical problem, the embodiment of the present application provides a model registration method, including steps 1 to 3, specifically:
[0086] Step 1: The third device sends a third message to the first device, where the third message includes a first model file corresponding to the first model, and the third message is used to register the first model.
[0087] Optionally, the third message is used to register the first model, which can also be understood as the third message being used to publish the first model. The subsequent steps are the same and will not be described in detail.
[0088] Optionally, the first device is used to store the model, and the first device may include at least one of the following: a) a model store (Model Store), used to store the model; b) NEF; c) an analytical database function (Analytic Data Repository Function, ADRF); d) NRF.
[0089] Optionally, the first model file includes: structural information of the first model or parameter information of the first model. The parameter information includes: a parameter type list or parameter values.
[0090] Optionally, the first model file can be understood as the model file itself; or, storage address information of the model file, such as information indicating the download address, Uniform Resource Locator (URL) location, or storage device of the first model file.
[0091] The third device is used to generate or train a model, and the third device may include at least one of the following: a) a core network device of a 3rd Generation Partnership Project (3GPP) network; b) a RAN of a 3GPP network; c) a target server device, such as a third-party server (3rd Server); d) a second terminal, the second terminal being a terminal capable of generating or training a model; the second terminal may be the same as or different from the first terminal. Optionally, the core network device includes but is not limited to the types of network-side devices 12 listed above, and this application does not limit this.
[0092] Optionally, the third message may further include first information, where the first information may include at least one of the following:
[0093] 1) Information indicating the version number of the first model;
[0094] 2) information indicating at least one parameter corresponding to the first model;
[0095] 3) information indicating the usage conditions or usage scenarios of the first model;
[0096] 4) information indicating at least one function supported by the first model;
[0097] 5) Information indicating a generating device or a training device for the first model;
[0098] 6) Information indicating a device using the first model.
[0099] Optionally, the version number information of the first model can also be understood as the version number (version number) or version identification (identity, ID) of the first model. The subsequent description is the same and will not be repeated.
[0100] Optionally, the information of the at least one parameter corresponding to the first model can also be understood as a parameter list. The information of the at least one parameter corresponding to the first model is used to indicate the input and output parameters of the model. The following is the same and will not be repeated here.
[0101] Optionally, the information used to indicate the use conditions or use scenarios of the first model can also be understood as activation conditions. This information is used to indicate the environmental parameters used to activate the first model. The environmental parameters may include: time parameters, such as activation at a certain time point or time period; location parameters, such as activation at a certain tracking area (TA), cell, site, or longitude and latitude; channel environment parameters, such as antenna port, carrier frequency, bandwidth, and separate channel signaling (SCS); channel characteristics, such as signal-to-noise ratio (SNR), line of sight (LOS), and non-line of sight (NLOS). The following is the same and will not be repeated.
[0102] Optionally, the information about the functions supported by the first model is used to indicate that the first model is specifically used for reasoning purposes of a certain task type (identified by a function type or analytic ID). For example, the reasoning purpose may be trajectory prediction, channel state information (CSI) compression / decompression, AI positioning, beam prediction, etc.
[0103] Optionally, the information of the generating device or training device of the first model can also be understood as the producer identification, such as ID, fully qualified domain name (FQDN), IP address, etc.; the producer manufacturer can include the vendor name (vendor name) or vendor identification (vendor ID).
[0104] Optionally, the information of the device using the first model can also be understood as a consumer identifier, such as an ID, FQDN, IP address, Type Allocation Code (TAC), or consumer manufacturer, which can include a vendor name or vendor ID. This information is used to limit the consumers of the model to a specific range.
[0105] Step 2: The first device receives a third message sent by a third device; the first device assigns a first identifier to the first model;
[0106] Optionally, the first identifier may be a model identifier (model ID).
[0107] Step 3: The first device sends a fourth message to the third device;
[0108] The fourth message may include at least one of the following:
[0109] Processing result information, used to indicate the result obtained by processing the first model file;
[0110] The second model file is obtained by converting the first model file.
[0111] Optionally, the processing result information may be used to indicate a result obtained by the second device processing the first model file; the second model file is obtained by the second device converting the first model file.
[0112] Optionally, the fourth message may further include at least one of the following: a first identifier; the first model file; first information; and first indication information for indicating information of a manufacturer corresponding to the second device.
[0113] The model registration method provided in the embodiment of the present application is that when the first device receives the third message for registering the first model, it assigns a globally unique first identifier to the first model, thereby registering the first model to the first device, thereby realizing unified management of models generated by different model training parties, so that the network side and the terminal have a unified understanding of the first model.
[0114] FIG2 is a flow chart of a model processing method according to an embodiment of the present application. As shown in FIG2 , the method includes steps 201-202; wherein:
[0115] Step 201: A first device sends a first message to at least one second device, where the first message includes a first model file corresponding to a first model.
[0116] It should be noted that the embodiments of the present application can be applied to scenarios where models generated or trained by different devices are managed and adapted. Because different UE vendors in the related art have their own optimization methods for models, when different model trainers send their trained models to the first terminal, some models may not be correctly executed in the first terminal, resulting in model incompatibility.
[0117] To solve this problem, in an embodiment of the present application, a first device sends a first message to at least one second device, requesting each second device to process the first model separately, including at least one of the following:
[0118] a) testing various parameters that can be used by the model to check whether the first model can be used normally in the second device, or whether the first model can be used normally in the terminal device associated with the second device;
[0119] b) Perform security authentication on the model.
[0120] Optionally, the second device may also convert the first model file, such as mode conversion or format conversion. Mode conversion or format conversion refers to converting the first model from a model mode not supported by the terminal manufacturer or terminal device associated with the second device to a model mode supported by the terminal manufacturer. The first model may include an AI model or an ML model, etc. The model mode of the first model is, for example, the encoding format of the first model file, for example, the Open Neural Network Exchange (ONNX) model format.
[0121] Optionally, the second device processes the first model, which can also be understood as the second device checking, testing, validating, checking out, or verifying the first model. The subsequent steps are the same and will not be repeated here.
[0122] Alternatively, the model mode can also be understood as the model format. The following is the same and will not be repeated here.
[0123] Optionally, the first device includes at least one of the following: a) Model Store; b) NEF; c) ADRF; d) NRF. Optionally, the second device includes a UE vendor server (UE Vendor Sever).
[0124] It should be noted that the first message includes the first model file corresponding to the first model. Since different UE Vendors have their own optimization methods for models, after the second device receives the first message including the first model file from the first device, the second device verifies the model mode of the first model based on the first message; accordingly, the second device may support the model mode of the first model file, or may not support the model mode of the first model file.
[0125] Optionally, verifying the model pattern of the first model can also be understood as checking, testing, confirming, or inspecting the first model. The subsequent steps are the same and will not be described in detail.
[0126] When the second device supports the model mode of the first model file, it is necessary to process the first model file, such as check, test, confirm, inspect or verify, to determine whether the first model file can be normally used on the second device.
[0127] Optionally, the first model file can be used normally on the second device, which can also be understood as the first model file can be used normally on the terminal device associated with the second device.
[0128] If the second device does not support the model mode of the first model file, it needs to convert the first model file to generate a second model file of the first model, and the model mode of the second model file is the model mode supported by the second device; after the second device generates the second model file, it needs to check, test, confirm, inspect or verify the second model file to ensure that the second model file can be used normally on the second device.
[0129] Optionally, the second model file can be used normally on the second device, which can also be understood as the second model file can be used normally on the terminal device associated with the second device.
[0130] Optionally, in this application, normal use can also be understood as normal operation, correct operation, or problem-free operation. The following is the same and will not be repeated.
[0131] Step 202: The first device receives a second message returned by the second device; the second message includes at least one of the following: processing result information, used to indicate the result obtained by processing the first model file; a second model file, which is obtained by converting the first model file.
[0132] Optionally, the second model file can be correctly executed in a first terminal associated with the second device. Optionally, the first terminal includes but is not limited to the types of terminals 11 listed above.
[0133] In the embodiment of the present application, the second message is, for example, a model check response message or a model conversion response message.
[0134] Optionally, the processing result information may indicate a result of the second device checking, testing, confirming, verifying or validating the first model file. For example, if the processing result information indicates a test result, the test result includes a test success or a test failure.
[0135] If the test result includes a test success, it indicates that the first model file or the second model file can be used normally on the second device. If the test result includes a test failure, it indicates that the first model file or the second model file cannot be used normally on the second device, that is, the second device does not support the first model.
[0136] For example, if the second device tests that the first model file can be used normally on the second device, the second device returns processing result information to the first device, indicating that the test result is successful. For another example, if the second device tests that the first model file cannot be used normally on the second device, the second device converts the first model file to generate a second model file; then the second device tests the second model file. If the second model file can be used normally on the second device, the second device returns processing result information and the second model file to the first device, indicating that the test result of the second model file is successful.
[0137] In the case that the test result indicated by the processing result information is a test failure, the second device only returns the processing result information to the first device, indicating that the test result is a test failure, indicating that the second device does not support the first model.
[0138] Optionally, the second message may further include first indication information, used to indicate the manufacturer associated with the second device; the first indication information may be, for example, a vendor ID.
[0139] Accordingly, when the first device receives the second messages returned by multiple second devices, the first device can determine which second device the second message comes from based on the first indication information.
[0140] In an embodiment of the present application, a first message is sent to at least one second device via a first device, the first message including a first model file corresponding to the first model, and then the first device receives a second message returned by the second device, the second message including processing result information and at least one item in the second model file, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. The first device can know whether the first model file can be correctly executed in the second device and the terminal associated with the second device based on the processing result information, and the first device can determine whether the second model file can be correctly executed in the second device and the terminal associated with the second device based on the second message including the second model file. Therefore, the first device can determine whether the first model can be correctly executed in the terminal associated with the second device based on the received processing result information or the second model file, thereby avoiding the problem that the terminal cannot use the model normally.
[0141] Optionally, before the first device sends the first message to at least one second device, it is also necessary to receive a registration request from the model training party, and then register the model trained by the model training party to the first model. This can be achieved by the following steps:
[0142] The first device receives a third message sent by a third device, where the third message includes the first model file and is used to register the first model.
[0143] In this embodiment of the present application, the third device is used to train the model and generate a trained first model. After generating the first model, the third device sends a third message to the first device. In response, the first device needs to receive the third message sent by the third device. The third message is used to register the first model with the first device, so that the first device can store the first model.
[0144] Optionally, the third device includes at least one of the following:
[0145] a) Core network equipment;
[0146] b)RAN;
[0147] c) Target server device, such as 3rd Server;
[0148] d) The second terminal can be the same as the first terminal or different from the first terminal; that is, the other terminal can train the model and then register the trained first model to the first device, or the first terminal can train the model and then register the trained first model to the first device.
[0149] Optionally, before the first device sends the first message to at least one second device, the first device assigns a first identifier to the first model.
[0150] That is to say, the first device needs to assign a globally unique identifier to the first model associated with the first model file, such as a first identifier, model ID, etc.; the first device assigns the first identifier to the first model, thereby achieving unified management of models generated by different model training parties, so that the network side and the terminal have a unified understanding of the first model.
[0151] For example, in a scenario where the network side instructs the first terminal to execute the first model, the first device assigns a globally unique first identifier to the first model, and the network side instructs the first terminal to execute the first model based on the first identifier, so that the network side and the first terminal have a unified understanding of the first model, thereby achieving unified management of models generated by different model training parties.
[0152] Optionally, the first message further includes the first identifier. That is, the first message sent by the first device to the second device further includes the first identifier of the first model, so that the first device and the second device have a unified understanding of the first model.
[0153] Optionally, the first message further includes first information, where the first information includes at least one of the following:
[0154] a) Information indicating the version number of the first model, such as the version ID or version number corresponding to the first model.
[0155] b) Information indicating at least one parameter corresponding to the first model, for example, a Parameter list corresponding to the first model.
[0156] c) Information used to indicate the usage conditions or usage scenarios of the first model, for example, the conditions under which the first model can be used, that is, the validity conditions (validity condition) or applicable conditions (applicable condition) of the first model, including time conditions, for example, it can only be activated at a certain time point or time period; location conditions, for example, activation at a certain TA, cell, site, longitude and latitude position; channel conditions, such as antenna port, carrier frequency, bandwidth, SCS, channel characteristics; scene conditions; speed conditions; access cell range or cell ID list, etc.
[0157] d) Information indicating at least one function (Function type or analytic ID) supported by the first model, for example, a function (Function) supported by the first model, such as predicting cell signal strength.
[0158] e) Information indicating a generating device or a training device of the first model, for example, the producer of the first model.
[0159] f) Information indicating a device using the first model, for example, a consumer of the first model.
[0160] In the above embodiment, the second device can test the first information included in the first message to ensure that the first information can be correctly executed in the second device, and further ensure that the first information can be correctly executed in the first terminal associated with the second device, avoiding the problem of incompatibility of the first terminal when using the first model.
[0161] Optionally, after the first device receives the second message returned by the second device, it is also necessary to send a registration result or a publishing result, such as a model publish ack message, to the third device, indicating that the first model has been registered or published in the first device. This can be achieved by the following steps:
[0162] The first device sends a fourth message to the third device;
[0163] The fourth message includes at least one of the following:
[0164] a) the processing result information;
[0165] b) The second model file is obtained by converting the first model file.
[0166] Optionally, the fourth message further includes at least one of the following:
[0167] a) first identification;
[0168] b) the first model file;
[0169] c) first information;
[0170] d) First indication information, used to indicate information about the manufacturer corresponding to the second device, for example, a vendor ID.
[0171] Optionally, after the first device sends the registration or publishing result to the third device, it is necessary to send information indicating that the second device has successfully tested and associated with the first model to the first terminal to ensure that the first model can be correctly executed in the first terminal. This can be achieved by the following steps:
[0172] The first device sends a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
[0173] When the fifth message includes the first model file, it means that the first device directly sends the first model file that can be correctly executed in the second device to the first terminal; when the fifth message includes the second model file, it means that the first model file cannot be correctly executed in the second device. After the second device converts the first model file, the first device sends the second model file that can be correctly executed in the second device to the first terminal.
[0174] Optionally, the fifth message further includes at least one of the following: a first identifier; and first information, where the first information includes at least one of the following:
[0175] a) information indicating the version number of the first model;
[0176] b) information indicating at least one parameter corresponding to the first model;
[0177] c) information indicating a usage condition or usage scenario of the first model;
[0178] d) information indicating at least one function supported by the first model;
[0179] e) information indicating a generating device or a training device of the first model;
[0180] f) Information indicating a device using the first model.
[0181] In the above embodiment, the first device sends the first identifier to the first terminal so that when the network side instructs the first terminal to execute the first model, the first terminal and the network side can have a unified understanding; since the first information is information that has been successfully tested by the second device associated with the first terminal, the first device sends the first information to the first terminal, which can ensure that the first terminal can correctly execute the first information.
[0182] Optionally, a manner in which the first device sends the fifth message to the first terminal includes at least one of the following:
[0183] Mode 1: When the first terminal has been registered with the first device, the first device sends the fifth message to the first terminal.
[0184] For example, the first terminal first registers in the Model Store, and then the Model Store can directly send the fifth message to the first terminal.
[0185] Method 2: When the first device receives a first request from the first terminal, it sends a sixth message to the first terminal, where the first request is used to query whether there is any model file that has not been downloaded or has been updated, and the sixth message is used to indicate whether there is a model file that has not been downloaded or has been updated in the first device; when the first device receives a second request from the first terminal, it sends the fifth message to the first terminal; the second request is used to obtain the fifth message.
[0186] Optionally, the first request includes second indication information for indicating the manufacturer corresponding to the first terminal, such as a vendor ID or terminal model information. The second indication information is carried in the International Mobile Equipment Identity (IMEI) of the first terminal.
[0187] For example, the first terminal periodically requests the Model Store whether there is any model file that has not been downloaded or has been updated; when there is any model file that has not been downloaded or has been updated in the Model Store, the first terminal is notified; and the first terminal obtains the model file from the Model Store according to the notification.
[0188] Method 3: When the first device receives the seventh message from the first terminal, it sends the fifth message to the first terminal; the seventh message includes at least one of the following: a first identifier; information used to indicate the version number of the first model; second indication information, used to indicate information about the manufacturer or terminal model corresponding to the first terminal.
[0189] For example, in a scenario where the network side instructs the first terminal to execute the first model, the network side may first send a list of model identifiers (model ID list) supported by it to the first terminal, wherein the model ID list includes the first identifier; the first terminal compares the model ID list supported by itself with the model ID list supported by the network side, and if the first terminal determines that it lacks the first identifier, it sends a seventh message to the Model Store to request the Model Store to send the fifth message to the first terminal;
[0190] Optionally, the first terminal determining that it lacks the first identifier can also be understood as the first terminal determining that it does not save the first identifier, or does not save the first model corresponding to the first identifier.
[0191] Among them, the seventh message includes at least one of the following items: a first identifier; information used to indicate the version number of the first model, such as version ID; second indication information, used to indicate the manufacturer corresponding to the first terminal, such as vender ID, or terminal model information, wherein the second indication information is carried in the IMEI of the first terminal.
[0192] Optionally, the seventh message may further carry model capability information supported by the first terminal, and the Model Store may send the fifth message to the first terminal based on the model capability information supported by the first terminal.
[0193] FIG3 is a second flow chart of the model processing method provided in an embodiment of the present application. As shown in FIG3 , the method includes steps 301-302; wherein:
[0194] Step 301: The second device receives a first message sent by the first device, where the first message includes a first model file corresponding to a first model.
[0195] Optionally, the first message may be, for example, a model check request message, used to instruct the second device to verify the first model, thereby checking whether the first model can be used normally in the second device. The first model may be an AI model, an ML model, etc.
[0196] Optionally, the first device is used to store the first model trained by each model training party, and the first device includes at least one of the following: a) Model Store; b) NEF; c) ADRF; d) NRF.
[0197] Optionally, the second device is used to verify the first model, and the second device includes a UE vendor server (UE Vendor Sever).
[0198] It should be noted that the first message includes the first model file corresponding to the first model. Since different UE Vendors have their own optimization methods for models, after the first device sends the first message including the first model file to the second device, the second device needs to process the first model file of the first model, such as checking, testing, confirming, inspecting or verifying, to determine whether the first model file can be used normally on the second device.
[0199] It should be noted that the terminal device associated with the second device can be understood as a terminal device produced, developed or sold by the UE manufacturer (UE Vendor) corresponding to the UE manufacturer server.
[0200] For example, the second device verifies the encoding format of the first model file; accordingly, the second device may support the encoding format of the first model file, or may not support the encoding format of the first model file.
[0201] Step 302: The second device returns a second message to the first device; the second message includes at least one of the following: processing result information, used to indicate the result obtained by processing the first model file; a second model file, which is obtained by converting the first model file.
[0202] Optionally, the second model file can be correctly executed in a first terminal associated with the second device.
[0203] Optionally, the second message may further include first indication information, used to indicate the manufacturer associated with the second device; the first indication information may be, for example, a vendor ID.
[0204] Correspondingly, when the second device returns the second message to the first device, the first device can determine which second device the second message comes from based on the first indication information.
[0205] In an embodiment of the present application, the second device receives a first message sent by the first device, the first message includes a first model file corresponding to the first model, and then the second device returns a second message to the first device, the second message includes processing result information and at least one item in the second model file, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. The first device can know whether the first model file can be correctly executed in the second device and the terminal associated with the second device based on the processing result information, and the first device can determine whether the second model file can be correctly executed in the second device and the terminal associated with the second device based on the inclusion of the second model file in the second message. Therefore, the second device returns the processing result information or the second model file to the first device, which can enable the first device to determine whether the first model can be correctly executed in the terminal associated with the second device, thereby avoiding the problem of the terminal being unable to use the model normally.
[0206] Optionally, the first message further includes a first identifier, which is allocated by the first device to the first model.
[0207] That is, the first message received by the second device from the first device includes the first identifier of the first model, so that the first device and the second device have a unified understanding of the first model.
[0208] Optionally, the first message further includes first information, where the first information includes at least one of the following:
[0209] a) information indicating the version number of the first model;
[0210] b) information indicating at least one parameter corresponding to the first model;
[0211] c) information indicating a usage condition or usage scenario of the first model;
[0212] d) information indicating at least one function supported by the first model;
[0213] e) information indicating a generating device or a training device of the first model;
[0214] f) Information indicating a device using the first model.
[0215] Optionally, after the second device returns the second message to the first device, it is also necessary to send information indicating that the second device has successfully tested and associated with the first model to the first terminal to ensure that the first model can be correctly executed in the first terminal. This can be achieved by the following steps:
[0216] The second device sends a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
[0217] Specifically, when the fifth message includes the first model file, it means that the second device directly sends the first model file that can be correctly executed in the second device to the first terminal; when the fifth message includes the second model file, it means that the first model file cannot be correctly executed in the second device, and the second device converts the first model file and then sends the second model file that can be correctly executed in the second device to the first terminal.
[0218] Optionally, the fifth message further includes at least one of the following: a first identifier; and first information, where the first information includes at least one of the following:
[0219] a) information indicating the version number of the first model;
[0220] b) information indicating at least one parameter corresponding to the first model;
[0221] c) information indicating a usage condition or usage scenario of the first model;
[0222] d) information indicating at least one function supported by the first model;
[0223] e) information indicating a generating device or a training device of the first model;
[0224] f) Information indicating a device using the first model.
[0225] Optionally, the second device sends the fifth message to the first terminal in at least one of the following ways:
[0226] Mode 1: When the first terminal has been registered with the second device, the second device sends the fifth message to the first terminal.
[0227] For example, the first terminal first registers with the UE Vendor Sever, and then the UE Vendor Sever may directly send the fifth message to the first terminal.
[0228] Method 2: When the second device receives the third request from the first terminal, it sends the eighth message to the first terminal, where the third request is used to query whether there is any model file that has not been downloaded or has been updated, and the eighth message is used to indicate that there is a model file that has not been downloaded or has been updated in the second device; when the second device receives the fourth request from the first terminal, it sends the fifth message to the first terminal; the fourth request is used to obtain the fifth message.
[0229] Optionally, the third request includes second indication information for indicating the manufacturer corresponding to the first terminal, such as vendor ID or terminal model information. The second indication information is carried in the IMEI of the first terminal.
[0230] The UE Vendor Sever may push the fifth message to the first terminal through the application layer, such as an Over The Top (OTT) service that goes beyond the operator's pipeline.
[0231] For example, the first terminal periodically requests the UE Vendor Sever whether there is any model file that has not been downloaded or has been updated; when the UE Vendor Sever has any model file that has not been downloaded or has been updated, the first terminal is notified; and the first terminal obtains the model file from the UE Vendor Sever according to the notification.
[0232] Method 3: When the second device receives the ninth message from the first terminal, it sends the fifth message to the first terminal; the ninth message includes at least one of the following: a first identifier; information used to indicate the version number of the first model; second indication information, used to indicate the manufacturer information or terminal model information corresponding to the first terminal.
[0233] For example, in a scenario where the network side instructs the first terminal to execute the first model, the network side may first send a model identifier list (model ID list) supported by it to the first terminal, wherein the model ID list includes the first identifier; the first terminal compares the model ID list supported by itself with the model ID list supported by the network side. When the first terminal determines that it lacks the first identifier, it sends a ninth message to the UE Vendor Sever to request the UE Vendor Sever to send a fifth message to the first terminal;
[0234] Among them, the ninth message includes at least one of the following items: a first identifier; information used to indicate the version number of the first model, such as version ID; second indication information, used to indicate the manufacturer corresponding to the first terminal, such as vender ID, or terminal model information, wherein the second indication information is carried in the IMEI of the first terminal.
[0235] Optionally, the ninth message may further carry model capability information supported by the first terminal, and the UE Vendor Sever may send the fifth message to the first terminal based on the model capability information supported by the first terminal.
[0236] FIG4 is a flow chart of a model execution method provided in an embodiment of the present application. As shown in FIG4 , the method includes steps 401-402; wherein:
[0237] Step 401: The first terminal obtains a fifth message, where the fifth message includes a first model file or a second model file corresponding to a first model, where the second model file is obtained by converting the first model file.
[0238] Step 402: The first terminal executes the first model file or the second model file.
[0239] It should be noted that, when the fifth message includes the first model file, it means that the first model file can be normally used on the second device associated with the first terminal;
[0240] When the second model file is included in the fifth message, it means that the first model file cannot be used normally on the second device associated with the first terminal. The second device converts the first model file to generate a second model file that can be used normally on the second device to ensure that the second model file can be used normally on the first terminal.
[0241] Optionally, the second model file is obtained by converting the first model file by a second device associated with the first terminal, and the second model file can be correctly executed in the first terminal associated with the second device.
[0242] The first terminal avoids the problem of incompatibility when using the model by executing the first model file or the second model file.
[0243] In an embodiment of the present application, the first terminal obtains a fifth message, which includes a first model file or a second model file corresponding to the first model, wherein the second model file is obtained by converting the first model file. The first terminal executes the first model file or the second model file, which can ensure that the first model file or the second model file can be correctly executed in the first terminal, thereby avoiding the problem that the first terminal cannot use the model normally.
[0244] Optionally, the fifth message further includes at least one of the following:
[0245] A first identifier, used to identify the first model; and first information, wherein the first information includes at least one of the following:
[0246] a) information indicating the version number of the first model;
[0247] b) information indicating at least one parameter corresponding to the first model;
[0248] c) information indicating a usage condition or usage scenario of the first model;
[0249] d) information indicating at least one function supported by the first model;
[0250] e) information indicating a generating device or a training device of the first model;
[0251] f) Information indicating a device using the first model.
[0252] In the above embodiment, the first terminal obtains the first identifier so that when the network side instructs the first terminal to execute the first model, the first terminal and the network side can have a unified understanding; since the first information is information that has been successfully tested by the second device associated with the first terminal, it can be guaranteed that the first terminal can correctly execute the first information.
[0253] Optionally, the first terminal may obtain the fifth message by at least one of the following methods 1 and 2:
[0254] Mode 1: The first terminal receives the fifth message from the first device.
[0255] Optionally, the first device is used to store the first model trained by each model training party, and the first device includes at least one of the following: a) Model Store; b) NEF; c) ADRF; d) NRF.
[0256] For example, the first terminal first registers in the Model Store, and then the Model Store can directly send the fifth message to the first terminal.
[0257] Method 2: The first terminal sends a first request to the first device; the first terminal receives a sixth message from the first device, where the first request is used to query whether there is any model file that has not been downloaded or has been updated, and the sixth message is used to indicate that there is a model file that has not been downloaded or has been updated in the first device; the first terminal sends a second request to the first device, where the second request is used to obtain the fifth message; the first terminal receives the fifth message from the first device.
[0258] Optionally, the first request includes second indication information for indicating the manufacturer corresponding to the first terminal, such as vender ID, or terminal model information; the second indication information is carried in the IMEI of the first terminal.
[0259] For example, the first terminal periodically requests the Model Store whether there is any model file that has not been downloaded or has been updated; when there is any model file that has not been downloaded or has been updated in the Model Store, the first terminal is notified; and the first terminal obtains the model file from the Model Store according to the notification.
[0260] Optionally, the first terminal may obtain the fifth message by at least one of the following methods 3 and 4:
[0261] Mode 3: The first terminal receives the fifth message from the second device.
[0262] For example, the first terminal first registers with the UE Vendor Sever, and then the UE Vendor Sever may directly send the fifth message to the first terminal.
[0263] Method 4: The first terminal sends a third request to the second device; the first terminal receives an eighth message from the second device, the third request is used to query whether there is any model file that has not been downloaded or has been updated, and the eighth message is used to indicate that there is a model file that has not been downloaded or has been updated in the second device; the first terminal sends a fourth request to the second device, and the fourth request is used to obtain the fifth message; the first terminal receives the fifth message from the second device.
[0264] Optionally, the third request includes second indication information for indicating the manufacturer corresponding to the first terminal, such as vendor ID or terminal model information. The second indication information is carried in the IMEI of the first terminal.
[0265] For example, the first terminal periodically requests the UE Vendor Sever whether there is any model file that has not been downloaded or has been updated; when the UE Vendor Sever has any model file that has not been downloaded or has been updated, the first terminal is notified; and the first terminal obtains the model file from the UE Vendor Sever according to the notification.
[0266] Optionally, the first terminal may obtain the fifth message by at least one of the following methods 5 and 6:
[0267] Mode 5: The first terminal receives the fifth message from the third device.
[0268] Optionally, the third device includes at least one of the following: a) a core network device; b) an access network device; c) a target server device, such as a 3rd Server; d) a second terminal, which may be the same as or different from the first terminal.
[0269] For example, the first terminal first registers with the core network device, and then the core network device can directly send the fifth message to the first terminal.
[0270] Method 6: The first terminal sends a fifth request to the third device; the first terminal receives the tenth message from the third device, the fifth request is used to query whether there is any undownloaded or updated model file in the third device, and the tenth message is used to indicate that there is any undownloaded or updated model file in the third device; the first terminal sends a sixth request to the third device, and the sixth request is used to obtain the fifth message; the first terminal receives the fifth message from the third device.
[0271] Optionally, the fifth request includes second indication information for indicating the manufacturer corresponding to the first terminal, such as vendor ID or terminal model information. The second indication information is carried in the IMEI of the first terminal.
[0272] For example, the first terminal periodically requests the core network device whether there is a model file that has not been downloaded or has been updated; when there is a model file that has not been downloaded or has been updated in the core network device, the first terminal is notified; the first terminal obtains the model file from the core network device according to the notification.
[0273] Optionally, the first terminal may obtain the fifth message by using method 7, which specifically includes the following steps [1] to [2]:
[0274] Step [1], the first terminal sends an eleventh message to the third device;
[0275] Step [2], the first terminal receives the fifth message from the third device;
[0276] The eleventh message includes at least one of the following:
[0277] a) second instruction information;
[0278] b) A first model list, including an identifier or at least one version number of at least one model supported by the first terminal.
[0279] In an embodiment of the present application, the first terminal sends at least one of its corresponding manufacturer information, terminal model information, the identifier (model ID) of at least one model supported by the first terminal, and the version number (version ID) of at least one model supported by the first terminal to a third device. The third device can filter out the information associated with the first model supported by the first terminal based on the above information, and carry it in a fifth message to send to the first terminal.
[0280] Optionally, the eleventh message may further carry model capability information supported by the first terminal, and the third device may send the fifth message to the first terminal based on the model capability information supported by the first terminal.
[0281] Optionally, the eleventh message is carried by at least one of the following messages: a radio resource control (RRC) message; a non-access stratum (NAS) message.
[0282] In actual applications, there are many ways for the first terminal to send the eleventh message to the third device. For example, the first terminal can send the eleventh message to the third device in the radio capability of the RRC layer; for example, the first terminal can send the eleventh message to the access and mobility management function (Access and Mobility Management Function, AMF) through the fifth generation mobile communication technology (5th Generation Mobile Communication Technology, 5G) mobility management (mobility management, MM) capability, and then the AMF sends the eleventh message to the third device.
[0283] Optionally, the first terminal may obtain the fifth message by using method 8, which specifically includes the following steps [1] to [2]:
[0284] Step [1]: The first terminal receives a twelfth message from the third device; the twelfth message is used to indicate a second model list; the second model list includes at least one of the following: an identifier and at least one version number of at least one model supported by the third device;
[0285] Step [2]: The first terminal obtains the fifth message based on the first model list and the second model list, if it is determined that the first identifier or the information indicating the version number of the first model exists in the second model list but does not exist in the first model list.
[0286] In an embodiment of the present application, the first terminal first needs to receive a twelfth message from the third device, where the twelfth message is used to indicate a second model list, where the second model list includes an identifier of at least one model supported by the third device, such as a model ID list, and at least one version number, such as a version ID list.
[0287] Optionally, the twelfth message is carried by at least one of the following messages: a) System Information Block (SIB) message; b) RRC message; c) NAS message.
[0288] After receiving the twelfth message, the first terminal compares the first model list with the second model list, wherein the first model list includes an identifier of at least one model supported by the first terminal, such as a model ID list, and at least one version number, such as a version ID list.
[0289] If the first terminal determines that the first identifier or information indicating the version number of the first model, such as the version ID, exists in the second model list but not in the first model list, the first terminal obtains the fifth message based on the missing information in the first model list. In other words, if the second model list includes a model ID or version ID that is not in the first model list; or if the first terminal determines, based on the second model list, that the model ID or version ID that is not stored in the first model list exists in the first model list, the first terminal obtains the fifth message.
[0290] Optionally, in mode 8, when the first terminal determines that the first identifier or the information indicating the version number of the first model exists in the second model list but does not exist in the first model list, there are multiple ways to obtain the fifth message, specifically including the following modes 8.1 to 8.3:
[0291] Method 8.1: The first terminal sends a seventh message to the first device based on the first identifier or the information used to indicate the version number of the first model, and the seventh message includes the first identifier, the information used to indicate the version number of the first model and at least one of the second indication information; the first terminal receives the fifth message from the first device.
[0292] In the embodiment of the present application, the first terminal may obtain the fifth message from the Model Store.
[0293] For example, in a scenario where the network side instructs the first terminal to execute the first model, the network side may first send a list of model identifiers (model ID list) supported by it to the first terminal, wherein the model ID list includes the first identifier; the first terminal compares the model ID list supported by itself with the model ID list supported by the network side, and if the first terminal determines that it lacks the first identifier, it sends a seventh message to the Model Store to request the Model Store to send the fifth message to the first terminal;
[0294] Among them, the seventh message includes at least one of the following items: a first identifier; information used to indicate the version number of the first model, such as version ID; second indication information, used to indicate the manufacturer corresponding to the first terminal, such as vender ID, or terminal model information, wherein the second indication information is carried in the IMEI of the first terminal.
[0295] Optionally, the seventh message may further carry model capability information supported by the first terminal, and the Model Store may send the fifth message to the first terminal based on the model capability information supported by the first terminal.
[0296] Method 8.2: The first terminal sends a ninth message to the second device based on the first identifier or the information used to indicate the version number of the first model, and the ninth message includes at least one of the first identifier, the information used to indicate the version number of the first model and the second indication information; the first terminal receives the fifth message from the second device.
[0297] In this embodiment of the present application, the first terminal may obtain the fifth message from the UE Vendor Sever.
[0298] For example, when the first terminal determines that it lacks the first identifier, the first terminal sends a ninth message to the UE Vendor Sever to request the UE Vendor Sever to send the fifth message to the first terminal.
[0299] Among them, the ninth message includes at least one of the following items: a first identifier; information used to indicate the version number of the first model, such as version ID; second indication information, used to indicate the manufacturer corresponding to the first terminal, such as vender ID, or terminal model information, wherein the second indication information is carried in the IMEI of the first terminal.
[0300] Optionally, the ninth message may further carry model capability information supported by the first terminal, and the UE Vendor Sever may send the fifth message to the first terminal based on the model capability information supported by the first terminal.
[0301] Method 8.3: The first terminal sends a thirteenth message to the third device based on the first identifier or the information used to indicate the version number of the first model, and the thirteenth message includes the first identifier, the information used to indicate the version number of the first model and at least one of the second indication information; the first terminal receives the fifth message from the third device.
[0302] In this embodiment of the present application, the first terminal may obtain the fifth message from the third device.
[0303] For example, when the first terminal determines that it lacks the first identifier, the first terminal sends the thirteenth message to the core network device to request the core network device to send the fifth message to the first terminal.
[0304] Among them, the thirteenth message includes at least one of the following items: a first identifier; information used to indicate the version number of the first model, such as version ID; second indication information, used to indicate the manufacturer corresponding to the first terminal, such as vender ID, or terminal model information, wherein the second indication information is carried in the IMEI of the first terminal.
[0305] Optionally, the thirteenth message may further carry model capability information supported by the first terminal, and the UE Vendor Sever may send the fifth message to the first terminal based on the model capability information supported by the first terminal.
[0306] Optionally, in the above embodiment, after the first terminal obtains the fifth message from the first device or the second device, it is also necessary to send the first model list supported by the first terminal to the third device, so that the third device sends an activation message of the first model to the first terminal, thereby enabling the first terminal to correctly execute the first model; this can be specifically achieved by the following steps:
[0307] The first terminal sends a fifteenth message to the third device, where the fifteenth message is used to indicate a first model list, where the first model list includes an identifier or at least one version number of at least one model supported by the first terminal.
[0308] In an embodiment of the present application, the first terminal sends a first model list to the third device, so that the third device can determine whether the first terminal meets the triggering conditions of the model and whether the first terminal supports the parameters to be activated based on the identifier of at least one model supported by the first terminal or at least one version number; when the third device determines that the first terminal supports the usable conditions and parameters to be activated of the model, it can send an activation indication to activate the first model to the first terminal.
[0309] In actual applications, there are many ways for the first terminal to send the fifteenth message to the third device. For example, the first terminal can send the fifteenth message to the third device in the radio capability of the RRC layer; for example, the first terminal can send the fifteenth message to the AMF through the 5G MM capability, and then the AMF sends the fifteenth message to the third device.
[0310] Optionally, the first terminal activates the first model, which can be achieved by following the steps [1] to [2]:
[0311] Step [1], the first terminal receives a fourteenth message from the third device;
[0312] Step [2]: The first terminal activates the first model based on the fourteenth message.
[0313] Optionally, the fourteenth message includes at least one of the following:
[0314] a) first identification;
[0315] b) information indicating the version number of the first model, such as version ID;
[0316] c) At least one parameter to be activated.
[0317] In an embodiment of the present application, the third device can ensure that the model that needs to be activated by the third device is consistent with the model executed by the first terminal based on the first identifier or information indicating the version number of the first model, thereby avoiding inconsistent understanding of the first model by the first terminal and the third device.
[0318] The third device needs to determine whether the first terminal meets the triggering conditions of the model and whether the first terminal supports the parameters to be activated, where the triggering conditions of the model may be, for example, usable conditions of the model, including time conditions, location conditions, channel conditions, scene conditions, speed conditions, accessed cell range or cell ID list, etc.;
[0319] When the third device determines the usable conditions and parameters to be activated for the first terminal to support the model, the third device sends the fourteenth message to the first terminal. Accordingly, the first terminal activates the first model based on the fourteenth message, thereby enabling the first terminal to correctly execute the first model.
[0320] FIG5 is a flow chart of a model registration method provided in an embodiment of the present application. As shown in FIG5 , the method includes steps 501-502; wherein:
[0321] Step 501: The third device sends a third message to the first device, where the third message includes a first model file corresponding to a first model, and the third message is used to register the first model.
[0322] Step 502: The third device receives a fourth message returned by the first device; the fourth message includes at least one of the following: processing result information, used to indicate the result obtained by processing the first model file; a second model file, which is obtained by converting the first model file.
[0323] Optionally, the third device generates a model or trains a model to obtain the first model. After obtaining the first model, the third device sends a third message to the first device. The third message may be, for example, a registration request or a publish request. The third message is used to register the first model with the first device so that the first device stores the first model.
[0324] Then the third device receives a fourth message returned by the first device, where the fourth message may be, for example, a registration or publishing response message, used to indicate that the first model has been registered with the first device, or that the first device has stored the first model.
[0325] Optionally, the second model file can be correctly executed in a first terminal associated with the second device.
[0326] Optionally, the first device includes at least one of the following: a) Model Store; b) NEF; c) ADRF; d) NRF.
[0327] Optionally, the second device includes a UE Vendor Sever.
[0328] Optionally, the third device includes at least one of the following: a) core network device; b) access network device; c) target server device; d) second terminal.
[0329] In an embodiment of the present application, the third device sends a third message to the first device, so that the first device registers the first model based on the first model file corresponding to the first model included in the third message; then the third device receives the processing result information and the second model file returned by the first device, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. The third device can know whether the first model file can be correctly executed in the second device and the terminal associated with the second device based on the processing result information, and the third device can determine whether the second model file can be correctly executed in the second device and the terminal associated with the second device based on the second model file included in the second message. Therefore, the third device can determine whether the first model can be correctly executed in the terminal associated with the second device based on the processing result information or the second model file, thereby avoiding the problem that the terminal cannot use the model normally.
[0330] Optionally, the third message further includes first information, where the first information includes at least one of the following:
[0331] a) information indicating the version number of the first model;
[0332] b) information indicating at least one parameter corresponding to the first model;
[0333] c) information indicating a usage condition or usage scenario of the first model;
[0334] d) information indicating at least one function supported by the first model;
[0335] e) information indicating a generating device or a training device of the first model;
[0336] f) Information indicating a device using the first model.
[0337] Optionally, the fourth message further includes at least one of the following:
[0338] a) first identification;
[0339] b) the first model file;
[0340] c) first information;
[0341] d) First indication information, used to indicate information about the manufacturer corresponding to the second device.
[0342] Optionally, after receiving the fourth message returned by the first device, the third device further needs to perform the following steps:
[0343] The third device sends a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
[0344] In an embodiment of the present application, the third device sends the first model file or the second model file to the first terminal, so that the first model file can be correctly executed in the terminal, or the second model file can be correctly executed in the terminal, thereby avoiding the problem of terminal incompatibility when using the model.
[0345] Optionally, the fifth message further includes at least one of the following: a first identifier; and first information, where the first information includes at least one of the following:
[0346] a) information indicating the version number of the first model;
[0347] b) information indicating at least one parameter corresponding to the first model;
[0348] c) information indicating a usage condition or usage scenario of the first model;
[0349] d) information indicating at least one function supported by the first model;
[0350] e) information indicating a generating device or a training device of the first model;
[0351] f) Information indicating a device using the first model.
[0352] Optionally, the third device sends the fifth message to the first terminal, which can be implemented by the following method 1 or method 2:
[0353] Mode 1: When the first terminal has been registered with the third device, the third device sends the fifth message to the first terminal.
[0354] For example, the first terminal first registers with the core network device, and then the core network device can directly send the fifth message to the first terminal.
[0355] Method 2: When the third device receives the fifth request from the first terminal, it sends the tenth message to the first terminal, where the fifth request is used to query whether there are any undownloaded or updated model files in the third device, and the tenth message is used to indicate that there are any undownloaded or updated model files in the third device; when the third device receives the sixth request from the first terminal, it sends the fifth message to the first terminal; the sixth request is used to obtain the fifth message.
[0356] Optionally, the fifth request includes second indication information, which is used to indicate manufacturer information or terminal model information corresponding to the first terminal.
[0357] For example, the first terminal periodically requests the core network device whether there is a model file that has not been downloaded or has been updated; when there is a model file that has not been downloaded or has been updated in the core network device, the first terminal is notified; the first terminal obtains the model file from the core network device according to the notification.
[0358] Optionally, the third device sending the fifth message to the first terminal may also be implemented by the following method 3, which specifically includes steps [1] to [2]:
[0359] Step [1], the third device receives the eleventh message from the first terminal;
[0360] Step [2], the third device sends a fifth message to the first terminal based on the eleventh message;
[0361] The eleventh message includes at least one of the following:
[0362] a) second instruction information;
[0363] b) A first model list, including an identifier or at least one version number of at least one model supported by the first terminal.
[0364] In an embodiment of the present application, the third device receives the eleventh message sent from the first terminal. The third device can filter out information associated with the first model supported by the first terminal based on the eleventh message, and carry it in the fifth message to send to the first terminal.
[0365] Optionally, the eleventh message is carried by at least one of the following messages: RRC message; NAS message.
[0366] Optionally, the third device sending the fifth message to the first terminal may also be implemented by the following method 4, which specifically includes steps [1] to [2]:
[0367] Step [1], the third device receives the thirteenth message from the first terminal;
[0368] Step [2]: The third device sends a fifth message to the first terminal based on the thirteenth message; the thirteenth message includes at least one of the following: a first identifier, information indicating the version number of the first model, and second indication information.
[0369] For example, in a scenario where the network side instructs the first terminal to execute the first model, the first terminal compares the model ID list supported by itself with the model ID list supported by the network side. When the first terminal determines that it lacks the first identifier, it sends the thirteenth message to the core network device to request the core network device to send the fifth message to the first terminal.
[0370] Optionally, the second indication information is carried in IMEI.
[0371] Optionally, before the third device receives the thirteenth message from the first terminal, the third device further needs to perform the following steps:
[0372] The third device sends a twelfth message to the first terminal; the twelfth message is used to indicate a second model list; the second model list includes at least one of the following: an identifier and at least one version number of at least one model supported by the third device.
[0373] In an embodiment of the present application, the third device needs to send the second model list it supports to the first terminal, so that the first terminal can compare the first model list it supports with the second model list supported by the third device. When the first terminal determines that it lacks the first identifier in the second model list, it sends a thirteenth message to the third device to request the third device to send a fifth message to the first terminal.
[0374] Optionally, the twelfth message is carried by at least one of the following messages: 1) SIB message; 2) RRC message; 3) NAS message.
[0375] Optionally, when the first terminal obtains the fifth message from the first device or the second device, the first terminal needs to send a first model list supported by the first terminal to the third device, so that the third device sends an activation message of the first model to the first terminal based on the first model list, thereby enabling the first terminal to correctly execute the first model; this can be specifically achieved through the following steps:
[0376] The third device receives a fifteenth message from the first terminal, where the fifteenth message is used to indicate a first model list.
[0377] In an embodiment of the present application, the third device receives a list of first models supported by the first terminal, so that the third device can determine whether the first terminal meets the triggering conditions of the model and whether the first terminal supports the parameters to be activated based on the identifier of at least one model supported by the first terminal or at least one version number; when the third device determines that the first terminal supports the usable conditions and parameters to be activated of the model, it can send an activation indication to the first terminal to activate the first model.
[0378] Optionally, the fifteenth message is carried by at least one of the following messages: RRC message; NAS message.
[0379] Optionally, the first device instructs the first terminal to activate the first model, which can be specifically implemented by the following steps:
[0380] When the first terminal meets the triggering condition of the first model and the first terminal supports at least one parameter to be activated, the third device sends a fourteenth message to the first terminal, where the fourteenth message is used to instruct activation of the first model.
[0381] Optionally, the fourteenth message includes at least one of the following:
[0382] a) first identification;
[0383] b) information indicating the version number of the first model;
[0384] c) At least one parameter to be activated.
[0385] In an embodiment of the present application, the third device can ensure that the model that needs to be activated by the third device is consistent with the model executed by the first terminal based on the first identifier or information indicating the version number of the first model, thereby avoiding inconsistent understanding of the first model by the first terminal and the third device.
[0386] The third device needs to determine whether the first terminal meets the triggering conditions of the model and whether the first terminal supports the parameters to be activated, where the triggering conditions of the model can be, for example, the usable conditions of the model, including time conditions, location conditions, channel conditions, scene conditions, speed conditions, access cell range or cell ID list, etc.
[0387] When the third device determines the usable conditions and parameters to be activated for the first terminal to support the model, the third device sends the fourteenth message to the first terminal. Accordingly, the first terminal activates the first model based on the fourteenth message, thereby enabling the first terminal to correctly execute the first model.
[0388] FIG6 is a schematic diagram of signaling interaction of a model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG6 , the method includes the following steps 601-606, where step 605 includes steps 605a, 605b, and 605c:
[0389] Step 601: The third device sends a third message to the first device. The third message includes a first model file corresponding to the first model. The third message is used to register the first model.
[0390] Optionally, the first device includes at least one of the following: a model store; a NEF; and an ADRF. The third device includes at least one of the following: a core network device; an access network device; a target server device; and a second terminal.
[0391] Step 602: The first device sends a first message to at least one second device, where the first message includes a first model file.
[0392] Optionally, the second device includes a UE manufacturer server.
[0393] Step 603: The second device returns a second message to the first device; the second message includes at least one of the following: processing result information, used to indicate the result obtained by processing the first model file; and a second model file, which is obtained by converting the first model file.
[0394] Step 604: The third device receives the fourth message returned by the first device; the fourth message includes at least one of the following: processing result information; and the second model file.
[0395] Step 605a: The first terminal obtains a fifth message from the third device;
[0396] Step 605b: The first terminal obtains a fifth message from the first device;
[0397] Step 605c: The first terminal obtains a fifth message from the second device;
[0398] It should be noted that the first terminal may execute at least one of steps 605a, 605b, and 605c.
[0399] Optionally, the fifth message includes the first model file or the second model file corresponding to the first model.
[0400] Step 606: The first terminal executes the first model file or the second model file.
[0401] In an embodiment of the present application, a third device sends a third message to a first device, so that the first device can register the first model. Then, the first device sends a first message to different second devices, requesting each second device to test the first model separately, and then receives a second message from each second device. Based on the processing result information or the second model file included in the second message, it can be determined that the first model file can be correctly executed in the terminal, or that the second model file can be correctly executed in the terminal. The first terminal obtains the first model file or the second model file and executes it, avoiding the problem of incompatibility of the first terminal when using the model.
[0402] FIG7 is a second schematic diagram of signaling interaction of the model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG7 , the method includes steps 701-711, wherein step 701 includes steps 701a and 701b, step 702 includes steps 702a and 702b, and step 707 includes steps 707a and 707b. FIG7 shows an implementation scheme in which the UE manufacturer server sends the model file to the UE, wherein:
[0403] Step 701a: The 3GPP network side performs model training.
[0404] Step 701b: The third-party server performs model training.
[0405] It should be noted that, in practice, only step 701a or only step 701b may be performed, and this application does not limit this. Optionally, the model may be an AI model or an ML model.
[0406] Step 702a: The 3GPP network sends a model registration or release message to the model store, where the model registration or release message is used to register or release the trained model to the model store.
[0407] Step 702b: The third-party server sends a model registration or publishing message to the model store, where the model registration or publishing message is used to register or publish the trained model to the model store.
[0408] It should be noted that step 702a corresponds to step 701a, and step 702b corresponds to step 701b, that is, when the 3GPP network side executes step 701a, the 3GPP network side executes step 702a; and when the third-party server executes step 701b, the third-party server executes step 702b.
[0409] Optionally, a model registration or publishing message contains at least one of the following information:
[0410] Model file (open format) and model configuration information, where the model configuration information includes at least one of the following:
[0411] a) The version number corresponding to the model, b) The parameter list corresponding to the model, c) The conditions under which the model can be used, d) The functions supported by the model, e) The producer of the model, f) The consumer of the model.
[0412] Step 703: The model store stores the model file and the model configuration information, and assigns a globally unique model identifier to the model.
[0413] Step 704: The model store sends a model checking request to each UE manufacturer server.
[0414] Specifically, the model checking request includes at least one of the following information:
[0415] a) Model identification, b) Model file, c) Model version number, d) Model parameter list, e) Model usable conditions, f) Model supported functions, g) Model producer, h) Model consumer.
[0416] Step 705: Each UE manufacturer's server checks the model internally.
[0417] It should be noted that checking the model refers to testing each parameter that can be used in the model or performing corresponding conversions.
[0418] Here, conversion means converting from a model mode (such as an encoding format of a model file) that is not supported by the UE manufacturer to a model mode that is supported by the UE manufacturer.
[0419] Step 706: Each UE manufacturer server sends a model check response to the model store, including conversion results and test results.
[0420] It should be noted that the test results include test success results and test failure results.
[0421] Step 707a: The model store sends a model registration or release confirmation message to the 3GPP network side, including the registration or release result.
[0422] Step 707b: The model store sends a model registration or publishing confirmation (ack) message to the third-party server, including the registration result or publishing result.
[0423] It should be noted that, in practice, the model store may perform both step 707a and step 707b, or may only perform step 707a, or may only perform step 707b. In the case where the model store performs both step 707a and step 707b, step 707a and step 707b may be performed simultaneously, or step 707a may be performed first and then step 707b, or step 707b may be performed first and then step 707a. This application does not limit this.
[0424] Optionally, the registration result or publishing result includes a model identifier and a version number corresponding to the model.
[0425] Step 708: The UE manufacturer server sends the model identifier, model file, model version number, model parameter list, model usable conditions, model supported functions, model producer, and model consumer to the UE through the application layer, such as OTT.
[0426] Optionally, the model file includes the model file before conversion by the UE manufacturer server, or the model file after conversion by the UE manufacturer server.
[0427] The specific push method includes: the UE periodically requests the UE manufacturer server whether there is a new model file; when the UE manufacturer server has a new model file, it notifies the UE; and then the UE obtains the new model file from the UE manufacturer server according to the notification.
[0428] Step 709: The UE performs a model capability indication, that is, the UE sends a list of model identifiers and version numbers supported by the UE to the 3GPP network side.
[0429] Specifically, the UE may send the model identifier and version number list supported by the UE in the radio capability of the RRC layer to the 3GPP network side, such as the base station;
[0430] or,
[0431] The UE can use the 5G MM capability to send the model identifier and version number list supported by the UE to the AMF, and the AMF then sends it to the 3GPP network side.
[0432] Step 710: On the 3GPP network side, for example, the access network device / core network device determines that the UE meets the triggering conditions of the model and performs model activation management if the UE supports the parameters to be activated.
[0433] That is, the 3GPP network side sends a message for activating the model to the UE, and the message carries at least one of the following information: a) a model identifier of the model to be activated; b) a parameter to be activated.
[0434] Step 711: The UE activates the model using the parameters to be activated.
[0435] Specifically, the UE executes the model according to the model identifier of the model to be activated and the parameters to be activated to obtain a result.
[0436] FIG8 is a third schematic diagram of signaling interaction of the model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG8 , the method includes steps 801-811, wherein step 801 includes steps 801a and 801b, step 802 includes steps 802a and 802b, and step 807 includes steps 807a and 807b. FIG8 shows an implementation scheme of the model store sending the model file to the UE, wherein:
[0437] Step 801a: The 3GPP network side performs model training.
[0438] Step 801b: The third-party server performs model training.
[0439] It should be noted that, in practice, only step 801a or only step 801b may be performed, and this application does not limit this. Optionally, the model may be an AI model, an ML model, etc.
[0440] Step 802a: The 3GPP network sends a model registration or release message to the model store, where the model registration or release message is used to register or release the trained model to the model store.
[0441] Step 802b: The third-party server sends a model registration or publishing message to the model store, where the model registration or publishing message is used to register or publish the trained model to the model store.
[0442] It should be noted that step 802a corresponds to step 801a, and step 802b corresponds to step 801b, that is, when step 801a is executed on the 3GPP network side, the 3GPP network side executes step 802a; and when step 801b is executed on the third-party server, the third-party server executes step 802b.
[0443] Optionally, a model registration or publishing message contains at least one of the following information:
[0444] Model file (open format) and model configuration information, where the model configuration information includes at least one of the following:
[0445] a) The version number corresponding to the model, b) The parameter list corresponding to the model, c) The conditions under which the model can be used, d) The functions supported by the model, e) The producer of the model, f) The consumer of the model.
[0446] Step 803: The model store stores the model file and the model configuration information, and assigns a globally unique model identifier to the model.
[0447] Step 804: The model store sends a model check request to each UE manufacturer server.
[0448] Specifically, the model checking request includes at least one of the following information:
[0449] a) Model identification, b) Model file, c) Model version number, d) Model parameter list, e) Model usable conditions, f) Model supported functions, g) Model producer, h) Model consumer.
[0450] Step 805: Each UE manufacturer's server checks the model internally.
[0451] It should be noted that checking the model refers to testing each parameter that can be used in the model or performing corresponding conversions.
[0452] Here, conversion means converting from a model mode (such as an encoding format of a model file) that is not supported by the UE manufacturer to a model mode that is supported by the UE manufacturer.
[0453] Step 806: Each UE manufacturer server sends a model check response to the model store, including conversion results and test results.
[0454] It should be noted that the test results include test success results and test failure results.
[0455] Step 807a: The model store sends a model registration or release confirmation message to the 3GPP network side, including the registration or release result.
[0456] Step 807b: The model store sends a model registration or publishing confirmation message to the third-party server, including the registration or publishing result.
[0457] It should be noted that, in practice, the model store may perform both step 807a and step 807b, or may only perform step 807a, or may only perform step 807b. When the model store performs both step 807a and step 807b, step 807a and step 807b may be performed simultaneously, or step 807a may be performed first and then step 807b, or step 807b may be performed first and then step 807a. This application does not limit this.
[0458] Optionally, the registration or publishing result includes a model identifier, a version number corresponding to the model, and a model file.
[0459] Step 808: The model store sends the model identifier, model file, model version number, model parameter list, model usable conditions, model supported functions, model producer, and model consumer to the UE.
[0460] Optionally, the model file includes the model file before conversion by the UE manufacturer server, or the model file after conversion by the UE manufacturer server.
[0461] Specifically, the model store may actively push the model to the UE, or the UE may periodically obtain the model from the model store.
[0462] The UE periodically obtains from the model store, which can be specifically achieved in the following manner: the UE periodically requests the model store whether there is a new model file; when the model store has a new model file, the UE is notified; and then the UE obtains the new model file from the model store according to the notification.
[0463] Step 809: The UE sends the model identifiers and version number list supported by the UE to the 3GPP network side.
[0464] Specifically, the UE may send the model identifier and version number list supported by the UE to the 3GPP network side in the radio capability of the RRC layer, such as the base station;
[0465] or,
[0466] The UE can use the 5G MM capability to send the model identifier and version number list supported by the UE to the AMF, and the AMF then sends it to the 3GPP network side.
[0467] Step 810: On the 3GPP network side, for example, the access network device / core network device determines that the UE meets the triggering conditions of the model and performs model activation management if the UE supports the parameters to be activated.
[0468] That is, the 3GPP network sends a message to the UE for activating the model, and the message carries at least one of the following:
[0469] a) Model identifier of the model to be activated; b) Parameters to be activated.
[0470] Step 811: The UE activates the model using the parameters to be activated.
[0471] Specifically, the UE executes the model according to the model identifier of the model to be activated and the parameters to be activated to obtain a result.
[0472] FIG9 is a fourth schematic diagram of signaling interaction of the model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG9 , the method includes steps 901-911, wherein step 901 includes steps 901a and 901b, step 902 includes steps 902a and 902b, and step 907 includes steps 907a and 907b. FIG9 shows an implementation scheme in which the 3GPP network side sends the model file to the UE, wherein:
[0473] Step 901a: The 3GPP network side performs model training.
[0474] Step 901b: The third-party server performs model training.
[0475] It should be noted that, in practice, only step 901a or only step 901b may be performed, and this application does not limit this. Optionally, the model may be an AI model, an ML model, etc.
[0476] Step 902a: The 3GPP network sends a model registration or release message to the model store, where the model registration or release message is used to register or release the trained model to the model store.
[0477] Step 902b: The third-party server sends a model registration or publishing message to the model store, where the model registration or publishing message is used to register or publish the trained model to the model store.
[0478] It should be noted that step 902a corresponds to step 901a, and step 902b corresponds to step 901b, that is, when step 901a is executed on the 3GPP network side, the 3GPP network side executes step 902a; and when step 901b is executed on the third-party server, the third-party server executes step 902b.
[0479] A model registration or publishing message contains at least one of the following information:
[0480] Model file (open format) and model configuration information, where the model configuration information includes at least one of the following:
[0481] a) The version number corresponding to the model, b) The parameter list corresponding to the model, c) The conditions under which the model can be used, d) The functions supported by the model, e) The producer of the model, f) The consumer of the model.
[0482] Step 903: The model store stores the model file and the model configuration information, and assigns a globally unique model identifier to the model.
[0483] Step 904: The model store sends a model check request to each UE manufacturer server.
[0484] Specifically, the model checking request includes at least one of the following information:
[0485] a) Model identification, b) Model file, c) Model version number, d) Model parameter list, e) Model usable conditions, f) Model supported functions, g) Model producer, h) Model consumer.
[0486] Step 905: Each UE manufacturer's server internally checks the model.
[0487] It should be noted that checking the model refers to testing each parameter that can be used in the model or performing corresponding conversions.
[0488] Here, conversion means converting from a model mode (such as an encoding format of a model file) that is not supported by the UE manufacturer to a model mode that is supported by the UE manufacturer.
[0489] Step 906: Each UE manufacturer server sends a model check response to the model store, including conversion results and test results.
[0490] It should be noted that the test results include test success results and test failure results.
[0491] Step 907a: The model store sends a model registration or release confirmation message to the 3GPP network side, including the registration or release result.
[0492] Step 907b: The model store sends a model registration or publishing confirmation message to the third-party server, including the registration or publishing result.
[0493] It should be noted that, in practice, the model store may perform both step 907a and step 907b, or may only perform step 907a, or may only perform step 907b. When the model store performs both step 907a and step 907b, step 907a and step 907b may be performed simultaneously, or step 907a may be performed first and then step 907b, or step 907b may be performed first and then step 907a. This application does not limit this.
[0494] Optionally, the registration or publishing result includes a model identifier, a version number corresponding to the model, and a model file.
[0495] Step 908: The UE sends at least one of the following information to the 3GPP network:
[0496] a) Model capability information supported by the UE, b) UE manufacturer server identifier, where the UE manufacturer server identifier can be carried in the IMEI.
[0497] It should be noted that the UE may send the capability information supported by the UE and the version number corresponding to the model in the radio capability of the RRC layer to the 3GPP network side, such as the base station;
[0498] or,
[0499] The UE can use the 5G MM capability to send the capability information supported by the UE and the version number corresponding to the model to the AMF, and the AMF will then send it to the 3GPP network side.
[0500] Step 909: The 3GPP network side performs model transmission, that is, the 3GPP network side sends the model identifier and the model file to the UE.
[0501] Optionally, the model file includes the model file before conversion by the UE manufacturer server, or the model file after conversion by the UE manufacturer server.
[0502] Step 910: On the 3GPP network side, for example, the access network device / core network device determines that the UE meets the triggering conditions of the model and the UE supports the parameters to be activated, and performs model activation management.
[0503] That is, the 3GPP network sends a message to the UE for activating the model, and the message carries at least one of the following:
[0504] a) Model identifier of the model to be activated; b) Parameters to be activated.
[0505] Step 911: The UE activates the model using the parameters to be activated.
[0506] Specifically, the UE executes the model according to the model identifier of the model to be activated and the parameters to be activated to obtain a result.
[0507] FIG10 is a fifth signaling interaction diagram of the model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG10 , the method includes steps 1001-1013, wherein step 1001 includes steps 1001a and 1001b, step 1002 includes steps 1002a and 1002b, and step 1007 includes steps 1007a and 1007b. FIG10 shows another implementation scheme in which the UE manufacturer server sends the model file to the UE, wherein:
[0508] Step 1001a: The 3GPP network side performs model training.
[0509] Step 1001b: The third-party server performs model training.
[0510] It should be noted that, in practice, only step 1001a or only step 1001b may be performed, and this application does not limit this. Optionally, the model may be an AI model, an ML model, etc.
[0511] Step 1002a: The 3GPP network sends a model registration or release message to the model store, where the model registration or release message is used to register or release the trained model to the model store.
[0512] Step 1002b: The third-party server sends a model registration or publishing message to the model store, where the model registration or publishing message is used to register or publish the trained model to the model store.
[0513] It should be noted that step 1002a corresponds to step 1001a, and step 1002b corresponds to step 1001b, that is, when step 1001a is executed on the 3GPP network side, the 3GPP network side executes step 1002a; and when step 1001b is executed on the third-party server, the third-party server executes step 1002b.
[0514] Optionally, a model registration or publishing message contains at least one of the following information:
[0515] Model file (open format) and model configuration information, where the model configuration information includes at least one of the following:
[0516] a) The version number corresponding to the model, b) The parameter list corresponding to the model, c) The conditions under which the model can be used, d) The functions supported by the model, e) The producer of the model, f) The consumer of the model.
[0517] Step 1003: The model store stores the model file and the model configuration information, and assigns a globally unique model identifier to the model.
[0518] Step 1004: The model store sends a model check request to each UE manufacturer server.
[0519] Specifically, the model checking request includes at least one of the following information:
[0520] a) Model identification, b) Model file, c) Model version number, d) Model parameter list, e) Model usable conditions, f) Model supported functions, g) Model producer, h) Model consumer.
[0521] Step 1005: Each UE manufacturer's server checks the model internally.
[0522] It should be noted that checking the model refers to testing each parameter that can be used in the model or performing corresponding conversions.
[0523] Here, conversion means converting from a model mode (such as an encoding format of a model file) that is not supported by the UE manufacturer to a model mode that is supported by the UE manufacturer.
[0524] Step 1006: Each UE manufacturer server sends a model check response to the model store, including conversion results and test results.
[0525] It should be noted that the test results include test success results and test failure results.
[0526] Step 1007a: The model store sends a model registration or release confirmation message to the 3GPP network side, including the registration or release result.
[0527] Step 1007b: The model store sends a model registration or publishing confirmation message to the third-party server, including the registration or publishing result.
[0528] It should be noted that, in practice, the model store may perform both step 1007a and step 1007b, or may only perform step 1007a, or may only perform step 1007b. When the model store performs both step 1007a and step 1007b, step 1007a and step 1007b may be performed simultaneously, or step 1007a may be performed first and then step 1007b, or step 1007b may be performed first and then step 1007a. This application does not limit this.
[0529] Optionally, the registration or publishing result includes a model identifier and a version number corresponding to the model.
[0530] Step 1008: The 3GPP network side sends a list of model identifiers supported by the 3GPP network side to the UE.
[0531] Specifically, the 3GPP network side may send the model identification list it supports to the UE via the SIB; or,
[0532] The 3GPP network side may send a list of model identifiers supported by the 3GPP network side to the UE via an RRC message or a NAS message.
[0533] Step 1009: The UE compares the model identifier list it supports with the model identifier list supported by the network side. If there is a model identifier that the UE lacks, the UE actively sends the model identifier that the UE lacks to the UE manufacturer server to which it belongs, to request downloading information such as the model file corresponding to the model identifier.
[0534] Step 1010: The UE manufacturer server sends the model identifier, model file, model version number, model parameter list, model usable conditions, model supported functions, model producer, and model consumer to the UE through the application layer, such as OTT (Over The Top).
[0535] Optionally, the model file includes the model file before conversion by the UE manufacturer server, or the model file after conversion by the UE manufacturer server.
[0536] The specific push method includes: the UE periodically requests the UE manufacturer server whether there is a new model file; when the UE manufacturer server has a new model file, it notifies the UE; and then the UE obtains the new model file from the UE manufacturer server according to the notification.
[0537] Step 1011: The UE sends a list of model identifiers and version numbers supported by the UE to the 3GPP network side.
[0538] Specifically, the UE may send the model identifier and version number list supported by the UE to the 3GPP network side in the radio capability of the RRC layer, such as the base station;
[0539] or,
[0540] The UE can use the 5G MM capability to send the model identifier and version number list supported by the UE to the AMF, and the AMF then sends it to the 3GPP network side.
[0541] Step 1012: On the 3GPP network side, for example, the access network device / core network device determines that the UE meets the triggering conditions of the model and the UE supports the parameters to be activated, and performs model activation management.
[0542] That is, the 3GPP network sends a message to the UE for activating the model, and the message carries at least one of the following:
[0543] a) Model identifier of the model to be activated; b) Parameters to be activated.
[0544] Step 1013: The UE activates the model using the parameters to be activated.
[0545] Specifically, the UE executes the model according to the model identifier of the model to be activated and the parameters to be activated to obtain a result.
[0546] FIG11 is a sixth signaling interaction diagram of the model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG11 , the method includes steps 1101-1113, wherein step 1101 includes steps 1101a and 1101b, step 1102 includes steps 1102a and 1102b, and step 1107 includes steps 1107a and 1107b. FIG11 shows another implementation scheme in which the model store sends the model file to the UE, wherein:
[0547] Step 1101a: The 3GPP network side performs model training.
[0548] Step 1101b: The third-party server performs model training.
[0549] It should be noted that, in practice, only step 1101a or only step 1101b may be performed, and this application does not limit this. Optionally, the model may be an AI model, an ML model, etc.
[0550] Step 1102a: The 3GPP network sends a model registration or release message to the model store, where the model registration or release message is used to register or release the trained model to the model store.
[0551] Step 1102b: The third-party server sends a model registration or publishing message to the model store, where the model registration or publishing message is used to register or publish the trained model to the model store.
[0552] It should be noted that step 1102a corresponds to step 1101a, and step 1102b corresponds to step 1101b, that is, when step 1101a is executed on the 3GPP network side, the 3GPP network side executes step 1102a; and when step 1101b is executed on the third-party server, the third-party server executes step 1102b.
[0553] Optionally, a model registration or publishing message contains at least one of the following information:
[0554] Model file (open format) and model configuration information, where the model configuration information includes at least one of the following:
[0555] a) The version number corresponding to the model, b) The parameter list corresponding to the model, c) The conditions under which the model can be used, d) The functions supported by the model, e) The producer of the model, f) The consumer of the model.
[0556] Step 1103: The model store stores the model file and the model configuration information, and assigns a globally unique model identifier to the model.
[0557] Step 1104: The model store sends a model checking request to each UE manufacturer server.
[0558] Specifically, the model checking request includes at least one of the following information:
[0559] a) Model identification, b) Model file, c) Model version number, d) Model parameter list, e) Model usable conditions, f) Model supported functions, g) Model producer, h) Model consumer.
[0560] Step 1105: Each UE manufacturer's server internally checks the model.
[0561] It should be noted that checking the model refers to testing each parameter that can be used in the model or performing corresponding conversions.
[0562] Here, conversion means converting from a model mode (such as an encoding format of a model file) that is not supported by the UE manufacturer to a model mode that is supported by the UE manufacturer.
[0563] Step 1106: Each UE manufacturer server sends a model check response to the model store, including conversion results and test results.
[0564] It should be noted that the test results include test success results and test failure results.
[0565] Step 1107a: The model store sends a model registration or release confirmation message to the 3GPP network side, including the registration or release result.
[0566] Step 1107b: The model store sends a model registration or publishing confirmation message to the third-party server, including the registration or publishing result.
[0567] It should be noted that, in practice, the model store may perform both step 1107a and step 1107b, or may only perform step 1107a, or may only perform step 1107b. When the model store performs both step 1107a and step 1107b, step 1107a and step 1107b may be performed simultaneously, or step 1107a may be performed first and then step 1107b, or step 1107b may be performed first and then step 1107a. This application does not limit this.
[0568] Optionally, the registration or publishing result includes a model identifier and a version number corresponding to the model.
[0569] Step 1108: The 3GPP network side sends a list of model identifiers supported by the 3GPP network side to the UE.
[0570] Specifically, the 3GPP network side may send the model identification list it supports to the UE via the SIB; or,
[0571] The 3GPP network side may send a list of model identifiers supported by the 3GPP network side to the UE via an RRC message or a NAS message.
[0572] Step 1109: The UE compares the model identification list it supports with the model identification list supported by the network side. If there is a model identification that the UE lacks, the UE sends at least one of the following information to the model store: model capability information supported by the UE and the version number corresponding to the model, where the version number corresponding to the model can be carried in the IMEI.
[0573] Step 1110: The model store sends the model identifier, model file, version number corresponding to the model, parameter list corresponding to the model, conditions under which the model can be used, functions supported by the model, producer of the model, and consumer of the model to the UE.
[0574] Optionally, the model file includes the model file before conversion by the UE manufacturer server, or the model file after conversion by the UE manufacturer server.
[0575] Specifically, the model store may actively push the model to the UE, or the UE may periodically obtain the model from the model store.
[0576] The UE periodically obtains from the model store, which can be specifically achieved in the following manner: the UE periodically requests the model store whether there is a new model file; when the model store has a new model file, the UE is notified; and then the UE obtains the new model file from the model store according to the notification.
[0577] Step 1111: The UE sends a list of model identifiers and version numbers supported by the UE to the 3GPP network side.
[0578] Specifically, the UE may send the model identifier and version number list supported by the UE to the 3GPP network side in the radio capability of the RRC layer, such as the base station;
[0579] or,
[0580] The UE can use the 5G MM capability to send the model identifier and version number list supported by the UE to the AMF, and the AMF then sends it to the 3GPP network side.
[0581] Step 1112: On the 3GPP network side, for example, the access network device / core network device determines that the UE meets the triggering conditions of the model and performs model activation management if the UE supports the parameters to be activated.
[0582] That is, the 3GPP network sends a message to the UE for activating the model, and the message carries at least one of the following:
[0583] a) Model identifier of the model to be activated; b) Parameters to be activated.
[0584] Step 1113: The UE activates the model using the parameters to be activated.
[0585] Specifically, the UE executes the model according to the model identifier of the model to be activated and the parameters to be activated to obtain a result.
[0586] FIG12 is a seventh signaling interaction diagram of the model processing, registration, and execution method provided in an embodiment of the present application. As shown in FIG12 , the method includes steps 1201-1212, wherein step 1201 includes steps 1201a and 1201b, step 1202 includes steps 1202a and 1202b, and step 1207 includes steps 1207a and 1207b. FIG12 shows another implementation scheme in which the 3GPP network side sends the model file to the UE, wherein:
[0587] Step 1201a: The 3GPP network side performs model training.
[0588] Step 1201b: The third-party server performs model training.
[0589] It should be noted that, in practice, only step 1201a or only step 1201b may be performed, and this application does not limit this. Optionally, the model may be an AI model, an ML model, or the like.
[0590] Step 1202a: The 3GPP network sends a model registration or release message to the model store, where the model registration or release message is used to register or release the trained model to the model store.
[0591] Step 1202b: The third-party server sends a model registration or publishing message to the model store, where the model registration or publishing message is used to register or publish the trained model to the model store.
[0592] It should be noted that step 1202a corresponds to step 1201a, and step 1202b corresponds to step 1201b, that is, when step 1201a is executed on the 3GPP network side, the 3GPP network side executes step 1202a; and when step 1201b is executed on the third-party server, the third-party server executes step 1202b.
[0593] Optionally, a model registration or publishing message contains at least one of the following information:
[0594] Model file (open format) and model configuration information, where the model configuration information includes at least one of the following:
[0595] a) The version number corresponding to the model, b) The parameter list corresponding to the model, c) The conditions under which the model can be used, d) The functions supported by the model, e) The producer of the model, f) The consumer of the model.
[0596] Step 1203: The model store stores the model file and the model configuration information, and assigns a globally unique model identifier to the model.
[0597] Step 1204: The model store sends a model check request to each UE manufacturer server.
[0598] Specifically, the model checking request includes at least one of the following information:
[0599] a) Model identification, b) Model file, c) Model version number, d) Model parameter list, e) Model usable conditions, f) Model supported functions, g) Model producer, h) Model consumer.
[0600] Step 1205: Each UE manufacturer's server checks the model internally.
[0601] It should be noted that checking the model refers to testing each parameter that can be used in the model or performing corresponding conversions.
[0602] Here, conversion means converting from a model mode (such as an encoding format of a model file) that is not supported by the UE manufacturer to a model mode that is supported by the UE manufacturer.
[0603] Step 1206: Each UE manufacturer server sends a model check response to the model store, including conversion results and test results.
[0604] It should be noted that the test results include test success results and test failure results.
[0605] Step 1207a: The model store sends a model registration or release confirmation message to the 3GPP network side, including the registration or release result.
[0606] Step 1207b: The model store sends a model registration or publishing confirmation message to the third-party server, including the registration or publishing result.
[0607] It should be noted that, in practice, the model store may perform both step 1207a and step 1207b, or may only perform step 1207a, or may only perform step 1207b. When the model store performs both step 1207a and step 1207b, step 1207a and step 1207b may be performed simultaneously, or step 1207a may be performed first and then step 1207b, or step 1207b may be performed first and then step 1207a. This application does not limit this.
[0608] In step 1207a and step 1207b, the registration or publishing result includes the model identifier, the version number corresponding to the model, and the model file.
[0609] Step 1208: The 3GPP network side sends a list of model identifiers supported by the 3GPP network side to the UE.
[0610] Specifically, the 3GPP network side may send the model identification list it supports to the UE via the SIB; or,
[0611] The 3GPP network side may send a list of model identifiers supported by the 3GPP network side to the UE via an RRC message or a NAS message.
[0612] Step 1209: The UE compares its supported model identifier list with the model identifier list supported by the network. If there is a model identifier that the UE lacks, the UE sends at least one of the following information to the 3GPP network:
[0613] a) Model capability information supported by the UE, b) Version number corresponding to the model, where the version number corresponding to the model can be carried in the IMEI.
[0614] It should be noted that the UE may send the capability information supported by the UE and the version number corresponding to the model in the radio capability of the RRC layer to the 3GPP network side, such as the base station;
[0615] or,
[0616] The UE can use the 5G MM capability to send the capability information supported by the UE and the version number corresponding to the model to the AMF, and the AMF will then send it to the 3GPP network side.
[0617] Step 1210: The 3GPP network side performs model transmission, that is, the 3GPP network side sends the model identifier and the model file to the UE.
[0618] Optionally, the model file includes the model file before conversion by the UE manufacturer server, or the model file after conversion by the UE manufacturer server.
[0619] Step 1211: On the 3GPP network side, for example, the access network device / core network device determines that the UE meets the triggering conditions of the model and performs model activation management if the UE supports the parameters to be activated.
[0620] That is, the 3GPP network sends a message to the UE for activating the model, and the message carries at least one of the following:
[0621] a) Model identifier of the model to be activated; b) Parameters to be activated.
[0622] Step 1212: The UE activates the model using the parameters to be activated.
[0623] Specifically, the UE executes the model according to the model identifier of the model to be activated and the parameters to be activated to obtain a result.
[0624] The model processing method provided in the embodiment of the present application can be executed by a model processing device. In the embodiment of the present application, the model processing device provided in the embodiment of the present application is described by taking the execution of the model processing method by the model processing device as an example.
[0625] FIG13 is a schematic diagram of a structure of a model processing device according to an embodiment of the present application. As shown in FIG13 , the model processing device 1300 is applied to a first device and includes:
[0626] A first sending module 1301 is configured to send a first message to at least one second device, where the first message includes a first model file corresponding to a first model;
[0627] A first receiving module 1302 is configured to receive a second message returned by the second device;
[0628] The second message includes at least one of the following:
[0629] Processing result information, used to indicate the result obtained by processing the first model file;
[0630] A second model file is obtained by converting the first model file.
[0631] In the model processing device provided in the embodiment of the present application, a first message is sent to at least one second device, the first message includes a first model file corresponding to the first model, and then a second message returned by the second device is received, the second message includes processing result information and at least one item in the second model file, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. Based on the processing result information, it can be known whether the first model file can be correctly executed in the second device and the terminal associated with the second device, and based on the second message including the second model file, it can be determined that the second model file can be correctly executed in the second device and the terminal associated with the second device. Therefore, based on the received processing result information or the second model file, it can be determined whether the first model can be correctly executed in the terminal associated with the second device, thereby avoiding the problem of the terminal being unable to use the model normally.
[0632] Optionally, the device further comprises:
[0633] An allocating module is configured to allocate a first identifier to the first model, where the first identifier is used to identify the first model.
[0634] Optionally, the first message also includes the first identifier.
[0635] Optionally, the first message further includes first information, where the first information includes at least one of the following:
[0636] Information indicating the version number of the first model;
[0637] Information indicating at least one parameter corresponding to the first model;
[0638] Information indicating a usage condition or usage scenario of the first model;
[0639] Information indicating at least one function supported by the first model;
[0640] Information for indicating a generating device or a training device of the first model;
[0641] Information indicating a device using the first model.
[0642] Optionally, the device further comprises:
[0643] The fourth receiving module is used to receive a third message sent by a third device, where the third message includes the first model file and is used to register the first model.
[0644] Optionally, the device further comprises:
[0645] A third sending module, configured to send a fourth message to the third device;
[0646] The fourth message includes at least one of the following:
[0647] the processing result information;
[0648] the second model file.
[0649] Optionally, the fourth message further includes at least one of the following:
[0650] a first identifier, used to identify the first model;
[0651] the first model file;
[0652] First information, the first information including at least one of the following: information indicating a version number of the first model; information indicating at least one parameter corresponding to the first model; information indicating a usage condition or usage scenario of the first model; information indicating at least one function supported by the first model; information indicating a generation device or training device for the first model; information indicating a device using the first model;
[0653] The first indication information is used to indicate information of a manufacturer corresponding to the second device.
[0654] Optionally, the device further comprises:
[0655] The fourth sending module is configured to send a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
[0656] Optionally, the fifth message further includes at least one of the following: a first identifier, used to identify the first model;
[0657] First information, the first information includes at least one of the following: information for indicating the version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating the usage conditions or usage scenarios of the first model; information for indicating at least one function supported by the first model; information for indicating the generation device or training device of the first model; information for indicating the usage device of the first model.
[0658] Optionally, the fourth sending module is further configured to perform at least one of the following:
[0659] In a case where the first terminal has been registered with the first device, sending the fifth message to the first terminal;
[0660] Upon receiving a first request from the first terminal, sending a sixth message to the first terminal, where the first request is used to query whether there is a model file that has not been downloaded or has been updated, and the sixth message is used to indicate that there is a model file that has not been downloaded or has been updated in the first device; upon receiving a second request from the first terminal, sending the fifth message to the first terminal, where the second request is used to obtain the fifth message;
[0661] Upon receiving the seventh message from the first terminal, the fifth message is sent to the first terminal; the seventh message includes at least one of the following: a first identifier for identifying the first model; information for indicating the version number of the first model; and second indication information for indicating information on the manufacturer or terminal model corresponding to the first terminal.
[0662] Optionally, the first request includes second indication information.
[0663] Optionally, the second indication information is carried in the International Mobile Equipment Identity (IMEI) of the first terminal.
[0664] Optionally, the second model file can be correctly executed in a first terminal associated with the second device.
[0665] Optionally, the first device includes at least one of the following: a model store Model Store; a network exposure function NEF; an analysis database function ADRF; and a network storage function NRF.
[0666] Optionally, the third device includes at least one of the following: a core network device; an access network device; a target server device; or a second terminal.
[0667] FIG14 is a second structural diagram of a model processing device provided in an embodiment of the present application. As shown in FIG14 , the model processing device 1400 is applied to a second device and includes:
[0668] The second receiving module 1401 is configured to receive a first message sent by a first device, where the first message includes a first model file corresponding to a first model;
[0669] A returning module 1402 is configured to return a second message to the first device;
[0670] The second message includes at least one of the following:
[0671] Processing result information, used to indicate the result obtained by processing the first model file;
[0672] A second model file is obtained by converting the first model file.
[0673] In the model processing device provided in the embodiment of the present application, a first message sent by a first device is received, the first message includes a first model file corresponding to the first model, and then a second message is returned to the first device, the second message includes processing result information and at least one item in the second model file, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. The first device can know whether the first model file can be correctly executed in the second device and the terminal associated with the second device based on the processing result information, and the first device can determine whether the second model file can be correctly executed in the second device and the terminal associated with the second device based on the second message including the second model file. Therefore, returning the processing result information or the second model file to the first device can enable the first device to determine whether the first model can be correctly executed in the terminal associated with the second device, thereby avoiding the problem of the terminal being unable to use the model normally.
[0674] Optionally, the first message further includes a first identifier, which is allocated to the first model by the first device and is used to identify the first model.
[0675] Optionally, the first message further includes first information, where the first information includes at least one of the following:
[0676] Information indicating the version number of the first model;
[0677] Information indicating at least one parameter corresponding to the first model;
[0678] Information indicating a usage condition or usage scenario of the first model;
[0679] Information indicating at least one function supported by the first model;
[0680] Information for indicating a generating device or a training device of the first model;
[0681] Information indicating a device using the first model.
[0682] Optionally, the device further comprises:
[0683] The fifth sending module is configured to send a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
[0684] Optionally, the fifth message also includes at least one of the following: a first identifier for identifying the first model; first information, the first information including at least one of the following: information for indicating the version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating the usage conditions or usage scenarios of the first model; information for indicating at least one function supported by the first model; information for indicating the generation device or training device of the first model; information for indicating the usage device of the first model.
[0685] Optionally, the fifth sending module is further configured to perform at least one of the following:
[0686] In a case where the first terminal has been registered with the second device, sending the fifth message to the first terminal;
[0687] Upon receiving a third request from the first terminal, sending an eighth message to the first terminal, wherein the third request is used to query whether there is an undownloaded or updated model file, and the eighth message is used to indicate that an undownloaded or updated model file exists in the second device; upon receiving a fourth request from the first terminal, sending the fifth message to the first terminal; the fourth request is used to obtain the fifth message;
[0688] Upon receiving the ninth message from the first terminal, the fifth message is sent to the first terminal; the ninth message includes at least one of the following: a first identifier for identifying the first model; information for indicating the version number of the first model; and second indication information for indicating the manufacturer information or terminal model information corresponding to the first terminal.
[0689] Optionally, the third request includes second indication information.
[0690] Optionally, the second indication information is carried in the International Mobile Equipment Identity (IMEI) of the first terminal.
[0691] Optionally, the second model file can be correctly executed in a first terminal associated with the second device.
[0692] Optionally, the first device includes at least one of the following: a model store Model Store; a network exposure function NEF; an analysis database function ADRF; and a network storage function NRF.
[0693] The model execution method provided in the embodiment of the present application can be executed by a model execution device. In the embodiment of the present application, the model execution device is used as an example to illustrate the model execution method.
[0694] FIG15 is a schematic diagram of the structure of a model execution device provided in an embodiment of the present application. As shown in FIG15 , the model execution device 1500 is applied to a first terminal and includes:
[0695] An acquisition module 1501 is configured to acquire a fifth message, where the fifth message includes a first model file or a second model file corresponding to the first model, where the second model file is obtained by converting the first model file.
[0696] The execution module 1502 is configured to execute the first model file or the second model file.
[0697] In the model execution device provided in the embodiment of the present application, by obtaining the fifth message, the fifth message includes the first model file or the second model file corresponding to the first model, wherein the second model file is obtained by converting the first model file, and executing the first model file or the second model file can ensure that the first model file or the second model file can be correctly executed in the first terminal, thereby avoiding the problem that the first terminal cannot use the model normally.
[0698] Optionally, the fifth message further includes at least one of the following:
[0699] a first identifier, used to identify the first model;
[0700] First information, the first information including at least one of the following:
[0701] Information indicating the version number of the first model;
[0702] Information indicating at least one parameter corresponding to the first model;
[0703] Information indicating a usage condition or usage scenario of the first model;
[0704] Information indicating at least one function supported by the first model;
[0705] Information for indicating a generating device or a training device of the first model;
[0706] Information indicating a device using the first model.
[0707] Optionally, the acquisition module 1501 is further configured to perform at least one of the following:
[0708] receiving the fifth message from the first device;
[0709] Send a first request to a first device; receive a sixth message from the first device, wherein the first request is used to query whether there is any model file that has not been downloaded or has been updated, and the sixth message is used to indicate that there is a model file that has not been downloaded or has been updated in the first device; send a second request to the first device, wherein the second request is used to obtain the fifth message; receive the fifth message from the first device.
[0710] Optionally, the first request includes second indication information, which is used to indicate information about a manufacturer or terminal model corresponding to the first terminal.
[0711] Optionally, the acquisition module 1501 is further configured to perform at least one of the following:
[0712] receiving the fifth message from the second device;
[0713] Sending a third request to the second device; receiving an eighth message from the second device, wherein the third request is used to query whether there is any model file that has not been downloaded or has been updated, and the eighth message is used to indicate that there is a model file that has not been downloaded or has been updated in the second device; sending a fourth request to the second device, wherein the fourth request is used to obtain the fifth message; receiving the fifth message from the second device.
[0714] Optionally, the third request includes second indication information.
[0715] Optionally, the acquisition module 1501 is further configured to perform at least one of the following:
[0716] receiving the fifth message from the third device;
[0717] Send a fifth request to a third device; receive a tenth message from the third device, wherein the fifth request is used to query whether there are any undownloaded or updated model files in the third device, and the tenth message is used to indicate that there are any undownloaded or updated model files in the third device; send a sixth request to the third device, wherein the sixth request is used to obtain the fifth message; receive the fifth message from the third device.
[0718] Optionally, the fifth request includes second indication information.
[0719] Optionally, the acquisition module 1501 is further configured to:
[0720] sending an eleventh message to the third device;
[0721] receiving the fifth message from the third device;
[0722] The eleventh message includes at least one of the following:
[0723] Second indication information, used to indicate manufacturer information or terminal model information corresponding to the first terminal;
[0724] The first model list includes an identifier or at least one version number of at least one model supported by the first terminal.
[0725] Optionally, the eleventh message is carried by at least one of the following messages: a radio resource control RRC message; a non-access stratum NAS message.
[0726] Optionally, the acquisition module 1501 is further configured to:
[0727] receiving a twelfth message from a third device; wherein the twelfth message is used to indicate a second model list; wherein the second model list includes at least one of the following: an identifier and at least one version number of at least one model supported by the third device;
[0728] Based on the first model list and the second model list, the fifth message is obtained when it is determined that the first identifier or information indicating the version number of the first model exists in the second model list but does not exist in the first model list; the first identifier is used to identify the first model.
[0729] Optionally, the acquisition module 1501 is further configured to perform at least one of the following:
[0730] Based on the first identifier or the information used to indicate the version number of the first model, sending a seventh message to the first device, the seventh message including at least one of the first identifier, the information used to indicate the version number of the first model, and second indication information, where the second indication information is used to indicate manufacturer information or terminal model information corresponding to the first terminal; and receiving the fifth message from the first device;
[0731] Based on the first identifier or the information used to indicate the version number of the first model, sending a ninth message to the second device, the ninth message including at least one of the first identifier, the information used to indicate the version number of the first model, and second indication information, where the second indication information is used to indicate manufacturer information or terminal model information corresponding to the first terminal; and receiving the fifth message from the second device;
[0732] Based on the first identifier or the information used to indicate the version number of the first model, a thirteenth message is sent to the third device, the thirteenth message including the first identifier, the information used to indicate the version number of the first model and at least one of the second indication information, the second indication information being used to indicate the manufacturer information or terminal model information corresponding to the first terminal; and the fifth message is received from the third device.
[0733] Optionally, the second indication information is carried in the International Mobile Equipment Identity (IMEI) of the first terminal.
[0734] Optionally, the twelfth message is carried by at least one of the following messages: a system information block SIB message; a radio resource control RRC message; or a non-access stratum NAS message.
[0735] Optionally, the device further comprises:
[0736] a fifth receiving module, configured to receive a fourteenth message from a third device;
[0737] An activation module is used to activate the first model based on the fourteenth message.
[0738] Optionally, the fourteenth message includes at least one of the following:
[0739] a first identifier, used to identify the first model;
[0740] Information indicating the version number of the first model;
[0741] At least one parameter to be activated.
[0742] Optionally, the device further comprises:
[0743] The sixth sending module is used to send a fifteenth message to the third device, where the fifteenth message is used to indicate the first model list.
[0744] Optionally, the fifteenth message is carried by at least one of the following messages: RRC message; NAS message.
[0745] Optionally, the second model file can be correctly executed in a first terminal associated with the second device.
[0746] Optionally, the first device includes at least one of the following: a model store Model Store; a network exposure function NEF; an analysis database function ADRF; and a network storage function NRF.
[0747] Optionally, the third device includes at least one of the following: a core network device; an access network device; a target server device; or a second terminal.
[0748] The model registration method provided in the embodiment of the present application can be executed by a model registration device. In the embodiment of the present application, the model registration device is used as an example to illustrate the model registration method provided in the embodiment of the present application.
[0749] FIG16 is a schematic diagram of the structure of a model registration device provided in an embodiment of the present application. As shown in FIG16 , the model registration device 1600 is applied to a third device and includes:
[0750] A second sending module 1601 is configured to send a third message to the first device, where the third message includes a first model file corresponding to the first model, and the third message is used to register the first model;
[0751] The third receiving module 1602 is configured to receive a fourth message returned by the first device;
[0752] The fourth message includes at least one of the following:
[0753] Processing result information, used to indicate the result obtained by processing the first model file;
[0754] A second model file is obtained by converting the first model file.
[0755] In the model registration device provided in the embodiment of the present application, by sending a third message to the first device, the first device registers the first model based on the first model file corresponding to the first model included in the third message; then the processing result information and the second model file returned by the first device are received, wherein the processing result information is used to indicate the result obtained by processing the first model file, and the second model file is obtained by converting the first model file. Based on the processing result information, it can be known whether the first model file can be correctly executed in the second device and the terminal associated with the second device, and based on the second model file included in the second message, it can be determined whether the second model file can be correctly executed in the second device and the terminal associated with the second device. Therefore, based on the processing result information or the second model file, it can be determined whether the first model can be correctly executed in the terminal associated with the second device, thereby avoiding the problem of the terminal being unable to use the model normally.
[0756] Optionally, the third message further includes first information, where the first information includes at least one of the following:
[0757] Information indicating the version number of the first model;
[0758] Information indicating at least one parameter corresponding to the first model;
[0759] Information indicating a usage condition or usage scenario of the first model;
[0760] Information indicating at least one function supported by the first model;
[0761] Information for indicating a generating device or a training device of the first model;
[0762] Information indicating a device using the first model.
[0763] Optionally, the fourth message also includes at least one of the following: a first identifier for identifying the first model; the first model file; first indication information for indicating information about the manufacturer corresponding to the second device; first information, the first information including at least one of the following: information for indicating the version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating the usage conditions or usage scenarios of the first model; information for indicating at least one function supported by the first model; information for indicating the generation device or training device of the first model; information for indicating the usage device of the first model.
[0764] Optionally, the device further comprises:
[0765] The seventh sending module is configured to send a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
[0766] Optionally, the fifth message also includes at least one of the following: a first identifier for identifying the first model; first information, the first information including at least one of the following: information for indicating the version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating the usage conditions or usage scenarios of the first model; information for indicating at least one function supported by the first model; information for indicating the generation device or training device of the first model; information for indicating the usage device of the first model.
[0767] Optionally, the seventh sending module is further configured to perform at least one of the following:
[0768] In a case where the first terminal has been registered with the third device, sending the fifth message to the first terminal;
[0769] In the event of receiving the fifth request from the first terminal, a tenth message is sent to the first terminal, the fifth request is used to query whether there is any undownloaded or updated model file in the third device, and the tenth message is used to indicate that there is any undownloaded or updated model file in the third device; in the event of receiving the sixth request from the first terminal, the fifth message is sent to the first terminal; the sixth request is used to obtain the fifth message.
[0770] Optionally, the fifth request includes second indication information, which is used to indicate manufacturer information or terminal model information corresponding to the first terminal.
[0771] Optionally, the seventh sending module is further configured to:
[0772] receiving an eleventh message from the first terminal;
[0773] Based on the eleventh message, sending a fifth message to the first terminal;
[0774] The eleventh message includes at least one of the following:
[0775] Second indication information, used to indicate manufacturer information or terminal model information corresponding to the first terminal;
[0776] The first model list includes an identifier or at least one version number of at least one model supported by the first terminal.
[0777] Optionally, the eleventh message is carried by at least one of the following messages: a radio resource control RRC message; a non-access stratum NAS message.
[0778] Optionally, the seventh sending module is further configured to:
[0779] receiving a thirteenth message from the first terminal;
[0780] Based on the thirteenth message, a fifth message is sent to the first terminal; the thirteenth message includes at least one of the following: a first identifier for identifying the first model, information for indicating the version number of the first model, and second indication information, the second indication information is used to indicate the manufacturer information or terminal model information corresponding to the first terminal.
[0781] Optionally, the second indication information is carried in the International Mobile Equipment Identity (IMEI).
[0782] Optionally, the device further comprises:
[0783] An eighth sending module is configured to send a twelfth message to the first terminal; the twelfth message is used to indicate a second model list; the second model list includes at least one of the following: an identifier and at least one version number of at least one model supported by the third device.
[0784] Optionally, the twelfth message is carried by at least one of the following messages: a system information block SIB message; a radio resource control RRC message; or a non-access stratum NAS message.
[0785] Optionally, the device further comprises:
[0786] The ninth sending module is used to send a fourteenth message to the first terminal when the first terminal meets the triggering condition of the first model and the first terminal supports at least one parameter to be activated, wherein the fourteenth message is used to indicate the activation of the first model.
[0787] Optionally, the fourteenth message includes at least one of the following:
[0788] a first identifier, used to identify the first model;
[0789] Information indicating the version number of the first model;
[0790] At least one parameter to be activated.
[0791] Optionally, the device further comprises:
[0792] The sixth receiving module is configured to receive a fifteenth message from the first terminal, where the fifteenth message is used to indicate a first model list.
[0793] Optionally, the fifteenth message is carried by at least one of the following messages: RRC message; NAS message.
[0794] Optionally, the second model file can be correctly executed in a first terminal associated with the second device.
[0795] Optionally, the first device includes at least one of the following: a model store Model Store; a network exposure function NEF; an analysis database function ADRF; and a network storage function NRF.
[0796] Optionally, the third device includes at least one of the following: a core network device; an access network device; a target server device; or a second terminal.
[0797] The model processing device, model execution device, and model registration device in the embodiments of the present application can be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or other device other than a terminal. For example, the terminal can include but is not limited to the types of terminals 11 listed above, and other devices can be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiments of the present application.
[0798] The model processing device provided in the embodiment of the present application can implement the various processes implemented by the method embodiments of Figures 2 to 3, and achieve the same technical effects; the model execution device provided in the embodiment of the present application can implement the various processes implemented by the method embodiment of Figure 4, and achieve the same technical effects; the model registration device provided in the embodiment of the present application can implement the various processes implemented by the method embodiment of Figure 5, and achieve the same technical effects; to avoid repetition, they will not be repeated here.
[0799] Figure 17 is a structural diagram of a communication device provided in an embodiment of the present application. As shown in Figure 17, an embodiment of the present application further provides a communication device 1700, including a processor 1701 and a memory 1702. The memory 1702 stores a program or instruction that can be run on the processor 1701. For example, when the communication device 1700 is a first device, the program or instruction is executed by the processor 1701 to implement the various steps of the model processing method embodiment shown in Figure 2 above, and can achieve the same technical effect. When the communication device 1700 is a second device, the program or instruction is executed by the processor 1701 to implement the various steps of the model processing method embodiment shown in Figure 3 above, and can achieve the same technical effect. When the communication device 1700 is a first terminal, the program or instruction is executed by the processor 1701 to implement the various steps of the model execution method embodiment shown in Figure 4 above, and can achieve the same technical effect. When the communication device 1700 is a third device, the program or instruction is executed by the processor 1701 to implement the various steps of the model registration method embodiment shown in Figure 5 above, and can achieve the same technical effect. To avoid repetition, I will not go into details here.
[0800] The present application also provides a first terminal embodiment, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps in the model execution method embodiment shown in FIG4 . This first terminal embodiment corresponds to the first terminal-side method embodiment described above, and each implementation process and implementation method of the above method embodiment can be applied to this first terminal embodiment and can achieve the same technical effect. Specifically, FIG18 is a schematic diagram of the structure of the first terminal provided in the present application embodiment.
[0801] The first terminal 1800 includes but is not limited to: a radio frequency unit 1801, a network module 1802, an audio output unit 1803, an input unit 1804, a sensor 1805, a display unit 1806, a user input unit 1807, an interface unit 1808, a memory 1809 and at least some of the components of the processor 1810.
[0802] Those skilled in the art will appreciate that the first terminal 1800 may further include a power source (such as a battery) for powering various components. The power source may be logically connected to the processor 1810 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The first terminal structure shown in FIG18 does not limit the first terminal. The first terminal may include more or fewer components than shown, or may combine certain components, or have different component arrangements, which will not be described in detail here.
[0803] It should be understood that in an embodiment of the present application, the input unit 1804 may include a graphics processing unit (GPU) 18041 and a microphone 18042, and the graphics processor 18041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 1806 may include a display panel 18061, and the display panel 18061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1807 includes a touch panel 18071 and at least one of other input devices 18072. The touch panel 18071 is also called a touch screen. The touch panel 18071 may include two parts: a touch detection device and a touch controller. Other input devices 18072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.
[0804] In the embodiment of the present application, after receiving downlink data from a network-side device, the RF unit 1801 may transmit the data to the processor 1810 for processing. Furthermore, the RF unit 1801 may send uplink data to the network-side device. Typically, the RF unit 1801 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like.
[0805] The memory 1809 can be used to store software programs or instructions and various data. The memory 1809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 1809 may include a volatile memory or a non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM). The memory 1809 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.
[0806] Processor 1810 may include one or more processing units. Optionally, processor 1810 integrates an application processor and a modem processor. The application processor primarily handles operations related to the operating system, user interface, and application programs, while the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into processor 1810.
[0807] It can be understood that the implementation process of each implementation method mentioned in the embodiments of the present application can refer to the relevant description of the model execution method embodiment shown in Figure 4 above, and achieve the same or corresponding technical effects. To avoid repetition, it will not be repeated here.
[0808] The present application also provides a first device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the model processing method embodiment shown in FIG2 . This first device embodiment corresponds to the first device method embodiment described above, and each implementation process and implementation method of the above method embodiment are applicable to this first device embodiment and can achieve the same technical effects.
[0809] The present application also provides a second device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the model processing method embodiment shown in FIG3 . This second device embodiment corresponds to the aforementioned second device method embodiment, and each implementation process and implementation method of the aforementioned method embodiment are applicable to this second device embodiment and can achieve the same technical effects.
[0810] The present application also provides a third device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the model processing method embodiment shown in FIG5 . This third device embodiment corresponds to the aforementioned third device method embodiment, and each implementation process and implementation method of the aforementioned method embodiment are applicable to this third device embodiment and can achieve the same technical effects.
[0811] Figure 19 is one of the structural diagrams of the third device provided in an embodiment of the present application. As shown in Figure 19, the third device 1900 includes: an antenna 1901, a radio frequency device 1902, a baseband device 1903, a processor 1904, and a memory 1905. Antenna 1901 is connected to radio frequency device 1902. In the uplink direction, radio frequency device 1902 receives information through antenna 1901 and sends the received information to baseband device 1903 for processing. In the downlink direction, baseband device 1903 processes the information to be sent and sends it to radio frequency device 1902. Radio frequency device 1902 processes the received information and sends it out through antenna 1901.
[0812] The method executed by the third device in the above embodiment may be implemented in the baseband device 1903 , which includes a baseband processor.
[0813] The baseband device 1903 may, for example, include at least one baseband board, on which multiple chips are arranged, as shown in Figure 19, one of the chips is, for example, a baseband processor, which is connected to the memory 1905 through a bus interface to call the program in the memory 1905 and execute the network device operations shown in the above method embodiment.
[0814] The third device may further include a network interface 1906 , which is, for example, a Common Public Radio Interface (CPRI).
[0815] Specifically, the third device 1900 of an embodiment of the present invention also includes: instructions or programs stored in the memory 1905 and executable on the processor 1904. The processor 1904 calls the instructions or programs in the memory 1905 to execute the methods shown in Figures 2, 3, and 5 above, and achieve the same technical effects. To avoid repetition, they will not be elaborated here.
[0816] FIG20 is a second structural diagram of the third device provided in an embodiment of the present application. As shown in FIG20 , the third device 2000 includes a processor 2001, a network interface 2002, and a memory 2003. The network interface 2002 is, for example, a common public radio interface (CPRI).
[0817] Specifically, the third device 2000 of the embodiment of the present invention also includes: instructions or programs stored in the memory 2003 and executable on the processor 2001. The processor 2001 calls the instructions or programs in the memory 2003 to execute the methods shown in Figures 2, 3 and 5 above, and achieve the same technical effects. To avoid repetition, they will not be elaborated here.
[0818] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the method embodiments of Figures 2 to 5 above are implemented, and the same technical effects can be achieved. To avoid repetition, they will not be repeated here.
[0819] The processor is the processor in the terminal described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. In some examples, the readable storage medium may be a non-transitory readable storage medium.
[0820] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the method embodiments of Figures 2 to 5 above, and can achieve the same technical effects. To avoid repetition, they will not be repeated here.
[0821] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
[0822] An embodiment of the present application further provides a computer program / program product, which is stored in a storage medium. The computer program / program product is executed by at least one processor to implement the various processes of the method embodiments of Figures 2 to 5 above, and can achieve the same technical effects. To avoid repetition, they are not described here.
[0823] An embodiment of the present application also provides a model processing, execution and registration system, including: a first device, a second device, a first terminal and a third device, wherein the first device can be used to execute the steps of the model processing method shown in Figure 2 above, the second device can be used to execute the steps of the model processing method shown in Figure 3 above, the first terminal can be used to execute the steps of the model execution method shown in Figure 4 above, and the third device can be used to execute the steps of the model registration method shown in Figure 5 above.
[0824] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted or combined. In addition, the features described with reference to certain examples may be combined in other examples.
[0825] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of a computer software product plus a necessary general-purpose hardware platform, or of course, by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes a number of instructions for enabling a terminal or network-side device to execute the methods described in each embodiment of the present application.
[0826] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms of implementation methods without departing from the purpose of this application and the scope of protection of the claims. These implementation methods are all within the protection of this application.
Claims
1. A model processing method, comprising: The first device sends a first message to at least one second device, where the first message includes a first model file corresponding to the first model; The first device receives a second message returned by the second device; The second message includes at least one of the following: Processing result information, used to indicate the result obtained by processing the first model file; A second model file is obtained by converting the first model file.
2. The model processing method according to claim 1, further comprising: The first device allocates a first identifier to the first model, where the first identifier is used to identify the first model.
3. The model processing method according to claim 2, wherein: The first message also includes the first identifier.
4. The model processing method according to any one of claims 1 to 3, wherein: The first message also includes first information, and the first information includes at least one of the following: Information used to indicate a version number of the first model; Information used to indicate at least one parameter corresponding to the first model; Information indicating a usage condition or usage scenario of the first model; Information indicating at least one function supported by the first model; Information for indicating a generating device or a training device of the first model; Information indicating a device using the first model.
5. The model processing method according to any one of claims 1 to 4, further comprising: The first device receives a third message sent by a third device, where the third message includes the first model file and is used to register the first model.
6. The model processing method according to claim 5, further comprising: The first device sends a fourth message to the third device; The fourth message includes at least one of the following: The processing result information; the second model file.
7. The model processing method according to claim 6, wherein: The fourth message further includes at least one of the following: A first identifier, used to identify the first model; the first model file; First information, the first information including at least one of the following: information indicating a version number of the first model; information indicating at least one parameter corresponding to the first model; information indicating a use condition or use scenario of the first model; Information indicating at least one function supported by the first model; Information for indicating a generating device or a training device of the first model; information indicating a device using the first model; The first indication information is used to indicate information of a manufacturer corresponding to the second device.
8. The model processing method according to any one of claims 1 to 7, further comprising: The first device sends a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
9. The model processing method according to claim 8, wherein: The fifth message further includes at least one of the following: A first identifier, used to identify the first model; First information, the first information including at least one of the following: information indicating a version number of the first model; information indicating at least one parameter corresponding to the first model; information indicating a use condition or use scenario of the first model; Information indicating at least one function supported by the first model; Information used to indicate a device for generating or training the first model; information used to indicate a device for using the first model.
10. The model processing method according to claim 8 or 9, wherein: The first device sending a fifth message to the first terminal includes at least one of the following: The first device sends the fifth message to the first terminal when the first terminal has been registered with the first device; The first device sends a sixth message to the first terminal when receiving a first request from the first terminal, wherein the first request is used to query whether there is a model file that has not been downloaded or has been updated, and the sixth message is used to indicate that there is a model file that has not been downloaded or has been updated in the first device; the first device sends the fifth message to the first terminal when receiving a second request from the first terminal; The second request is used to obtain the fifth message; When receiving the seventh message from the first terminal, the first device sends the fifth message to the first terminal; The seventh message includes at least one of the following: a first identifier used to identify the first model; information used to indicate a version number of the first model; The second indication information is used to indicate information about a manufacturer or a terminal model corresponding to the first terminal.
11. A model processing method, comprising: The second device receives a first message sent by the first device, where the first message includes a first model file corresponding to the first model; The second device returns a second message to the first device; The second message includes at least one of the following: Processing result information, used to indicate the result obtained by processing the first model file; A second model file is obtained by converting the first model file.
12. The model processing method according to claim 11, wherein: The first message also includes a first identifier, which is allocated by the first device to the first model, and is used to identify the first model.
13. The model processing method according to claim 11 or 12, wherein: The first message also includes first information, and the first information includes at least one of the following: Information used to indicate a version number of the first model; Information used to indicate at least one parameter corresponding to the first model; Information indicating a usage condition or usage scenario of the first model; Information indicating at least one function supported by the first model; Information for indicating a generating device or a training device of the first model; Information indicating a device using the first model.
14. The model processing method according to any one of claims 11 to 13, further comprising: The second device sends a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
15. The model processing method according to claim 14, wherein: The fifth message further includes at least one of the following: a first identifier for identifying the first model; first information, the first information including at least one of the following: information for indicating a version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating a use condition or use scenario of the first model; Information indicating at least one function supported by the first model; Information used to indicate a device for generating or training the first model; information used to indicate a device for using the first model.
16. The model processing method according to claim 14 or 15, wherein: The second device sending a fifth message to the first terminal includes at least one of the following: The second device sends the fifth message to the first terminal when the first terminal has been registered with the second device; The second device sends an eighth message to the first terminal when receiving a third request from the first terminal, wherein the third request is used to query whether there is a model file that has not been downloaded or has been updated, and the eighth message is used to indicate that there is a model file that has not been downloaded or has been updated in the second device; the second device sends the fifth message to the first terminal when receiving a fourth request from the first terminal; The fourth request is used to obtain the fifth message; When receiving the ninth message from the first terminal, the second device sends the fifth message to the first terminal; The ninth message includes at least one of the following: a first identifier used to identify the first model; information used to indicate a version number of the first model; The second indication information is used to indicate the manufacturer information or terminal model information corresponding to the first terminal.
17. A model execution method, comprising: The first terminal obtains a fifth message, wherein the fifth message includes a first model file or a second model file corresponding to the first model, wherein the second model file is obtained by converting the first model file get; The first terminal executes the first model file or the second model file.
18. The model execution method according to claim 17, wherein: The fifth message further includes at least one of the following: A first identifier, used to identify the first model; First information, the first information includes at least one of the following: Information used to indicate a version number of the first model; Information used to indicate at least one parameter corresponding to the first model; Information indicating a usage condition or usage scenario of the first model; Information indicating at least one function supported by the first model; Information for indicating a generating device or a training device of the first model; Information indicating a device using the first model.
19. The model execution method according to claim 17 or 18, wherein: The first terminal acquiring the fifth message includes at least one of the following: The first terminal receives the fifth message from the first device; The first terminal sends a first request to the first device; the first terminal receives a sixth message from the first device, the first request is used to query whether there are any model files that have not been downloaded or have been updated, and the sixth message is used to indicate that there are model files that have not been downloaded or have been updated in the first device; the first terminal sends a second request to the first device, and the second request is used to obtain the fifth message; the first terminal receives the fifth message from the first device.
20. The model execution method according to claim 17 or 18, wherein: The first terminal acquiring the fifth message includes at least one of the following: The first terminal receives the fifth message from the second device; The first terminal sends a third request to the second device; the first terminal receives an eighth message from the second device, the third request is used to query whether there is a model file that has not been downloaded or has been updated, and the eighth message is used to indicate that there is a model file that has not been downloaded or has been updated in the second device; the first terminal sends a fourth request to the second device, and the fourth request is used to obtain the fifth message; the first terminal receives the fifth message from the second device interest.
21. The model execution method according to claim 17 or 18, wherein: The first terminal acquiring the fifth message includes at least one of the following: The first terminal receives the fifth message from the third device; The first terminal sends a fifth request to the third device; the first terminal receives a tenth message from the third device, the fifth request is used to query whether there are any undownloaded or updated model files in the third device, and the tenth message is used to indicate that there are any undownloaded or updated model files in the third device; the first terminal sends a sixth request to the third device, and the sixth request is used to obtain the fifth message; the first terminal receives the fifth message from the third device.
22. The model execution method according to claim 17 or 18, wherein: The first terminal acquiring the fifth message includes: The first terminal sends an eleventh message to the third device; The first terminal receives the fifth message from the third device; The eleventh message includes at least one of the following: The second indication information is used to indicate the manufacturer information or terminal model information corresponding to the first terminal; The first model list includes an identifier or at least one version number of at least one model supported by the first terminal.
23. The model execution method according to claim 17 or 18, wherein: The first terminal acquiring the fifth message includes: The first terminal receives a twelfth message from the third device; the twelfth message is used to indicate a second model list; the second model list includes at least one of the following: an identifier and at least one version number of at least one model supported by the third device; Based on the first model list and the second model list, the first terminal obtains the fifth message when it is determined that the first identifier or the information indicating the version number of the first model exists in the second model list but does not exist in the first model list; the first identifier is used to identify the first model.
24. The model execution method according to claim 23, wherein: The obtaining of the fifth message includes at least one of the following: The first terminal sends a seventh message to the first device based on the first identifier or the information used to indicate the version number of the first model, where the seventh message includes at least one of the first identifier, the information used to indicate the version number of the first model, and second indication information, where the second indication information is used to indicate manufacturer information or terminal model information corresponding to the first terminal; the first terminal receives the fifth message from the first device; The first terminal sends a ninth message to the second device based on the first identifier or the information used to indicate the version number of the first model, wherein the ninth message includes at least one of the first identifier, the information used to indicate the version number of the first model, and second indication information, wherein the second indication information is used to indicate manufacturer information or terminal model information corresponding to the first terminal; the first terminal receives the fifth message from the second device; The first terminal sends a thirteenth message to the third device based on the first identifier or the information used to indicate the version number of the first model, the thirteenth message including the first identifier, the information used to indicate the version number of the first model and at least one of the second indication information, the second indication information being used to indicate the manufacturer information or terminal model information corresponding to the first terminal; the first terminal receives the fifth message from the third device.
25. The model execution method according to claim 23, wherein: The twelfth message is carried by at least one of the following messages: a system information block SIB message; a radio resource control RRC message; a non-access layer NAS message.
26. The model execution method according to any one of claims 17 to 25, further comprising: The first terminal receives a fourteenth message from the third device; The first terminal activates the first model based on the fourteenth message.
27. The model execution method according to claim 26, wherein: The fourteenth message includes at least one of the following: A first identifier, used to identify the first model; Information used to indicate a version number of the first model; At least one parameter to be activated.
28. The model execution method according to claim 26 or 27, further comprising: The first terminal sends a fifteenth message to the third device, where the fifteenth message is used to indicate a first model list.
29. The model execution method according to claim 28, wherein: The fifteenth message is carried by at least one of the following messages: RRC message; NAS message.
30. A model registration method, comprising: The third device sends a third message to the first device, where the third message includes a first model file corresponding to the first model, and the third message is used to register the first model; The third device receives a fourth message returned by the first device; The fourth message includes at least one of the following: Processing result information, used to indicate the result obtained by processing the first model file; A second model file is obtained by converting the first model file.
31. The model registration method according to claim 30, wherein: The third message also includes first information, where the first information includes at least one of the following: Information used to indicate a version number of the first model; Information used to indicate at least one parameter corresponding to the first model; Information indicating a usage condition or usage scenario of the first model; Information indicating at least one function supported by the first model; Information for indicating a generating device or a training device of the first model; Information indicating a device using the first model.
32. The model registration method according to claim 30 or 31, wherein: The fourth message further includes at least one of the following: a first identifier for identifying the first model; the first model file; first indication information for indicating information about a manufacturer corresponding to the second device; first information, the first information including at least one of the following: information for indicating a version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating a use condition or use scenario of the first model; Information indicating at least one function supported by the first model; Information used to indicate a device for generating or training the first model; information used to indicate a device for using the first model.
33. The model registration method according to any one of claims 30 to 32, further comprising: The third device sends a fifth message to the first terminal, where the fifth message includes the first model file or the second model file.
34. The model registration method according to claim 33, wherein: The fifth message further includes at least one of the following: a first identifier for identifying the first model; first information, the first information including at least one of the following: information for indicating a version number of the first model; information for indicating at least one parameter corresponding to the first model; information for indicating a use condition or use scenario of the first model; Information indicating at least one function supported by the first model; Information used to indicate a device for generating or training the first model; information used to indicate a device for using the first model.
35. The model registration method according to claim 33 or 34, wherein: The third device sending a fifth message to the first terminal includes at least one of the following: The third device sends the fifth message to the first terminal when the first terminal has been registered with the third device; When the third device receives the fifth request from the first terminal, the third device sends the tenth message to the first terminal, where the fifth request is used to query whether there are any undownloaded or updated model files in the third device, and the tenth message is used to indicate that there are any undownloaded or updated model files in the third device; when the third device receives the sixth request from the first terminal, the third device sends the fifth message to the first terminal; the sixth request is used to obtain the fifth message.
36. The model registration method according to claim 33 or 34, wherein: The third device sending a fifth message to the first terminal includes: The third device receives an eleventh message from the first terminal; The third device sends a fifth message to the first terminal based on the eleventh message; The eleventh message includes at least one of the following: The second indication information is used to indicate the manufacturer information or terminal model information corresponding to the first terminal; The first model list includes an identifier or at least one version number of at least one model supported by the first terminal.
37. The model registration method according to claim 33 or 34, wherein: The third device sending a fifth message to the first terminal includes: The third device receives a thirteenth message from the first terminal; The third device sends a fifth message to the first terminal based on the thirteenth message; the thirteenth message includes at least one of the following: a first identifier for identifying the first model, information for indicating the version number of the first model, and second indication information, the second indication information is used to indicate the manufacturer information or terminal model information corresponding to the first terminal.
38. The model registration method according to claim 33 or 37, further comprising: The third device sends a twelfth message to the first terminal; The twelfth message is used to indicate the second model list; The second model list includes at least one of the following items: an identifier of at least one model supported by the third device and at least one version number.
39. The model registration method according to claim 38, wherein: The twelfth message is carried by at least one of the following messages: a system information block SIB message; a radio resource control RRC message; a non-access layer NAS message.
40. The model registration method according to any one of claims 30 to 39, further comprising: When the first terminal meets the triggering condition of the first model and the first terminal supports at least one parameter to be activated, the third device sends a fourteenth message to the first terminal, where the fourteenth message is used to indicate activation of the first model.
41. The model registration method according to claim 40, wherein: The fourteenth message includes at least one of the following: A first identifier, used to identify the first model; Information used to indicate a version number of the first model; At least one parameter to be activated.
42. The model registration method according to claim 39 or 40, further comprising: The third device receives a fifteenth message from the first terminal, where the fifteenth message is used to indicate a first model list.
43. The model registration method according to claim 42, wherein: The fifteenth message is carried by at least one of the following messages: RRC message; NAS message.
44. A model processing device comprising: A first sending module, configured to send a first message to at least one second device, wherein the first message includes a first model file corresponding to the first model; A first receiving module, configured to receive a second message returned by the second device; The second message includes at least one of the following: Processing result information, used to indicate the result obtained by processing the first model file; A second model file is obtained by converting the first model file.
45. A model processing device comprising: A second receiving module, configured to receive a first message sent by a first device, wherein the first message includes a first model file corresponding to the first model; A returning module, configured to return a second message to the first device; The second message includes at least one of the following: Processing result information, used to indicate the result obtained by processing the first model file; A second model file is obtained by converting the first model file.
46. A model execution device, comprising: an acquisition module, configured to acquire a fifth message, wherein the fifth message includes a first model file or a second model file corresponding to the first model, wherein the second model file is obtained by converting the first model file; An execution module is used to execute the first model file or the second model file.
47. A model registration device, comprising: A second sending module, configured to send a third message to the first device, wherein the third message includes a first model file corresponding to the first model, and the third message is used to register the first model; A third receiving module, configured to receive a fourth message returned by the first device; The fourth message includes at least one of the following: Processing result information, used to indicate the result obtained by processing the first model file; A second model file is obtained by converting the first model file.
48. A communication device, comprising a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, it implements the model processing method as described in any one of claims 1 to 10, or implements the model processing method as described in any one of claims 11 to 16, or implements the model execution method as described in any one of claims 17 to 29, or implements the steps of the model registration method as described in any one of claims 30 to 43.
49. A readable storage medium storing a program or instruction, wherein the program or instruction, when executed by a processor, implements the model processing method as described in any one of claims 1 to 10, or implements the model processing method as described in any one of claims 11 to 16, or implements the model execution method as described in any one of claims 17 to 29, or implements the steps of the model registration method as described in any one of claims 30 to 43.
Citation Information
Patent Citations
Model management method, electronic equipment and medium
CN112036558A
Artificial intelligence model publishing method and device
CN113849179A
Network model conversion method, device and system, storage medium and electronic device
CN114298284A
Model end side deployment method and device, equipment and storage medium
CN115756516A
Optimized deployment method and device for computer vision deep learning model
CN116048542A
