Model identification method and apparatus, terminal, and network side device

Abstract

Disclosed in the present application are a model identification method and apparatus, a terminal, and a network side device, the method comprising: a first terminal sends to a network side device a first model identification request, the first model identification request being used for requesting the network side device to perform a first model identification of a target artificial intelligence (AI) unit; and the first terminal receives a first model identifier configured by the network side device for the target AI unit, the first model identifier satisfying at least one of the following: a global identifier which is globally unique; a partial global identifier which is partially globally unique; and being used by a second terminal to send to the network side device a second model identification request.

Description

Model identification method, device, terminal and network side equipment

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application number 202311018872.6 filed in China on August 11, 2023, 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 recognition method, device, terminal and network-side equipment. Background Art

[0004] Artificial intelligence (AI) is currently gaining widespread application across various fields. Integrating AI into wireless communication networks to significantly improve technical indicators such as throughput, latency, and user capacity is a key task for future wireless communication networks. AI units or AI models can be implemented using neural networks, whose parameters can be optimized using gradient optimization algorithms.

[0005] In the online model registration of related technologies, when a communication device has a new model, each device needs to register the same new model with the network side, which leads to a large amount of signaling redundancy and wastes air interface resources.

[0006] Summary of the Invention

[0007] The embodiments of the present application provide a model identification method, apparatus, terminal, and network-side device, which can solve the problem that when registering a new AI model online, each communication device needs to request the network-side device to register the same new AI model, resulting in a large amount of signaling redundancy and waste of air interface resources.

[0008] In a first aspect, a model identification method is provided, the method comprising: a first terminal sending a first model identification request to a network device, wherein the first model identification request is used to request the network device to perform first model identification of a target artificial intelligence (AI) unit; and the first terminal receiving a first model identifier configured by the network device for the target AI unit, wherein the first model identifier satisfies at least one of the following:

[0009] A globally unique global identifier;

[0010] Partially globally unique partially global identifier;

[0011] Used by the second terminal to send a second model recognition request to the network side device.

[0012] In a second aspect, a model identification method is provided, which includes: a second terminal obtains a first model identifier obtained by the first terminal by triggering first model identification; the second terminal sends a second model identification request to a network side device, wherein the second model identification request is used to request the network side device to perform second model identification of the target AI unit, and the second model identification request carries the first model identifier; the second terminal obtains the second model identifier configured by the network side device for the target AI unit.

[0013] In a third aspect, a model identification method is provided, which includes: a network side device receives a first model identification request sent by a first terminal, wherein the first model identification request is used to request the network side device to perform a first model identification of a target AI unit; the network side device assigns a first model identifier to the target AI unit by performing the first model identification of the target AI unit, wherein the first model identifier satisfies at least one of the following: a globally unique global identifier; a partially globally unique partial global identifier; used by a second terminal to send a second model identification request to the network side device; and the network side device sends the first model identifier to the first terminal.

[0014] In a fourth aspect, a model identification device is provided, the device comprising: a sending module configured to send a first model identification request to a network-side device, wherein the first model identification request is used to request the network-side device to perform first model identification of a target artificial intelligence (AI) unit; and a receiving module configured to receive a first model identifier configured by the network-side device for the target AI unit, wherein the first model identifier satisfies at least one of the following:

[0015] A globally unique global identifier;

[0016] Partially globally unique partially global identifier;

[0017] Used by the second terminal to send a second model recognition request to the network side device.

[0018] In a fifth aspect, a model identification device is provided, which includes: a first acquisition module, used to obtain a first model identifier obtained by the first terminal by triggering the first model identification; a sending module, used to send a second model identification request to the network side device, wherein the second model identification request is used to request the network side device to perform a second model identification of the target AI unit, and the second model identification request carries the first model identifier; a second acquisition module, used to obtain the second model identifier configured by the network side device for the target AI unit.

[0019] In a sixth aspect, a model identification device is provided, which includes: a receiving module for receiving a first model identification request sent by a first terminal, wherein the first model identification request is used to request the network side device to perform a first model identification of a target AI unit; an allocation module for allocating a first model identifier to the target AI unit by performing the first model identification of the target AI unit, wherein the first model identifier satisfies at least one of the following: a globally unique global identifier; a partially globally unique partial global identifier; a second terminal for sending a second model identification request to the network side device; and a sending module for sending the first model identifier to the first terminal.

[0020] In the seventh aspect, a terminal is provided, which includes 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 first aspect are implemented, or the steps of the method described in the second aspect are implemented.

[0021] In an eighth aspect, a network side device is provided, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the third aspect are implemented.

[0022] In the ninth 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 method described in the first aspect are implemented, or the steps of the method described in the second aspect are implemented, or the steps of the method described in the third aspect are implemented.

[0023] In the tenth aspect, a wireless communication system is provided, comprising: a terminal and a network side device, wherein the terminal can be used to execute the steps of the method described in the first aspect, or the steps of the method described in the second aspect, and the network side device can be used to execute the steps of the method described in the second aspect.

[0024] In the eleventh aspect, a chip is provided, which includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the method as described in the first aspect, or the method as described in the second aspect, or the method as described in the third aspect.

[0025] In the twelfth 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 method described in the first aspect, or the steps of the method described in the second aspect, or the steps of the method described in the third aspect.

[0026] In an embodiment of the present application, the first terminal sends a first model identification request to the network side device, wherein the first model identification request is used to request the network side device to perform a first model identification of the target artificial intelligence AI unit; the first terminal receives a first model identifier configured by the network side device for the target AI unit, wherein the first model identifier satisfies at least one of the following: a globally unique global identifier; a partially globally unique partial global identifier; and is used for the second terminal to send a second model identification request to the network side device, so that the second terminal can obtain the second model identifier of the target AI unit from the network side device by using the above-mentioned first model identifier, thereby achieving the goal that both the first terminal and the second terminal do not need to request the network side device to register the same target AI unit, reducing signaling overhead and solving the problem of wasting air interface resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] FIG1 shows a block diagram of a wireless communication system to which embodiments of the present application may be applied;

[0028] FIG2 is a schematic diagram showing a flow chart of a model recognition method provided in an embodiment of the present application;

[0029] FIG3 is a schematic diagram showing a flow chart of another model recognition method provided in an embodiment of the present application;

[0030] FIG4 is a schematic flow chart showing another model recognition method provided in an embodiment of the present application;

[0031] FIG5 is a schematic flow chart showing another model recognition method provided in an embodiment of the present application;

[0032] FIG6 shows a schematic structural diagram of a model recognition device provided in an embodiment of the present application;

[0033] FIG7 is a schematic structural diagram of another model recognition device provided in an embodiment of the present application;

[0034] FIG8 is a schematic structural diagram of another model recognition device provided in an embodiment of the present application;

[0035] FIG9 shows a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0036] FIG10 is a schematic diagram showing the hardware structure of a terminal provided in an embodiment of the present application;

[0037] FIG11 shows a schematic diagram of the hardware structure of a network-side device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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 technology described 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 illustrative 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) systems. th Generation, 6G) communication system.

[0042] 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 may 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), an ATM, 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. 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 (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.

[0043] 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.

[0044] In an embodiment of the present application, a special model identification process is performed to save signaling overhead for model identification. UEs of a certain type / manufacturer are divided into two categories: Category 1 UEs and Category 2 UEs. When a manufacturer generates a new model, Category 1 UEs perform model identification on the network side and obtain a model ID, while Category 2 UEs do not need to perform model identification on the network side and can directly use the Model ID obtained through model identification. Therefore, it is not necessary for all UEs to repeatedly perform model identification on the same model, thereby reducing the signaling overhead of the system.

[0045] The first type of UE (referred to as the first terminal in the embodiment of this application): After obtaining the first model, the UE performs a first special model identification (model identification, or model registration, model authentication, model information reporting, model information confirmation, etc.) to the network side to obtain the model ID of the first model (first model ID).

[0046] The second type of UE (referred to as the second terminal in the embodiment of the present application): directly uses the Model ID (first model ID) of the first model obtained by the first type of UE; or uses the first model ID to perform a second special model identification to obtain a second model ID.

[0047] The first model (referred to as the target AI unit in the embodiment of the present application): a new model, an unregistered model; both the first type of UE and the second type of UE have the new model, or share the new model, or the new model can be used on the first type of UE and the second type of UE.

[0048] The following, in conjunction with the accompanying drawings, describes in detail the model recognition method, device, terminal and network-side equipment provided in the embodiments of the present application through some embodiments and their application scenarios.

[0049] FIG2 shows a flow chart of a model recognition method provided in an embodiment of the present application. Referring to FIG2 , the method includes the following steps.

[0050] In step 202, the first terminal sends a first model recognition request to the network side device, wherein the first model recognition request is used to request the network side device to perform a first model recognition of a target artificial intelligence (AI) unit.

[0051] Among them, the AI ​​unit in the embodiment of the present application (which may also be referred to as AI model, AI structure, AI function, AI characteristic, machine learning (ML) model, machine learning unit, neural network, neural network function, neural network function, etc.) may refer to a processing unit that can implement specific algorithms, formulas, processing procedures, capabilities, etc. related to AI; or the AI ​​unit may be a processing method, algorithm, function, module or unit for a specific data set; or the AI ​​unit may be a processing method, algorithm, function, module or unit running on AI / machine learning (ML) related hardware such as a graphics processing unit (GPU), a neural network processor (NPU), a tensor processing unit (TPU), an application specific integrated circuit (ASIC), etc., and this application does not make specific limitations on this. Optionally, the above-mentioned specific data set includes the input and / or output of the AI ​​unit.

[0052] In step 204, the first terminal receives a first model identifier configured by the network device for the target AI unit, wherein the first model identifier satisfies at least one of the following:

[0053] (1) A globally unique global identifier;

[0054] (2) partially globally unique partial global identifier;

[0055] (3) The second terminal sends a second model recognition request to the network side device.

[0056] Optionally, the first model identifier can be an AI model identifier, an AI structure identifier, an AI algorithm identifier, or an identifier of a specific data set associated with an AI unit, or an identifier of a specific scenario, environment, channel feature, or device related to AI / ML, or an identifier of an AI / ML-related function, feature, capability, or module. This application does not make any specific limitations on this.

[0057] The target AI unit is a new AI unit or an unregistered AI unit. Both the first terminal and the second terminal have the target AI unit, or the first terminal and the second terminal share the target AI unit, or the target AI unit can be used on the first terminal and the second terminal.

[0058] The second terminal mentioned above is a terminal different from the first terminal, or the second terminal is the first terminal at a different time. For example, if the first terminal loses connection with the network device or reconnects, or if the network device does not have or deletes the target AI unit information of the first terminal, the first terminal needs to resend the model recognition request to the network device, and the network device will treat the first terminal as the second terminal.

[0059] In an embodiment of the present application, a first terminal sends a first model identification request to a network-side device, and the first model identification request is used to request the network-side device to perform a first model identification of a target AI unit, such as model registration, model identification, model information reporting, model information confirmation, etc. of the target AI unit. Then, the first terminal receives a first model identifier configured by the network-side device for the target AI unit, and the first model identifier satisfies at least one of the following conditions: a globally unique global identifier, a partially globally unique partial global identifier, and a second terminal sending a second model identification request to the network-side device. This allows the second terminal to obtain the second model identifier of the target AI unit from the network-side device by using the above-mentioned first model identifier, thereby eliminating the need for both the first terminal and the second terminal to request registration of the same target AI unit from the network-side device, reducing signaling overhead, and solving the problem of wasting air interface resources.

[0060] In one implementation, the first model recognition request carries first information, and the first information may include at least one of the following:

[0061] (1) Type information of the first terminal, for example, a type identifier of the first terminal;

[0062] (2) manufacturer information of the first terminal, for example, the manufacturer identifier of the first terminal;

[0063] (3) vendor information of the first terminal, for example, a vendor identifier of the first terminal;

[0064] (4) information about the developer of the first terminal, for example, an identifier of the developer of the first terminal;

[0065] (5) dedicated signaling, the dedicated signaling being used to indicate that the first model identification request is used to trigger a specific model identification process;

[0066] (6) A dedicated identifier, where the dedicated identifier is used to indicate that the first model recognition request is used to trigger a specific model recognition process.

[0067] In one implementation, the first model identifier includes one of the following: the first information; and associated information about the first information. For example, the first model identifier may directly include the first information, or may also include associated information about the first information, such as associated information calculated from the first information.

[0068] For example, when a model identifier includes at least one of the first information and associated information of the first information, the model identifier is referred to as a globally unique global identifier or a partially globally unique partially global identifier. For a globally unique global identifier, AI units under different qualification conditions have different identifiers. For a partially globally unique partially global identifier, the qualification conditions are divided into global and local qualifications. AI units under different global qualification conditions have different identifiers, while AI units under different local qualification conditions can have the same identifier. For example, the qualification conditions include at least one of the following: unit structure, unit parameters, unit type, functions, characteristics, terminal capabilities, applicable scenarios, applicable conditions, associated datasets, manufacturer, vendor, developer, etc. For example, global qualifications include at least one of the following: functions, characteristics, manufacturer, vendor, developer, etc. For example, local qualifications include at least one of the following: unit structure, unit parameters, unit type, functions, characteristics, terminal capabilities, applicable scenarios, applicable conditions, associated datasets, etc. Optionally, the first information may also include the content of the qualification conditions.

[0069] In an embodiment of the present application, the first model identification request sent by the first terminal to the network side device carries relevant information of the first terminal, such as the type information of the first terminal, the manufacturer information of the first terminal, the seller information of the first terminal, and the developer information of the first terminal. By sending the relevant information of the first terminal to the network side device, the network side device can know the first terminal requesting model identification of the target AI unit; and the first model identification request also carries dedicated signaling or dedicated identification, so that when the network side device receives the dedicated signaling or dedicated identification, it can trigger the specific model identification process in the first model identification request indicated by the dedicated signaling or dedicated identification.

[0070] In one implementation, the first model identifier satisfies at least one of the following:

[0071] (1) Different target AI units have different first model identifiers;

[0072] (2) The same target AI unit has the same first model identifier;

[0073] (3) Different types of target AI units have different first model identifiers;

[0074] (4) Target AI units from different manufacturers have different first model identifiers;

[0075] (5) Target AI units of different vendors have different first model identifiers;

[0076] (6) Target AI units from different developers have different first model identifiers.

[0077] Among them, the first model identifier configured by the network side device for the target AI unit has many different situations. For example, different target AI units have different first model identifiers, and the terminal receives different first model identifiers and correspondingly registers different target AI units; for the same target AI unit with the same first model identifier, when multiple terminals have the same target AI unit, the network side device configures the same first model identifier for the target AI unit. After some of the multiple terminals obtain the first model identifier, the other terminals can also directly use the first model identifier; for different types of target AI units have different first model identifiers, target AI units from different manufacturers have different first model identifiers, target AI units from different sellers have different first model identifiers, and target AI units from different developers have different first model identifiers, the network side device configures different first model identifiers for the target AI units according to the different first model identifiers of different types, different manufacturers, different sellers, and different developers, distinguishes different target AI units, and accurately registers them.

[0078] FIG3 shows a flow chart of another model recognition method provided in an embodiment of the present application in one implementation manner. Referring to FIG3 , the method includes the following steps.

[0079] Step 302: The first terminal obtains the target AI unit.

[0080] Step 304: The first terminal sends a first model recognition request to the network device.

[0081] Among them, the first model recognition request is used to request the network side device to perform a first model recognition of the target artificial intelligence AI unit.

[0082] Step 306: The first terminal receives a first model identifier configured by the network-side device for the target AI unit.

[0083] The first model identifier satisfies at least one of the following: a globally unique global identifier; a partially globally unique partial global identifier; and is used by the second terminal to send a second model identification request to the network side device.

[0084] In one implementation, the above-mentioned step 302 in which the first terminal obtains the target AI unit may include: the first terminal downloads the target AI unit from a target server, wherein the target server includes at least one of the following: a server of the first terminal, a third-party server, a core network server, and a network-side server (for example, a server of another network vendor).

[0085] In an embodiment of the present application, the first terminal can download the target AI unit from at least one of the first terminal's server, a third-party server, a core network server, and a network side server to obtain the target AI unit, and then register the target AI unit by sending a first model recognition request to the network side device and receiving the first model identifier configured by the network side device for the target AI unit.

[0086] In one implementation, the first model identification request carries at least partial description information of the target AI unit.

[0087] Optionally, the description information of the target AI unit may include at least one of the following:

[0088] (1) the function of the target AI unit;

[0089] (2) the features or feature groups enabled by the target AI unit;

[0090] (3) applicable scenarios of the target AI unit;

[0091] (4) Applicable configuration of the target AI unit;

[0092] (5) input information of the target AI unit;

[0093] (6) the input information format of the target AI unit;

[0094] (7) output information of the target AI unit;

[0095] (8) the output information format of the target AI unit;

[0096] (9) Auxiliary information of the target AI unit.

[0097] The auxiliary information of the target AI unit may include hardware configuration information, antenna configuration information, wireless resource configuration information, wireless environment feature information, etc. that assist the AI ​​unit in reasoning. This information may be directly used by the target AI unit, or converted into certain identification or feature information for implicit use by the AI ​​unit. The description information of the target AI unit carried in the first model identification request is used to enable the network-side device to obtain relevant information about the target AI unit to be registered.

[0098] FIG4 shows a flow chart of another model recognition method provided in an embodiment of the present application. Referring to FIG4 , the method includes the following steps.

[0099] Step 402: The second terminal obtains the first model identifier obtained by the first terminal by triggering the first model identification;

[0100] Step 404: The second terminal sends a second model identification request to the network device, wherein the second model identification request is used to request the network device to perform second model identification of the target AI unit, and the second model identification request carries the first model identifier;

[0101] Step 406: The second terminal obtains a second model identifier configured by the network-side device for the target AI unit.

[0102] In an embodiment of the present application, the second terminal can obtain the second model identifier of the target AI unit from the network side device by using the first model identifier obtained by the first terminal by triggering the first model identification, thereby achieving the goal that both the first terminal and the second terminal do not need to request registration of the same target AI unit from the network side device, reducing signaling overhead and solving the problem of wasting air interface resources.

[0103] In one implementation, the second model identifier includes a local identifier. This local identifier is used for lifecycle management of the target AI unit. For example, AI units with different qualifying conditions can have the same local identifier. For example, the qualifying conditions include at least one of the following: unit structure, unit parameters, unit type, function, characteristics, terminal capabilities, applicable scenarios, applicable conditions, associated datasets, manufacturer, vendor, developer, etc.

[0104] In one implementation, the second model recognition request does not carry the description information of the target AI unit.

[0105] In an embodiment of the present application, since the first model identification request carries at least partial description information of the target AI unit, when the first terminal sends the first model identification request to the network side device, the network side device has already obtained at least partial description information of the target AI unit. Therefore, the second model identification request sent by the second terminal to the network side device does not need to carry all the description information of the target AI unit, further saving signaling overhead.

[0106] In one implementation, the second model recognition request carries partial description information of the target AI unit.

[0107] In an embodiment of the present application, since the first model identification request carries at least partial description information of the target AI unit, when the first terminal sends the first model identification request to the network side device, the network side device has already obtained at least partial description information of the target AI unit. Therefore, the second model identification request sent by the second terminal to the network side device only needs to carry partial description information of the target AI unit, such as the function of the target AI unit, the features or feature groups enabled by the target AI unit, further saving signaling overhead.

[0108] Optionally, the description information of the target AI unit includes at least one of the following:

[0109] (1) the function of the target AI unit;

[0110] (2) the features or feature groups enabled by the target AI unit;

[0111] (3) applicable scenarios of the target AI unit;

[0112] (4) Applicable configuration of the target AI unit;

[0113] (5) input information of the target AI unit;

[0114] (6) the input information format of the target AI unit;

[0115] (7) output information of the target AI unit;

[0116] (8) the output information format of the target AI unit;

[0117] (9) Auxiliary information of the target AI unit.

[0118] Among them, the auxiliary information of the target AI unit may include hardware configuration information, antenna configuration information, wireless resource configuration information, wireless environment feature information, etc. that assist the AI ​​unit in reasoning. This information may be directly used by the target AI unit, or converted into certain identification or feature information to be implicitly used by the AI ​​unit.

[0119] In one implementation, the first model identifier includes one of the following: first information, and associated information of the first information.

[0120] In one implementation, the first information may include at least one of the following:

[0121] (1) type information of the first terminal;

[0122] (2) manufacturer information of the first terminal;

[0123] (3) seller information of the first terminal;

[0124] (4) information about the developer of the first terminal;

[0125] (5) dedicated signaling, the dedicated signaling being used to indicate that the first model identification request is used to trigger a specific model identification process;

[0126] (6) A dedicated identifier, where the dedicated identifier is used to indicate that the first model recognition request is used to trigger a specific model recognition process.

[0127] In an embodiment of the present application, the second terminal obtains the first model identifier obtained by the first terminal by triggering the first model identification, and the first model identifier includes the above-mentioned information (1) to (6). The second terminal can determine the information that the network side device has already learned based on the information carried in the first model identifier, thereby avoiding repeatedly sending the same information to the network side device.

[0128] For example, when a model identifier includes at least one of the first information and associated information of the first information, the model identifier is referred to as a globally unique global identifier or a partially globally unique partially global identifier. For a globally unique global identifier, AI units under different qualification conditions have different identifiers. For a partially globally unique partially global identifier, the qualification conditions are divided into global and local qualifications. AI units under different global qualification conditions have different identifiers, while AI units under different local qualification conditions can have the same identifier. For example, the qualification conditions include at least one of the following: unit structure, unit parameters, unit type, functions, characteristics, terminal capabilities, applicable scenarios, applicable conditions, associated datasets, manufacturer, vendor, developer, etc. For example, global qualifications include at least one of the following: functions, characteristics, manufacturer, vendor, developer, etc. For example, local qualifications include at least one of the following: unit structure, unit parameters, unit type, functions, characteristics, terminal capabilities, applicable scenarios, applicable conditions, associated datasets, etc. Optionally, the first information may also include the content of the qualification conditions.

[0129] In one implementation, the second model identifier satisfies at least one of the following:

[0130] (1) The second model identifier does not include the first information of the first terminal;

[0131] (2) the second model identifier does not include associated information of the first information of the first terminal;

[0132] (3) different types of target AI units have the same second model identifier;

[0133] (4) Target AI units from different manufacturers have the same second model identifier;

[0134] (5) Target AI units from different vendors have the same second model identifier;

[0135] (6) Target AI units from different developers have the same second model identifier;

[0136] (7) Different target AI units have the same second model identifier;

[0137] (8) The same target AI unit has different second model identifiers.

[0138] In an embodiment of the present application, since the first model identifier received by the first terminal includes the first information or the associated information of the first information, the first information and the associated information of the first information are carried by the first terminal when sending the first model identification request to the network-side device, which includes the relevant information of the first terminal, the network-side device needs to send a second model identifier that does not include the first information of the first terminal or the management information of the first information to the second terminal. In addition, target AI units of different types / manufacturers may also have the same second model identifier. For example, the second model identifier of a target AI unit of a certain type / manufacturer may be the same as the second model identifier of a target AI unit of another type / manufacturer.

[0139] In one implementation, the first model identifier or the second model identifier is used for lifecycle management of the target AI unit.

[0140] The lifecycle management of the target AI unit includes at least one of the following:

[0141] (1) Activation of the target AI unit;

[0142] (2) deactivation of the target AI unit;

[0143] (3) selection of the target AI unit;

[0144] (4) switching of the target AI unit;

[0145] (5) reasoning of the target AI unit;

[0146] (6) Performance monitoring of the target AI unit;

[0147] (7) performance management of the target AI unit;

[0148] (8) Falling back to the non-target AI unit;

[0149] (9) training of the target AI unit;

[0150] (10) updating of the target AI unit;

[0151] (11) Registration of the target AI unit;

[0152] (12) Configuration of the target AI unit;

[0153] (13) Confirmation of the identity of the target AI unit.

[0154] In one implementation, the subject of lifecycle management includes one of the following: only the second terminal; only the network side device; the second terminal and the network side device.

[0155] In an embodiment of the present application, the process of using the target AI unit may include activation, deactivation, switching, inference, performance monitoring, performance management, training, updating, registration, configuration, and identity confirmation of the target AI unit, as well as fallback to a non-target AI unit. The first model identifier or the second model identifier can be used to manage the lifecycle of the target AI unit. The lifecycle management can be performed only by the second terminal, only by the network-side device, or by both the second terminal and the network-side device.

[0156] FIG5 shows a flow chart of another model recognition method provided in an embodiment of the present application. Referring to FIG5 , the method includes the following steps.

[0157] Step 502: The network-side device receives a first model recognition request sent by the first terminal, wherein the first model recognition request is used to request the network-side device to perform first model recognition of the target AI unit;

[0158] In step 504, the network-side device assigns a first model identifier to the target AI unit by performing a first model identification on the target AI unit, wherein the first model identifier satisfies at least one of the following conditions: a globally unique global identifier; a partially globally unique partial global identifier; and is used by the second terminal to send a second model identification request to the network-side device.

[0159] Step 506: The network-side device sends the first model identifier to the first terminal.

[0160] In an embodiment of the present application, a network-side device receives a first model identification request sent by a first terminal requesting the network-side device to perform first model identification of a target AI unit, and then assigns a first model identifier to the target AI unit through the first model identification of the target AI unit, and then sends the first model identifier to the first terminal, where the first model identifier satisfies at least one of the following: a globally unique global identifier, a partially globally unique partial global identifier, and a second terminal is used to send a second model identification request to the network-side device; the first terminal receives the first model identifier configured by the network-side device for the target AI unit, and the first model identifier can be used by the second terminal to send a second model identification request to the network-side device, so that the second terminal can obtain the second model identifier of the target AI unit from the network-side device by using the above-mentioned first model identifier, thereby achieving the goal that both the first terminal and the second terminal do not need to request registration of the same target AI unit to the network-side device, thereby reducing signaling overhead and solving the problem of wasting air interface resources.

[0161] In one implementation, the first model recognition request carries first information, and the first information may include at least one of the following:

[0162] (1) type information of the first terminal;

[0163] (2) manufacturer information of the first terminal;

[0164] (3) seller information of the first terminal;

[0165] (4) information about the developer of the first terminal;

[0166] (5) dedicated signaling, the dedicated signaling being used to indicate that the first model identification request is used to trigger a specific model identification process;

[0167] (6) A dedicated identifier, where the dedicated identifier is used to indicate that the first model recognition request is used to trigger a specific model recognition process.

[0168] In one implementation, the first model identifier includes one of the following: the first information, and associated information of the first information.

[0169] In an embodiment of the present application, the network side device can obtain the first terminal requesting model identification of the target AI unit by receiving the first information carried in the first model identification request sent by the first terminal; the dedicated signaling or dedicated identifier carried therein enables the network side device to trigger the specific model identification process in the first model identification request indicated by the dedicated signaling or dedicated identifier when receiving the dedicated signaling or dedicated identifier.

[0170] In one implementation, the first model identifier satisfies at least one of the following:

[0171] (1) Different target AI units have different first model identifiers;

[0172] (2) The same target AI unit has the same first model identifier;

[0173] (3) Different types of target AI units have different first model identifiers;

[0174] (4) Target AI units from different manufacturers have different first model identifiers;

[0175] (5) Target AI units of different vendors have different first model identifiers;

[0176] (6) Target AI units from different developers have different first model identifiers.

[0177] In one implementation, the first model identification request carries at least partial description information of the target AI unit.

[0178] For detailed description, please refer to the embodiment of the model recognition method in FIG2 , which will not be repeated here.

[0179] In one implementation, after the network-side device performs first model identification on the target AI unit and assigns a first model identifier to the target AI unit in step 504, the method may further include:

[0180] Step 508: The network-side device receives a second model identification request sent by the second terminal, wherein the second model identification request is used to request the network-side device to perform a second model identification of the target AI unit, and the second model identification request carries the first model identifier;

[0181] Step 510: The network-side device configures a second model identifier for the target AI unit;

[0182] Step 512: The network-side device sends the second model identifier to the second terminal.

[0183] In an embodiment of the present application, the network side device receives a first model recognition request sent by the first terminal requesting the network side device to perform first model recognition of the target AI unit, and then assigns a first model identifier to the target AI unit through the first model recognition of the target AI unit. After the second terminal obtains the first model identifier, it sends a second model recognition request to the network side device. After receiving the second model recognition request, the network side device configures the second model identifier for the target AI unit and then sends it to the second terminal. The second terminal also obtains the second model identifier configured for the target AI unit, thereby realizing the registration of the target AI unit in the first terminal and the second terminal.

[0184] In one implementation, the second model identifier includes a local identifier. This local identifier is used for lifecycle management of the target AI unit. For example, AI units with different qualifying conditions can have the same local identifier. For example, the qualifying conditions include at least one of the following: unit structure, unit parameters, unit type, function, characteristics, terminal capabilities, applicable scenarios, applicable conditions, associated datasets, manufacturer, vendor, developer, etc.

[0185] In one implementation, the second model recognition request does not carry the description information of the target AI unit.

[0186] In an embodiment of the present application, since the first model identification request carries at least partial description information of the target AI unit, when the first terminal sends the first model identification request to the network side device, the network side device has already obtained at least partial description information of the target AI unit. Therefore, the second model identification request sent by the second terminal to the network side device does not need to carry all the description information of the target AI unit, further saving signaling overhead.

[0187] In one implementation, the second model recognition request carries partial description information of the target AI unit.

[0188] In an embodiment of the present application, since the first model identification request carries at least partial description information of the target AI unit, when the first terminal sends the first model identification request to the network side device, the network side device has already obtained at least partial description information of the target AI unit. Therefore, the second model identification request sent by the second terminal to the network side device only needs to carry partial description information of the target AI unit, such as the function of the target AI unit, the features or feature groups enabled by the target AI unit, further saving signaling overhead.

[0189] Optionally, the description information of the target AI unit includes at least one of the following:

[0190] (1) the function of the target AI unit;

[0191] (2) the features or feature groups enabled by the target AI unit;

[0192] (3) applicable scenarios of the target AI unit;

[0193] (4) Applicable configuration of the target AI unit;

[0194] (5) input information of the target AI unit;

[0195] (6) the input information format of the target AI unit;

[0196] (7) output information of the target AI unit;

[0197] (8) the output information format of the target AI unit;

[0198] (9) Auxiliary information of the target AI unit.

[0199] Among them, the auxiliary information of the target AI unit may include hardware configuration information, antenna configuration information, wireless resource configuration information, wireless environment feature information, etc. that assist the AI ​​unit in reasoning. This information may be directly used by the target AI unit, or converted into certain identification or feature information to be implicitly used by the AI ​​unit.

[0200] In one implementation, the second model identifier satisfies at least one of the following:

[0201] (1) The second model identifier does not include the first information of the first terminal;

[0202] (2) the second model identifier does not include associated information of the first information of the first terminal;

[0203] (3) different types of target AI units have the same second model identifier;

[0204] (4) Target AI units from different manufacturers have the same second model identifier;

[0205] (5) Target AI units from different vendors have the same second model identifier;

[0206] (6) Target AI units from different developers have the same second model identifier;

[0207] (7) Different target AI units have the same second model identifier;

[0208] (8) The same target AI unit has different second model identifiers.

[0209] In an embodiment of the present application, since the first model identifier received by the first terminal includes the first information or the associated information of the first information, the first information and the associated information of the first information are carried by the first terminal when sending the first model identification request to the network-side device, which includes the relevant information of the first terminal, the network-side device needs to send a second model identifier that does not include the first information of the first terminal or the management information of the first information to the second terminal. In addition, target AI units of different types / manufacturers may also have the same second model identifier. For example, the second model identifier of a target AI unit of a certain type / manufacturer may be the same as the second model identifier of a target AI unit of another type / manufacturer.

[0210] In one implementation, the network-side device includes a core network and a base station; and step 504 in which the network-side device performs first model identification on the target AI unit and assigns a first model identifier to the target AI unit may include:

[0211] Step 5041: The base station sends a request for model identifier allocation to the core network;

[0212] Step 5042: The core network allocates a first model identifier to the target AI unit and sends it to the base station.

[0213] In an embodiment of the present application, after receiving the first model identification request of the first terminal, an application for model identifier allocation is sent to the core network through the base station, and then the core network allocates the first model identifier for the target AI unit and sends it to the base station, which sends the first model identifier to the first terminal.

[0214] In one implementation, the above method may also include: the core network sends allocation information of the first model identifier to the base station, wherein the allocation information is used to indicate one of the following: the rule of the first model identifier, the allocation suggestion of the first model identifier, and the allocation range of the first model identifier.

[0215] In an embodiment of the present application, the base station sends an application for model identifier allocation to the core network, and then the core network allocates a first model identifier for the target AI unit and sends it to the base station. The core network can also send allocation information of the first model identifier to the base station, and use the allocation information to indicate the rules, allocation suggestions, and allocation range of the first model identifier.

[0216] In one implementation, the network side device includes a core network and a base station; the network side device configures a second model identifier for the target AI unit, including: the base station configures the second model identifier for the target AI unit.

[0217] In an embodiment of the present application, after receiving the second model identification request from the second terminal, a second model identifier is allocated to the target AI unit through the base station, and then the second model identifier is sent to the second terminal.

[0218] The model recognition method provided in the embodiment of the present application can be executed by a model recognition device. In the embodiment of the present application, the model recognition device provided in the embodiment of the present application is described by taking the execution of the model recognition method by the model recognition device as an example.

[0219] FIG6 shows a schematic structural diagram of a model recognition device provided in an embodiment of the present application. Referring to FIG6 , the device 600 may include a sending module 61 and a receiving module 62 .

[0220] The sending module 61 is configured to send a first model identification request to the network-side device, wherein the first model identification request is used to request the network-side device to perform first model identification of the target artificial intelligence (AI) unit; and the receiving module 62 is configured to receive a first model identifier configured by the network-side device for the target AI unit, wherein the first model identifier satisfies at least one of the following:

[0221] (1) A globally unique global identifier;

[0222] (2) partially globally unique partial global identifier;

[0223] (3) The second terminal sends a second model recognition request to the network side device.

[0224] In one implementation, the first model identification request carries first information, where the first information includes at least one of the following:

[0225] (1) type information of the first terminal;

[0226] (2) manufacturer information of the first terminal;

[0227] (3) seller information of the first terminal;

[0228] (4) information about the developer of the first terminal;

[0229] (5) dedicated signaling, the dedicated signaling being used to indicate that the first model identification request is used to trigger a specific model identification process;

[0230] (6) A dedicated identifier, where the dedicated identifier is used to indicate that the first model recognition request is used to trigger a specific model recognition process.

[0231] In one implementation, the first model identifier includes one of the following: the first information, and associated information of the first information.

[0232] In one implementation, the first model identifier satisfies at least one of the following:

[0233] (1) Different target AI units have different first model identifiers;

[0234] (2) The same target AI unit has the same first model identifier;

[0235] (3) Different types of target AI units have different first model identifiers;

[0236] (4) Target AI units from different manufacturers have different first model identifiers;

[0237] (5) Target AI units of different vendors have different first model identifiers;

[0238] (6) Target AI units from different developers have different first model identifiers.

[0239] In one implementation, the apparatus 600 may further include an acquisition module configured to acquire the target AI unit.

[0240] In one implementation, the above-mentioned acquisition module can be used to download the target AI unit from a target server, wherein the target server includes at least one of the following: a server of the first terminal, a third-party server, a core network server, and a network side server.

[0241] In one implementation, the first model identification request carries at least partial description information of the target AI unit.

[0242] In one implementation, the description information of the target AI unit includes at least one of the following:

[0243] (1) the function of the target AI unit;

[0244] (2) the features or feature groups enabled by the target AI unit;

[0245] (3) applicable scenarios of the target AI unit;

[0246] (4) Applicable configuration of the target AI unit;

[0247] (5) input information of the target AI unit;

[0248] (6) the input information format of the target AI unit;

[0249] (7) output information of the target AI unit;

[0250] (8) the output information format of the target AI unit;

[0251] (9) Auxiliary information of the target AI unit.

[0252] FIG7 shows a schematic structural diagram of another model recognition device provided in an embodiment of the present application. Referring to FIG7 , the device 700 may include a first acquisition module 71 , a sending module 72 , and a second acquisition module 73 .

[0253] Among them, the first acquisition module 71 is used to obtain the first model identifier obtained by the first terminal by triggering the first model identification; the sending module 72 is used to send a second model identification request to the network side device, wherein the second model identification request is used to request the network side device to perform the second model identification of the target AI unit, and the second model identification request carries the first model identifier; the second acquisition module 73 is used to obtain the second model identifier configured by the network side device for the target AI unit.

[0254] In one implementation, the second model identifier includes: a local identifier.

[0255] In one implementation, the second model identification request does not carry description information of the target AI unit.

[0256] In one implementation, the second model identification request carries partial description information of the target AI unit.

[0257] In one implementation, the description information of the target AI unit includes at least one of the following:

[0258] (1) the function of the target AI unit;

[0259] (2) the features or feature groups enabled by the target AI unit;

[0260] (3) applicable scenarios of the target AI unit;

[0261] (4) Applicable configuration of the target AI unit;

[0262] (5) input information of the target AI unit;

[0263] (6) the input information format of the target AI unit;

[0264] (7) output information of the target AI unit;

[0265] (8) the output information format of the target AI unit;

[0266] (9) Auxiliary information of the target AI unit.

[0267] In one implementation, the first model identifier includes one of the following: first information, and associated information of the first information.

[0268] In one implementation, the first information includes at least one of the following:

[0269] (1) type information of the first terminal;

[0270] (2) manufacturer information of the first terminal;

[0271] (3) seller information of the first terminal;

[0272] (4) information about the developer of the first terminal;

[0273] (5) dedicated signaling, the dedicated signaling being used to indicate that the first model identification request is used to trigger a specific model identification process;

[0274] (6) A dedicated identifier, where the dedicated identifier is used to indicate that the first model recognition request is used to trigger a specific model recognition process.

[0275] In one implementation, the second model identifier satisfies at least one of the following:

[0276] (1) The second model identifier does not include the first information of the first terminal;

[0277] (2) the second model identifier does not include associated information of the first information of the first terminal;

[0278] (3) different types of target AI units have the same second model identifier;

[0279] (4) Target AI units from different manufacturers have the same second model identifier;

[0280] (5) Target AI units from different vendors have the same second model identifier;

[0281] (6) Target AI units from different developers have the same second model identifier;

[0282] (7) Different target AI units have the same second model identifier;

[0283] (8) The same target AI unit has different second model identifiers.

[0284] In one implementation, the first model identifier or the second model identifier is used for lifecycle management of the target AI unit.

[0285] In one implementation, the lifecycle management of the target AI unit includes at least one of the following:

[0286] (1) Activation of the target AI unit;

[0287] (2) deactivation of the target AI unit;

[0288] (3) selection of the target AI unit;

[0289] (4) switching of the target AI unit;

[0290] (5) reasoning of the target AI unit;

[0291] (6) Performance monitoring of the target AI unit;

[0292] (7) performance management of the target AI unit;

[0293] (8) Falling back to the non-target AI unit;

[0294] (9) training of the target AI unit;

[0295] (10) updating of the target AI unit;

[0296] (11) Registration of the target AI unit;

[0297] (12) Configuration of the target AI unit;

[0298] (13) Confirmation of the identity of the target AI unit.

[0299] In one implementation, the subject of the lifecycle management includes one of the following: only the second terminal; only the network side device; the second terminal and the network side device.

[0300] The model recognition 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 of an electronic device, such as an integrated circuit or chip. The electronic device can be a terminal. For example, the terminal can include but is not limited to the types of terminal 11 listed above, and the embodiments of the present application do not specifically limit this.

[0301] The model recognition device provided in the embodiment of the present application can implement the various processes implemented in the method embodiments of Figures 2 to 4 and achieve the same technical effects. To avoid repetition, they will not be described here.

[0302] FIG8 shows a schematic structural diagram of another model recognition device provided in an embodiment of the present application. Referring to FIG8 , the device 800 may include a receiving module 81 , an allocating module 82 , and a sending module 83 .

[0303] The receiving module 81 is configured to receive a first model recognition request sent by a first terminal, wherein the first model recognition request is used to request the network-side device to perform first model recognition of a target AI unit;

[0304] an allocating module 82 configured to allocate a first model identifier to the target AI unit by performing a first model identification on the target AI unit, wherein the first model identifier satisfies at least one of the following conditions: a globally unique global identifier; a partially globally unique partial global identifier; and is used by the second terminal to send a second model identification request to the network-side device;

[0305] The sending module 83 is configured to send the first model identifier to the first terminal.

[0306] In one implementation, the first model identification request carries first information, where the first information includes at least one of the following:

[0307] (1) type information of the first terminal;

[0308] (2) manufacturer information of the first terminal;

[0309] (3) seller information of the first terminal;

[0310] (4) information about the developer of the first terminal;

[0311] (5) dedicated signaling, the dedicated signaling being used to indicate that the first model identification request is used to trigger a specific model identification process;

[0312] (6) A dedicated identifier, where the dedicated identifier is used to indicate that the first model recognition request is used to trigger a specific model recognition process.

[0313] In one implementation, the first model identifier includes one of the following: the first information, and associated information of the first information.

[0314] In one implementation, the first model identifier satisfies at least one of the following:

[0315] (1) Different target AI units have different first model identifiers;

[0316] (2) The same target AI unit has the same first model identifier;

[0317] (3) Different types of target AI units have different first model identifiers;

[0318] (4) Target AI units from different manufacturers have different first model identifiers;

[0319] (5) Target AI units of different vendors have different first model identifiers;

[0320] (6) Target AI units from different developers have different first model identifiers.

[0321] In one implementation, the first model identification request carries at least partial description information of the target AI unit.

[0322] In one implementation, the above-mentioned receiving module 81 can also be used to receive a second model identification request sent by the second terminal, wherein the second model identification request is used to request the network side device to perform a second model identification of the target AI unit, and the second model identification request carries the first model identifier; the above-mentioned allocation module 82 can also be used to configure a second model identifier for the target AI unit; the above-mentioned sending module 83 can also be used to send the second model identifier to the second terminal.

[0323] In one implementation, the second model identifier includes: a local identifier.

[0324] In one implementation, the second model identification request does not carry description information of the target AI unit.

[0325] In one implementation, the second model identification request carries partial description information of the target AI unit.

[0326] In one implementation, the description information of the target AI unit includes at least one of the following:

[0327] (1) the function of the target AI unit;

[0328] (2) the features or feature groups enabled by the target AI unit;

[0329] (3) applicable scenarios of the target AI unit;

[0330] (4) Applicable configuration of the target AI unit;

[0331] (5) input information of the target AI unit;

[0332] (6) the input information format of the target AI unit;

[0333] (7) output information of the target AI unit;

[0334] (8) the output information format of the target AI unit;

[0335] (9) Auxiliary information of the target AI unit.

[0336] In one implementation, the second model identifier satisfies at least one of the following:

[0337] (1) The second model identifier does not include the first information of the first terminal;

[0338] (2) the second model identifier does not include associated information of the first information of the first terminal;

[0339] (3) different types of target AI units have the same second model identifier;

[0340] (4) Target AI units from different manufacturers have the same second model identifier;

[0341] (5) Target AI units from different vendors have the same second model identifier;

[0342] (6) Target AI units from different developers have the same second model identifier;

[0343] (7) Different target AI units have the same second model identifier;

[0344] (8) The same target AI unit has different second model identifiers.

[0345] In one implementation, the allocation module 82 allocates a first model identifier to the target AI unit by performing a first model identification on the target AI unit, which may include: sending an application for model identifier allocation to the core network; and obtaining the first model identifier allocated by the core network to the target AI unit.

[0346] In one implementation, the allocation module 82 is further used to receive allocation information of the first model identifier sent by the core network, wherein the allocation information is used to indicate one of the following: a rule of the first model identifier, an allocation suggestion of the first model identifier, and an allocation range of the first model identifier.

[0347] In one implementation, configuring the second model identifier for the target AI unit in the allocation module 82 may include: configuring the second model identifier for the target AI unit.

[0348] The model recognition 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 chip. The electronic device can be a network-side device, such as a server or a network attached storage (NAS), and the embodiments of the present application do not specifically limit this.

[0349] The model recognition device provided in the embodiment of the present application can implement the various processes implemented in the method embodiment of Figure 5 and achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0350] As shown in Figure 9, an embodiment of the present application further provides a communication device 900, including a processor 901 and a memory 902, wherein the memory 902 stores a program or instruction that can be run on the processor 901. For example, when the communication device 900 is a terminal, the program or instruction is executed by the processor 901 to implement the various steps of the embodiment of the recognition method of the model shown in Figures 2 to 4, and can achieve the same technical effect. When the communication device 900 is a network-side device, the program or instruction is executed by the processor 901 to implement the various steps of the embodiment of the recognition method of the model shown in Figure 5, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0351] The present application also provides a terminal 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 method embodiments shown in Figures 2 to 4. This terminal embodiment corresponds to the aforementioned terminal-side method embodiment, and each implementation process and implementation method of the aforementioned method embodiment is applicable to this terminal embodiment and can achieve the same technical effects. Specifically, Figure 10 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of the present application.

[0352] The terminal 1000 includes but is not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009 and at least some of the components of the processor 1010.

[0353] Those skilled in the art will appreciate that the terminal 1000 may also include a power supply (such as a battery) to power various components. The power supply may be logically connected to the processor 1010 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The terminal structure shown in FIG10 does not limit the terminal. The 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.

[0354] It should be understood that in an embodiment of the present application, the input unit 1004 may include a graphics processing unit (GPU) 10041 and a microphone 10042, and the graphics processing unit 10041 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 1006 may include a display panel 10061, and the display panel 10061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1007 includes a touch panel 10071 and at least one of other input devices 10072. The touch panel 10071 is also called a touch screen. The touch panel 10071 may include two parts: a touch detection device and a touch controller. Other input devices 10072 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 an operating stick, which will not be repeated here.

[0355] In the embodiment of the present application, after receiving downlink data from a network-side device, the RF unit 1001 may transmit the data to the processor 1010 for processing. Furthermore, the RF unit 1001 may send uplink data to the network-side device. Typically, the RF unit 1001 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like.

[0356] The memory 1009 can be used to store software programs or instructions and various data. The memory 1009 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 1009 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. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM), and a direct memory bus random access memory (DRRAM). The memory 1009 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0357] Processor 1010 may include one or more processing units. Optionally, processor 1010 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 1010.

[0358] The radio frequency unit 1001 is configured to send a first model identification request to a network device, wherein the first model identification request is used to request the network device to perform first model identification of a target artificial intelligence (AI) unit; and receive a first model identifier configured by the network device for the target AI unit, wherein the first model identifier satisfies at least one of the following:

[0359] A globally unique global identifier;

[0360] Partially globally unique partially global identifier;

[0361] Used by the second terminal to send a second model recognition request to the network side device.

[0362] It can be understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the method embodiment and achieve the same or corresponding technical effects. To avoid repetition, it will not be described here.

[0363] The present application also provides a network-side 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 method embodiment shown in FIG5 . This network-side device embodiment corresponds to the aforementioned network-side device method embodiment, and each implementation process and implementation method of the aforementioned method embodiment are applicable to this network-side device embodiment and can achieve the same technical effects.

[0364] Specifically, embodiments of the present application also provide a network-side device. As shown in Figure 11, the network-side device 1100 includes an antenna 1101, a radio frequency device 1102, a baseband device 1103, a processor 1104, and a memory 1105. Antenna 1101 is connected to radio frequency device 1102. In the uplink direction, radio frequency device 1102 receives information via antenna 1101 and sends the received information to baseband device 1103 for processing. In the downlink direction, baseband device 1103 processes the information to be transmitted and sends it to radio frequency device 1102. Radio frequency device 1102 processes the received information and then sends it through antenna 1101.

[0365] The method executed by the network-side device in the above embodiment may be implemented in the baseband device 1103 , which includes a baseband processor.

[0366] The baseband device 1103 may, for example, include at least one baseband board, on which multiple chips are arranged, as shown in Figure 11, one of the chips is, for example, a baseband processor, which is connected to the memory 1105 through a bus interface to call the program in the memory 1105 and execute the network device operations shown in the above method embodiment.

[0367] The network side device may further include a network interface 1106 , which is, for example, a Common Public Radio Interface (CPRI).

[0368] Specifically, the network side device 1100 of the embodiment of the present application also includes: instructions or programs stored in the memory 1105 and executable on the processor 1104. The processor 1104 calls the instructions or programs in the memory 1105 to execute the method of executing each module shown in FIG8 and achieve the same technical effect. To avoid repetition, it will not be described here.

[0369] 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 above-mentioned model recognition method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0370] 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.

[0371] 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 above-mentioned model recognition method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0372] 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.

[0373] 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 above-mentioned model recognition method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0374] 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.

[0375] 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.

[0376] 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 method for identifying a model, comprising: The first terminal sends a first model recognition request to the network side device, wherein the first model recognition request is used to request the network side device to perform a first model recognition of the target artificial intelligence AI unit; The first terminal receives a first model identifier configured by the network side device for the target AI unit, wherein the first model identifier satisfies at least one of the following: A globally unique global identifier; Partially globally unique partially globally identified; Used by the second terminal to send a second model identification request to the network side device.

2. The method according to claim 1, wherein: The first model identification request carries first information, wherein the first information includes at least one of the following: type information of the first terminal; manufacturer information of the first terminal; The seller information of the first terminal; Developer information of the first terminal; Dedicated signaling, where the dedicated signaling is used to indicate that the first model identification request is used to trigger a specific model identification process; A dedicated identifier, where the dedicated identifier is used to indicate that the first model identification request is used to trigger a specific model identification process.

3. The method according to claim 2, wherein: The first model identifier includes one of the following: the first information and associated information of the first information.

4. The method according to any one of claims 1 to 3, wherein: The first model identifier satisfies at least one of the following: Different target AI units have different first model identifiers; The same target AI unit has the same first model identifier; Different types of target AI units have different first model identifiers; Target AI units from different manufacturers have different first model identifiers; Target AI units of different vendors have different first model identifiers; Target AI units from different developers have different first model identifiers.

5. The method according to any one of claims 1 to 3, wherein: Before the first terminal sends the first model identification request to the network side device, the method further includes: The first terminal obtains the target AI unit.

6. The method according to claim 5, wherein: The first terminal obtains the target AI unit, including: The first terminal downloads the target AI unit from a target server, wherein the target server includes at least one of the following: a server of the first terminal, a third-party server, a core network server, and a network side server.

7. The method according to any one of claims 1 to 3, wherein: The first model identification request carries at least part of the description information of the target AI unit.

8. The method according to claim 7, wherein: The description information of the target AI unit includes at least one of the following: Functionality of the target AI unit; The feature or feature group enabled by the target AI unit; The applicable scenarios of the target AI unit; Applicable configuration of the target AI unit; Input information of the target AI unit; The input information format of the target AI unit; Output information of the target AI unit; The output information format of the target AI unit; Auxiliary information of the target AI unit.

9. A method for identifying a model, comprising: The second terminal obtains the first model identifier obtained by the first terminal by triggering the first model identification; The second terminal sends a second model identification request to the network side device, wherein the second model identification request is used to request the network side device to perform a second model identification of the target AI unit, and the second model identification request carries the first model identifier; The second terminal obtains a second model identifier configured by the network side device for the target AI unit.

10. The method according to claim 9, wherein: The second model identifier includes: a local identifier.

11. The method according to claim 9, wherein: The second model identification request does not carry the description information of the target AI unit.

12. The method according to claim 9, wherein: The second model identification request carries partial description information of the target AI unit.

13. The method according to claim 9 or 12, wherein: The description information of the target AI unit includes at least one of the following: Functionality of the target AI unit; The feature or feature group enabled by the target AI unit; The applicable scenarios of the target AI unit; Applicable configuration of the target AI unit; Input information of the target AI unit; The input information format of the target AI unit; Output information of the target AI unit; The output information format of the target AI unit; Auxiliary information of the target AI unit.

14. The method according to any one of claims 9 to 13, wherein: The first model identifier includes one of the following: first information and associated information of the first information.

15. The method according to claim 14, wherein: The first information includes at least one of the following: type information of the first terminal; manufacturer information of the first terminal; The seller information of the first terminal; Developer information of the first terminal; Dedicated signaling, where the dedicated signaling is used to indicate that the first model identification request is used to trigger a specific model identification process; A dedicated identifier, where the dedicated identifier is used to indicate that the first model identification request is used to trigger a specific model identification process.

16. The method according to claim 14, wherein: The second model identifier satisfies at least one of the following: The second model identifier does not include first information of the first terminal; The second model identifier does not include associated information of the first information of the first terminal; Different types of target AI units have the same second model identifier; Target AI units from different manufacturers have the same second model identifier; Target AI units from different vendors have the same second model identifier; Target AI units from different developers have the same second model identifier; Different target AI units have the same second model identifier; The same target AI unit has different second model identifiers.

17. The method according to any one of claims 9 to 16, wherein: The first model identifier or the second model identifier is used for lifecycle management of the target AI unit.

18. The method according to claim 17, wherein: The lifecycle management of the target AI unit includes at least one of the following: Activation of the target AI unit; deactivation of the target AI unit; Selection of the target AI unit; Switching of the target AI unit; Reasoning of the target AI unit; Performance monitoring of the target AI unit; performance management of the target AI unit; Fallback to non-target AI unit; Training of the target AI unit; Update of the target AI unit; Registration of the target AI unit; the configuration of the target AI unit; The identity of the target AI unit is confirmed.

19. The method according to claim 17, wherein: The subject of the lifecycle management includes one of the following: for the second terminal only; Only for network-side devices; The second terminal and the network side device.

20. A method for identifying a model, comprising: The network side device receives a first model identification request sent by the first terminal, wherein the first model identification request is used to request the network side device to perform first model identification of the target AI unit; The network side device assigns a first model identifier to the target AI unit by performing a first model identification of the target AI unit, wherein the first model identifier satisfies at least one of the following: a globally unique global identifier; a partially globally unique partial global identifier; and is used by the second terminal to send a second model identification request to the network side device; The network side device sends the first model identifier to the first terminal.

21. The method according to claim 20, wherein: The first model identification request carries first information, wherein the first information includes at least one of the following: type information of the first terminal; manufacturer information of the first terminal; The seller information of the first terminal; Developer information of the first terminal; Dedicated signaling, where the dedicated signaling is used to indicate that the first model identification request is used to trigger a specific model identification process; A dedicated identifier, where the dedicated identifier is used to indicate that the first model identification request is used to trigger a specific model identification process.

22. The method according to claim 21, wherein: The first model identifier includes one of the following: the first information and associated information of the first information.

23. The method according to any one of claims 20 to 22, wherein: The first model identifier satisfies at least one of the following: Different target AI units have different first model identifiers; The same target AI unit has the same first model identifier; Different types of target AI units have different first model identifiers; Target AI units from different manufacturers have different first model identifiers; Target AI units of different vendors have different first model identifiers; Target AI units from different developers have different first model identifiers.

24. The method according to any one of claims 20 to 23, wherein: The first model identification request carries at least part of the description information of the target AI unit.

25. The method according to any one of claims 20 to 23, wherein: After the network-side device assigns a first model identifier to the target AI unit by performing first model identification on the target AI unit, the method further includes: The network side device receives a second model identification request sent by the second terminal, wherein the second model identification request is used to request the network side device to perform a second model identification of the target AI unit, and the second model identification request carries the first model identifier; The network side device configures a second model identifier for the target AI unit; The network side device sends the second model identifier to the second terminal.

26. The method according to claim 25, wherein: The second model identifier includes: a local identifier.

27. The method according to claim 25, wherein: The second model identification request does not carry the description information of the target AI unit.

28. The method according to claim 25, wherein: The second model identification request carries partial description information of the target AI unit.

29. The method of claim 24, 27 or 28, wherein: The description information of the target AI unit includes at least one of the following: Functionality of the target AI unit; The feature or feature group enabled by the target AI unit; The applicable scenarios of the target AI unit; Applicable configuration of the target AI unit; Input information of the target AI unit; The input information format of the target AI unit; Output information of the target AI unit; The output information format of the target AI unit; Auxiliary information of the target AI unit.

30. The method according to any one of claims 20 to 29, wherein: The second model identifier satisfies at least one of the following: The second model identifier does not include first information of the first terminal; The second model identifier does not include associated information of the first information of the first terminal; Different types of target AI units have the same second model identifier; Target AI units from different manufacturers have the same second model identifier; Target AI units from different vendors have the same second model identifier; Target AI units from different developers have the same second model identifier; Different target AI units have the same second model identifier; The same target AI unit has different second model identifiers.

31. The method according to any one of claims 20 to 30, wherein: The network side device includes a core network and a base station; the network side device assigns a first model identifier to the target AI unit by performing a first model identification of the target AI unit, including: The base station sends an application for model identification allocation to the core network; The core network allocates a first model identifier to the target AI unit and sends it to the base station.

32. The method according to claim 31, wherein: The method also includes: the core network sending allocation information of the first model identifier to the base station, wherein the allocation information is used to indicate one of the following: a rule of the first model identifier, an allocation suggestion of the first model identifier, and an allocation range of the first model identifier.

33. The method according to any one of claims 25 to 30, wherein: The network side device includes a core network and a base station; the network side device configures a second model identifier for the target AI unit, including: The base station configures the second model identifier for the target AI unit.

34. A model recognition device, comprising: A sending module, used to send a first model recognition request to a network side device, wherein the first model recognition request is used to request the network side device to perform a first model recognition of a target artificial intelligence AI unit; A receiving module is configured to receive a first model identifier configured by a network side device for the target AI unit, wherein the first model identifier satisfies at least one of the following: A globally unique global identifier; Partially globally unique partially globally identified; Used by the second terminal to send a second model identification request to the network side device.

35. A model recognition device, comprising: A first acquisition module, used to acquire a first model identifier acquired by the first terminal by triggering first model recognition; A sending module, configured to send a second model identification request to a network side device, wherein the second model identification request is used to request the network side device to perform a second model identification of the target AI unit, and the second model identification request carries the first model identifier; The second acquisition module is used to obtain a second model identifier configured by the network side device for the target AI unit.

36. A model recognition device, comprising: A receiving module, configured to receive a first model identification request sent by a first terminal, wherein the first model identification request is used to request the network side device to perform first model identification of a target AI unit; An allocation module, configured to allocate a first model identifier to the target AI unit by performing a first model identification of the target AI unit, wherein the first model identifier satisfies at least one of the following: a globally unique global identifier; a partially globally unique partial global identifier; and is used by the second terminal to send a second model identification request to the network side device; A sending module is used to send the first model identifier to the first terminal.

37. A terminal 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 steps of the model recognition method as described in any one of claims 1 to 8, or implements the steps of the model recognition method as described in any one of claims 9 to 19.

38. A network side device, comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the model recognition method as described in any one of claims 20 to 33 are implemented.

39. A readable storage medium storing a program or instruction, wherein the program or instruction, when executed by a processor, implements the model recognition method as described in any one of claims 1 to 8, or implements the steps of the model recognition method as described in any one of claims 9 to 19, or implements the steps of the model recognition method as described in any one of claims 20 to 33.

Images