A wireless communication system AI model registration method and device
By introducing the AI model registration method in the wireless communication system, the problem of unregistered AI models of terminal devices is solved, the effective management and use of terminal-side models is achieved, and the system performance is improved.
Patent Information
- Application Number
- CN202211380586.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-04
- Publication Date
- 2025-09-09
- Estimated Expiration
- 2042-11-04
AI Technical Summary
The AI model registration and management process of terminal devices in wireless communication systems is not explicitly supported, resulting in the terminal-side AI model being unable to obtain auxiliary information from the base station, limiting the usage space and performance of the AI model.
A wireless communication system AI model registration method is provided, which realizes AI model registration on the terminal side and the network side through an information interaction process, including determining AI model registration instructions, identification and confirmation information, and supports the management and use of multiple models in different scenarios.
The registration and management of terminal-side AI models on the network side are realized, which improves the performance of wireless communication systems and the use effect of AI models.
Smart Images
Figure CN115767570B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of wireless communication technology, and in particular to a method and device for registering an AI model in a wireless communication system. Background Art
[0002] Mobile communication networks contain vast amounts of data resources. Leveraging artificial intelligence (AI) technology to rationally utilize and exploit these 5G data resources can effectively enhance mobile communication systems. The challenges faced by mobile communication systems are complex and diverse, and numerous studies have demonstrated that AI-based algorithms can effectively improve both network and radio performance. Improving mobile system performance using AI has become a key focus in future network design.
[0003] When wireless communication systems use AI technology to enhance positioning performance, they face the problem of how to use the model effectively. Since the AI model registration model needs to integrate the channel information of the positioning terminal and multiple base stations to calculate the position, the training of the AI model also requires a large amount of labeled data. In order to further improve the use scenarios of positioning algorithms based on AI models, it is necessary to consider reducing the computational complexity of the AI model, improving the generalization ability of the model, and reducing the training complexity of the AI model and the size of the training data set. The present invention provides a method and device that supports AI model registration, which can more efficiently use the AI model to realize the positioning function.
[0004] Leveraging AI models to enhance specific functions has become a key focus of wireless system evolution and enhancement. However, due to the inherent characteristics of AI models, their scope and scenarios are often relatively limited, necessitating dynamic model management to ensure AI model performance. Model management involves numerous steps, including model performance monitoring, updates, and delivery. For AI models on the terminal side to participate in model management, they must inform the network side of relevant AI model information. This process is called model registration.
[0005] Existing wireless communication systems do not explicitly support the registration and management of AI models. If terminal devices and network equipment do not support the registration process, the use of AI models on the terminal side will be entirely based on the terminal's implementation and will not be able to obtain auxiliary information from the base station. This will limit the use space and performance of AI model-based algorithms. The use of AI models is strongly dependent on the scenario. An AI model is often only used in specific scenarios. Scenario-related auxiliary information is very helpful for the use of AI models. Network equipment has very complete scenario information about the service area, which supports effective information transmission between network equipment and terminal devices. This will effectively expand the use space and effectiveness of AI model-based algorithms and improve the performance of the entire wireless communication system. Summary of the Invention
[0006] This application proposes a method and device for registering AI models in a wireless communication system. This method addresses the issue of how to effectively register multiple AI models for different functions in a terminal device in a wireless communication system, thereby enabling effective use of AI models in different application scenarios. In particular, when a terminal supports multiple models for a single function, this method can better support the management and use of multiple models in different scenarios, thereby improving the performance of the wireless communication system.
[0007] The registration of AI models requires corresponding processes to support it. The present invention provides an efficient AI model registration method and device, which can realize the rapid model registration function, so that the wireless system can quickly use and manage the terminal side model.
[0008] In a first aspect, the present application proposes a wireless communication system AI model registration method for a wireless communication system, comprising the following steps:
[0009] Determine an AI model registration indication included in the first downlink information, where the AI model registration indication includes an indication of a maximum number of allowed registered models;
[0010] Determine the AI model identifier included in the second uplink information;
[0011] Determine confirmation information included in the third downlink information, where the confirmation information corresponds to at least one of the AI model identifiers.
[0012] Furthermore, the wireless communication system AI model registration method is used for network-side devices and includes the following steps:
[0013] The first information sent includes an AI model registration instruction, wherein the AI model registration instruction includes an indication of a maximum number of allowed registered models;
[0014] The received second information includes an AI model identifier;
[0015] In response to the second information, the third information sent includes confirmation information corresponding to the AI model identifier.
[0016] The wireless communication system AI model registration method is used for a terminal side device and includes the following steps:
[0017] The received first information includes an AI model registration instruction, wherein the AI model registration instruction includes an indication of a maximum number of allowed registered models;
[0018] In response to the first information, the second information sent includes an AI model identifier;
[0019] The received third information includes confirmation information corresponding to the AI model identifier.
[0020] In any embodiment of the first aspect of the present application, the AI model registration indication further includes at least one of the following: a model function quantity indication; a model type quantity indication; a second information feedback time indication; and a second information feedback frequency indication.
[0021] In any embodiment of the first aspect of the present application, the AI model identifier includes a function identifier and a model number, and the number of the function identifiers does not exceed the number of model functions represented in the AI model registration indication; the number of the model numbers does not exceed the number of model types represented in the AI model registration indication.
[0022] In any embodiment of the first aspect of the present application, the number of the AI model identifiers does not exceed the maximum number of allowed registered models indicated in the AI model registration indication.
[0023] In any embodiment of the first aspect of the present application, the second information further includes at least one of the following model characteristics: a model type index, and a request to update training data.
[0024] In any embodiment of the first aspect of the present application, the third information includes the at least one AI model identifier, or the third information includes a bit symbol corresponding to the at least one AI model identifier.
[0025] In the second aspect, an embodiment of the present application proposes a network-side device for implementing the method described in any embodiment of the first aspect of the present application, and at least one module in the network-side device is used for at least one of the following functions: determining the AI model registration indication; sending the first information; receiving the second information; determining the AI model identifier; determining the confirmation information; and sending the third information.
[0026] In the third aspect, an embodiment of the present application proposes a terminal side device for implementing the method described in any embodiment of the first aspect of the present application, and at least one module in the terminal side device is used for at least one of the following functions: receiving the first information, determining the AI model registration indication; determining the AI model identifier; sending the second information; receiving the third information; and determining the confirmation information.
[0027] In a fourth aspect, the present application also proposes a communication device, comprising: a memory, a processor, and a computer program stored on the memory and runnable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described in any one of the embodiments of the first aspect of the present application.
[0028] In a fifth aspect, the present application further proposes a computer-readable medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any embodiment of the first aspect of the present application are implemented.
[0029] In a sixth aspect, the present application also proposes a mobile communication system comprising at least one network side device as described in any embodiment of the present application and / or at least one terminal side device as described in any embodiment of the present application.
[0030] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:
[0031] The method and apparatus provided by the present invention can implement the registration of terminal-side AI models on network-side devices, thereby supporting the management of registered AI models by both network-side and terminal-side devices. The solution provided by the present invention uses streamlined signaling to complete the exchange of terminal-side AI model ID identification and key AI model information, effectively supporting the use of more terminal-side AI models in mobile communication systems, thereby improving overall system performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0033] Figure 1 This is a flow chart of an embodiment of the method of this application;
[0034] Figure 2 This is a flowchart of an embodiment of the method of the present application for a network-side device;
[0035] Figure 3 This is a flow chart of an embodiment of the method of the present application used in a terminal-side device;
[0036] Figure 4 is a schematic diagram of an embodiment of the first information;
[0037] Figure 5 This is a schematic diagram of first information according to another embodiment;
[0038] Figure 6 Sending a location diagram for the first information and the second information;
[0039] Figure 7 This is a schematic diagram of an embodiment of a network-side device;
[0040] Figure 8 is a schematic diagram of an embodiment of a terminal-side device;
[0041] Figure 9This is a schematic structural diagram of a network-side device according to another embodiment of the present invention;
[0042] Figure 10 It is a block diagram of a terminal side device according to another embodiment of the present invention. DETAILED DESCRIPTION
[0043] To make the purpose, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0044] The following describes in detail the technical solutions provided by various embodiments of the present application in conjunction with the accompanying drawings.
[0045] This application proposes a wireless communication system AI model registration method for a wireless communication system, comprising the following steps:
[0046] Determine an AI model registration indication included in the first downlink information, where the AI model registration indication includes an indication of a maximum number of allowed registered models;
[0047] Determine the AI model identifier included in the second uplink information;
[0048] Determine confirmation information included in the third downlink information, where the confirmation information corresponds to at least one of the AI model identifiers.
[0049] It should be noted that the above steps are used for network entities of a wireless communication system, including terminal-side devices, network-side devices or other intermediate devices; the above steps can also be used for service devices that provide information processing for the network entity devices; the above steps can also be used for any device, system, subsystem, circuit, chip or software entity that provides information reception, transmission, identification, and processing for terminal-side devices or network-side devices.
[0050] In any embodiment of the present application, the AI model registration indication further includes at least one of the following: a model function quantity indication; a model type quantity indication; a second information feedback time indication; and a second information feedback frequency indication.
[0051] In any embodiment of the present application, the AI model identifier includes a function identifier and a model number, and the number of the function identifiers does not exceed the number of model functions represented in the AI model registration indication; the number of the model numbers does not exceed the number of model types represented in the AI model registration indication.
[0052] In any embodiment of the present application, the number of the AI model identifiers does not exceed the maximum number of allowed registered models indicated in the AI model registration indication.
[0053] In any embodiment of the present application, the second information further includes at least one of the following model characteristics: a model type index, and a request to update training data.
[0054] In any embodiment of the present application, the third information includes the at least one AI model identifier, or the third information includes a bit symbol corresponding to the at least one AI model identifier.
[0055] In a wireless communication system, the terminal side device and the network side device perform a series of information interactions to complete the registration of the AI model. The specific process of information interaction is as follows: Figure 1 As shown, it includes steps 110 to 130. The network side device first needs to trigger the terminal side device to report the AI model that needs to be managed through the AI model registration indication information. After the terminal side device receives the AI model registration indication information sent by the network side device, the terminal side device generates an AI model identifier (ID) and the required feedback information based on the content of the indication information and the AI model information that needs to be registered, and completes the AI model registration information reporting at the indicated location. After receiving the registration message, the network side device determines the model that can be registered, and notifies the terminal side device of the completed registered AI model through the AI model registration success information.
[0056] Figure 2 This is a flowchart of an embodiment of the method of this application for network-side devices. The wireless communication system AI model registration method is used for network-side devices and includes the following steps:
[0057] Step 210: Send first information, where the first information includes an AI model registration instruction, and the AI model registration instruction includes an indication of the maximum number of registered models allowed.
[0058] The network-side device sends the first information after determining the AI model registration indication. The network-side device sends the AI model registration indication information to the terminal through the first information. The first information content may include the main functions that the registered model can support, the maximum number of registered models allowed (AIModel_max), the number of model functions (AIModel_Fun); the number of model types (Modeltype_max), the feedback time indication (feedback time) and the feedback frequency indication (feedback frequency) indicating the feedback location of the second information.
[0059] The number of model types indicates the maximum number of registered models for each function.
[0060] Step 220: Receive second information, where the second information includes an AI model identifier; the network-side device determines the AI model identifier.
[0061] The second information comes from the terminal device, and the second information responds to the indication of the first information and is transmitted at the time position and frequency domain position determined by the second information feedback time indication and the second information feedback frequency indication.
[0062] Typically, the AI model registration information in the second information includes multiple AI model identification (ID) information, and each ID information can be composed of two parts: function identification and model number, where the number of model numbers does not exceed the maximum number of registered models for each function in the first information, and the number of AI model ID information in the second information does not exceed the maximum number of allowed registered models in the first information.
[0063] In addition to multiple model IDs, the second information can report multiple bits of model characteristics for each model. The number of bits and their specific meaning are agreed upon by the terminal and network devices. Typical agreed information includes model classification and whether the dataset needs to be updated.
[0064] Step 230: In response to the second information, send third information, where the third information includes confirmation information corresponding to the AI model identifier.
[0065] The network-side device sends the third information after confirming the confirmation information. The network-side device sends the AI model registration success information to the terminal-side device through the third information. The third information may include the model ID information fed back in the second information, or the third information may consist of multiple bits, each bit corresponding to the model ID fed back in the second information, indicating whether the model ID is successfully registered.
[0066] Figure 3 This is a flowchart of an embodiment of the method of this application for a terminal side device. The wireless communication system AI model registration method is used for a terminal side device, comprising the following steps:
[0067] Step 310: Receive first information, where the first information includes an AI model registration instruction, and the AI model registration instruction includes an instruction for the maximum number of registered models allowed.
[0068] After receiving the first information, the terminal device determines the AI model registration indication.
[0069] Step 320: In response to the first information, send second information, where the second information includes an AI model identifier.
[0070] First, the terminal side device determines the AI model identifier (and also determines the model characteristic information), and then sends the second information.
[0071] The terminal side device sends the AI model registration information to the network side device at the time position and frequency position indicated by the first information through the second information.
[0072] In the second information, the AI model identifier includes a function identifier and a model number, and the number of the function identifiers does not exceed the number of model functions indicated in the AI model registration indication; the number of the model numbers does not exceed the number of model types indicated in the AI model registration indication.
[0073] Preferably, the number of the AI model identifiers does not exceed the maximum number of allowed registered models indicated in the AI model registration indication.
[0074] Preferably, the second information further includes at least one of the following model characteristics: a model type index, and a request to update training data.
[0075] The second information is composed of a number of bits. For example, all-0 feedback indicates that no model is registered on the terminal side.
[0076] Step 330: Receive third information, the third information including confirmation information corresponding to the AI model identifier. The confirmation information indicates whether the corresponding AI model identifier is successfully registered. The terminal device determines the confirmation information based on the received third information.
[0077] Figure 4 This is a schematic diagram of an embodiment of the first information.
[0078] In this embodiment, the first information does not display the AI model function, but only indicates the maximum number of registered models (AImodel_max). The figure shows a schematic diagram indicating only the maximum number of registered models (AImodel_max), where AImodel_max is '1000', corresponding to a maximum of 16 registered models.
[0079] After receiving the first information, the terminal feeds back the second information at the position indicated by the first information (Feedback time and Feedback frequency). The specific time and frequency domain positions are shown in Figure 6. The second information feedback can add a description of the model itself, that is, model characteristic information, to the feedback AI model identifier (ID). The description content is agreed upon by the terminal and the network. For example, 2 bits are used to indicate the AI model function, 2 bits are used to indicate the AI model type, and 1 bit is used to indicate whether the model requires network-side data for training and updating. '0001 00011' indicates a channel information feedback model with a model ID of 0001, a model type of DNN, and requires some data for training and updating.
[0080] Figure 5 This is a schematic diagram of first information according to another embodiment.
[0081] In this embodiment, the network side device sends the first information via PDCCH. The first information content includes a 3-bit function indication (AImodel_Fun) and a 4-bit maximum number of allowed registered models (AImodel_max), a 3-bit maximum number of registered model types for each function (Modeltype_max), and the second information feedback time indication (Feedback time) and feedback frequency indication (Feedback frequency) information, such as Figure 5 As shown. The three bits and two 1s of AImodel_Fun indicate that the network device supports two of the three AI model functions for model management. The AI model function corresponding to each bit can be pre-agreed by the network device and the terminal device. For example, the three bits indicate whether the "channel information feedback" function, the "beam management" function, and the "positioning" function are supported. Specifically, for example, 110 indicates support for AI model management of the "channel information feedback" and "beam management" functions. The four bits of AImodel_max indicate the maximum number of models allowed to be registered. For example, 1000 indicates that a maximum of 16 models can be registered. The three bits of Modeltype_max indicate the maximum number of models that can be registered for a functional model. For example, 100 indicates that a maximum of 8 models of a type can be registered. Feedbacktime is feedback 16 time slots after the current PDCCH transmission time slot. '11' indicates the 16th time slot. The meaning of each bit is pre-agreed by the terminal and the network device. For example, 00, 01, 11, 11 indicate the 4th, 8th, 12th, and 16th time slots. Feedback frequency '100' indicates that RB numbered 8 carries the second information. The meaning of each bit combination is pre-agreed between the terminal and the network side device. For example, 000 to 111 indicate the 1st to 16th RBs.
[0082] Figure 6 A location diagram is sent for the first information and the second information.
[0083] After receiving the first information, the terminal feeds back the second information at the position indicated by the first information (Feedback time and Feedback frequency). The relationship between the sending position of the second information and the first information is shown in the figure. Figure 5 The first information content shown is that the network side device supports two functions (AImodel_Fun=110) and a maximum of 16 AI models are managed (AImodel_max=1000). The model ID in the second information consists of 6 bits, the first two bits are the model function identifier, indicating the model function selection, and the following 4 bits are the model number. Furthermore, the maximum number for each model (Modeltype_max=100) does not exceed 8. When the terminal is a channel information feedback registration model, there are multiple ways to choose:
[0084] Method 1: Single model ID feedback. The terminal selects a function, such as a model registration for "channel information feedback", and feedbacks a model ID number such as '100001'
[0085] Method 2: Multiple model ID feedback. The terminal registers multiple models for one function or multiple functions. Taking two functions and two models registered for each as an example, the second information will include '100001 10002010001 010002', indicating that there are four model identifiers in total.
[0086] Method 3: The model ID feedback is accompanied by some descriptions of the model. The description of the model is agreed upon by the terminal and the network. Two bits represent the AI model type, such as CNN, DNN, LSTM, or Transformer, and one bit indicates whether the model requires network-side data for training and updating. When the second information registers a model, there will be three bits in addition to the model ID information. For example, '100001 011' represents the first model of the channel information feedback, where '100001' is the AI model identifier, '10' indicates a request to register the model function identified by the first bit, '0001' is the model number, and '011' is the model feature information, where '01' is the model type index, specifically indicating that the model type is DNN. The last bit '1' is the update training data request identifier, indicating that the terminal device needs some data for training and updating.
[0087] Figure 7 A schematic diagram of an embodiment of a network-side device.
[0088] An embodiment of the present application also proposes a network-side device, using the method of any one of the embodiments of the present application, wherein at least one module in the network-side device is used for at least one of the following functions: determining the AI model registration indication; sending the first information; receiving the second information; determining the AI model identifier; determining the confirmation information; and sending the third information.
[0089] To implement the above technical solution, the present application also proposes a network side device 400, comprising a network sending module 401, a network receiving module 403, and a network determining module 402 connected to each other;
[0090] The network sending module is used to send the first information and the second information;
[0091] The network receiving module is configured to receive the second information according to the second information feedback time indication and the second information feedback frequency indication in the first information;
[0092] The network determination module is used to determine the AI model registration indication, the AI model identifier, the model characteristic information, and the confirmation information.
[0093] The specific methods for implementing the functions of the network sending module, network determination module, and network receiving module are as described in the various method embodiments of this application and will not be repeated here.
[0094] Figure 8 It is a schematic diagram of an embodiment of a terminal side device.
[0095] The present application also proposes a terminal side device, using the method of any embodiment of the present application, at least one module in the terminal side device is used to implement at least one of the following functions: receiving the first information, determining the AI model registration indication; determining the AI model identifier; sending the second information; receiving the third information; and determining the confirmation information.
[0096] To implement the above technical solution, the present application proposes a terminal side device 500, comprising a terminal determination module 502, a terminal sending module 501, and the terminal receiving module 503 connected to each other;
[0097] The terminal receiving module is configured to receive the first information and the third information in downlink.
[0098] The terminal determination module is configured to determine the AI model registration indication, the AI model identifier, the model characteristic information, and the confirmation information;
[0099] The terminal sending module is configured to send the second information according to the second information feedback time indication and the second information feedback frequency indication in the first information;
[0100] The specific methods for implementing the functions of the terminal sending module, the terminal determining module, and the terminal receiving module are as described in the various method embodiments of this application and will not be repeated here.
[0101] The terminal side device described in this application may refer to a personal mobile terminal side device.
[0102] Figure 9 A schematic structural diagram of a network side device according to another embodiment of the present invention is shown. As shown in the figure, the network side device 600 includes a processor 601, a wireless interface 602, and a memory 603. The wireless interface may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. The wireless interface implements the communication function with the terminal side device, processes wireless signals through receiving and transmitting devices, and the data carried by the signals is communicated with the memory or processor via an internal bus structure. The memory 603 contains a computer program for executing any one of the embodiments of the present application, and the computer program runs or changes on the processor 601. When the memory, processor, and wireless interface circuit are connected through a bus system. The bus system includes a data bus, a power bus, a control bus, and a status signal bus, which will not be described here.
[0103] Figure 10 This is a block diagram of a terminal device according to another embodiment of the present invention. Terminal device 700 includes at least one processor 701, memory 702, a user interface 703, and at least one network interface 704. The various components in terminal device 700 are coupled together via a bus system. The bus system is used to enable communication between these components. The bus system includes a data bus, a power bus, a control bus, and a status signal bus.
[0104] The user interface 703 may include a display, a keyboard, or a pointing device, such as a mouse, a trackball, a touch pad, or a touch screen.
[0105] Memory 702 stores executable modules or data structures. The memory may store an operating system and application programs. The operating system includes various system programs, such as a framework layer, a core library layer, and a driver layer, for implementing various basic services and processing hardware-based tasks. Application programs include various application programs, such as media players and browsers, for implementing various application services.
[0106] In an embodiment of the present invention, the memory 702 contains a computer program for executing any one of the embodiments of the present application, and the computer program is run or changed on the processor 701 .
[0107] Memory 702 includes a computer-readable storage medium. Processor 701 reads information from memory 702 and, in conjunction with its hardware, performs the steps of the above-described method. Specifically, the computer-readable storage medium stores a computer program that, when executed by processor 701, implements the steps of any of the above-described method embodiments.
[0108] The processor 701 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the method of the present application may be completed by hardware integrated logic circuits in the processor 701 or by instructions in the form of software. The processor 701 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, an off-the-shelf programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention may be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present invention may be directly implemented as being executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in the decoding processor.
[0109] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. In a typical configuration, the device of the present application includes one or more processors (CPUs), an input / output user interface, a network interface, and a memory.
[0110] Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0111] Therefore, the present application also provides a computer-readable medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the method described in any embodiment of the present application are implemented. For example, the memory 603, 702 of the present invention may include non-permanent memory, random access memory (RAM) and / or non-volatile memory in a computer-readable medium, such as read-only memory (ROM) or flash RAM.
[0112] based on Figures 7-10 In addition to the embodiments of the present application, the present application also proposes a mobile communication system, comprising at least one embodiment of any terminal side device in the present application and / or at least one embodiment of any network side device in the present application.
[0113] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.
[0114] It should also be noted that the terms "first", "second", etc. in this application are used to distinguish multiple objects with the same name and do not have any meaning of order or size unless specifically stated.
[0115] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.
Claims
1. A method for registering an AI model in a wireless communication system. When a terminal device in the wireless communication system has multiple AI models for different functions, the method registers the model in the system device, supports the management and use of multiple models in different scenarios, and is characterized by: The following steps are involved: Determine an AI model registration indication included in the downlink first information, where the AI model registration indication includes an indication of a maximum number of allowed registered models and a second information feedback time indication and / or a second time feedback frequency indication; Determine the AI model identifier included in the second information in the uplink; the second information also includes a request to update training data; Determine confirmation information included in the third downlink information, where the confirmation information corresponds to at least one of the AI model identifiers.
2. A wireless communication system AI model registration method, used for network-side devices, in which when a terminal device in a wireless communication system has multiple sets of AI models for different functions, the method registers the model in the system device, supports the management and use of multiple models in different scenarios, and is characterized by: The following steps are involved: The sent downlink first information includes an AI model registration indication, wherein the AI model registration indication includes an indication of a maximum number of allowed registered models, and further includes an indication of a second information feedback time and / or a second time feedback frequency indication; The received uplink second information includes an AI model identifier and a request to update training data; In response to the second information, the downlink third information sent includes confirmation information corresponding to the AI model identifier.
3. A wireless communication system AI model registration method, used for terminal side devices, in which when a terminal device in a wireless communication system has multiple sets of AI models for different functions, registration is performed on the system device, supporting the management and use of multiple models in different scenarios, characterized in that: The following steps are involved: The received downlink first information includes an AI model registration indication, wherein the AI model registration indication includes an indication of a maximum number of allowed registered models, and further includes an indication of a second information feedback time and / or an indication of a second time feedback frequency; In response to the first information, the uplink second information sent includes the AI model identifier and the request to update the training data; The received downlink third information includes confirmation information corresponding to the AI model identifier.
4. The wireless communication system AI model registration method according to any one of claims 1 to 3, characterized in that: The AI model registration instruction also includes at least one of the following: Model function quantity indication; model type quantity indication; second information feedback time indication; second information feedback frequency indication.
5. The wireless communication system AI model registration method according to any one of claims 1 to 3, characterized in that: The AI model identifier includes a function identifier and a model number. The number of function identifiers does not exceed the number of model functions indicated in the AI model registration instruction; The number of the model numbers does not exceed the number of model types indicated in the AI model registration indication.
6. The wireless communication system AI model registration method according to any one of claims 1 to 3, characterized in that: The number of the AI model identifiers does not exceed the maximum number of allowed registered models indicated in the AI model registration indication.
7. The wireless communication system AI model registration method according to any one of claims 1 to 3, characterized in that: The second information further includes at least one of the following model characteristics: Model type index, update training data request.
8. The wireless communication system AI model registration method according to any one of claims 1 to 3, characterized in that: The third information includes the at least one AI model identifier, or the third information includes a bit symbol corresponding to the at least one AI model identifier.
9. A network-side device, used to implement the method according to any one of claims 1 to 8, characterized in that: At least one module in the network-side device is configured to perform at least one of the following functions: Determine the AI model registration indication; send the first information; receive the second information; determine the AI model identifier; determine the confirmation information; and send the third information.
10. A terminal-side device, used to implement the method according to any one of claims 1 to 8, characterized in that: At least one module in the terminal-side device is used for at least one of the following functions: Receive the first information, determine the AI model registration indication; determine the AI model identifier; send the second information; receive the third information; and determine the confirmation information.
11. A communication device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method according to any one of claims 1 to 8.
12. A computer-readable medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 8.
13. A mobile communication system comprising at least one network-side device as claimed in claim 9 and at least one terminal-side device as claimed in claim 10.