Information transmission method and device

Abstract

The embodiment of the invention provides an information transmission method and device. The method comprises: a terminal device sends indication information related to AI / ML features and / or functions to a network device.

Description

Information transmission method and device Technical Field

[0001] The embodiments of the present application relate to the field of communication technologies. Background Art

[0002] As low-frequency spectrum resources become scarce, millimeter-wave (mmWave) bands, offering greater bandwidth, have become a key frequency band for 5G New Radio (NR) systems. Due to their shorter wavelengths, mmWaves exhibit different propagation characteristics than traditional low-frequency bands, such as higher propagation loss and poor reflection and diffraction performance. Consequently, larger antenna arrays are typically used to form shaped beams with greater gain, overcome propagation loss, and ensure system coverage.

[0003] With the advancement of artificial intelligence (AI) and machine learning (ML) technologies, applying AI / ML to wireless communications to overcome the difficulties of traditional methods has become a current technological trend. The application of AI / ML models to wireless communication systems, particularly in air interface transmission, is a new technology in the 5G-Advanced and 6G phases.

[0004] For example, for reporting Channel State Information (CSI), using the Autoencoder network in deep learning on the terminal device side, the AI ​​encoder (AI encoder) encodes / compresses the CSI, and the AI ​​decoder (AI decoder) on the network device side decodes / decompresses the CSI, which can reduce feedback overhead. For another example, for beam management, using AI / ML models to predict the spatially optimal beam pair based on the results of a small number of beam measurements can reduce system load and latency.

[0005] It should be noted that the above introduction to the technical background is merely intended to provide a clear and complete description of the technical solutions of this application and facilitate understanding by those skilled in the art. Simply because these solutions are described in the background technology section of this application, it should not be assumed that the above technical solutions are well known to those skilled in the art.

[0006] Summary of the Invention

[0007] In 5G and 6G systems, a wide range of terminal technical features (UE features) are defined. Not all features are required to be supported by the terminal. The network understands the UE's feature support by querying and reporting the terminal's capabilities, namely UE capabilities. This allows the network to understand the UE's capabilities and provides a basis for subsequent scheduling, configuration, and communication with the terminal.

[0008] According to the current technological trends of 5G-Advanced and 6G, a terminal device may support multiple AI / ML features. A feature can be described more specifically by the functional parameters of a predefined feature group. The features within each feature group can be regarded as a sub-feature, or a functionality. Another possibility is to send the configuration of predefined features or feature groups on the network side to the terminal device, thereby activating or enabling a terminal-side function. For an AI / ML feature or an AI / ML function enabled or activated by the network side, the network side monitors the performance of the AI / ML feature or function.

[0009] However, the inventors have found that the network-side response to monitoring the performance of AI / ML features or functions is slow or inaccurate, thus making it impossible to fully utilize AI / ML functions to improve communication quality.

[0010] To address at least one of the above problems, embodiments of the present application provide an information transmission method and apparatus.

[0011] According to a first aspect of an embodiment of the present application, an information transmission method is provided, the method comprising: a terminal device sending indication information related to AI / ML features and / or functions to a network device.

[0012] According to a second aspect of an embodiment of the present application, an information transmission method is provided, the method comprising: a network device receiving indication information related to AI / ML features and / or functions sent by a terminal device.

[0013] According to a third aspect of an embodiment of the present application, an information transmission device is provided, which is arranged in a terminal device, and the device includes: a first sending unit, which sends indication information related to AI / ML features and / or functions to a network device.

[0014] According to a fourth aspect of an embodiment of the present application, an information transmission device is provided, which is arranged in a network device, and includes: a second receiving unit, which receives indication information related to AI / ML features and / or functions sent by a terminal device.

[0015] According to the fifth aspect of the embodiment of the present application, a computer-readable program is provided, wherein when the program is executed in an information transmission device or a terminal device, the program causes the information transmission device or the terminal device to execute the information transmission method described in the first aspect of the embodiment of the present application.

[0016] According to the sixth aspect of the embodiment of the present application, a computer-readable program is provided, wherein when the program is executed in an information transmission device or a network device, the program causes the information transmission device or the network device to execute the information transmission method described in the second aspect of the embodiment of the present application.

[0017] According to the seventh aspect of the embodiment of the present application, a storage medium storing a computer-readable program is provided, wherein the computer-readable program enables an information transmission device or a terminal device to execute the information transmission method described in the first aspect of the embodiment of the present application.

[0018] According to an eighth aspect of the embodiments of the present application, a storage medium storing a computer-readable program is provided, wherein the computer-readable program enables an information transmission apparatus or a network device to execute the information transmission method described in the second aspect of the embodiments of the present application.

[0019] One of the beneficial effects of the embodiments of the present application is that: the terminal device sends indication information related to the AI / ML features and / or functions to the network device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of the AI / ML features and / or functions, thereby making full use of the AI / ML functions to improve communication quality and improve communication system performance.

[0020] Another beneficial effect of the embodiments of the present application is that the implementation of the AI / ML model of the terminal device is invisible to the network device, and the terminal device can observe and monitor the terminal-side model in a timely and accurate manner. The results of these monitoring can more accurately and timely reflect the performance of the functional level. The terminal device sends indication information or request information to the network device regarding the changes in the terminal-side model or its performance evaluation of the functional level, which can help the network device operate the function. Compared with the solution that does not use the terminal device to send the information, the AI / ML function can be used in a timely and reliable manner to improve the communication quality and improve the performance of the communication system.

[0021] With reference to the following description and accompanying drawings, specific embodiments of the present application are disclosed in detail, indicating the manner in which the principles of the present application can be employed. It should be understood that the embodiments of the present application are not limited in scope. Within the spirit and scope of the appended claims, the embodiments of the present application include many variations, modifications and equivalents.

[0022] Features described and / or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments, combined with features in other embodiments, or substituted for features in other embodiments.

[0023] It should be emphasized that the term "include / comprising" when used herein refers to the presence of features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps or components. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The elements and features described in one figure or one embodiment of the present application can be combined with the elements and features shown in one or more other figures or embodiments. In addition, in the accompanying drawings, similar reference numerals represent corresponding parts in several figures and can be used to indicate corresponding parts used in more than one embodiment.

[0025] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application;

[0026] FIG2 is an interaction diagram between a network device and a terminal device;

[0027] FIG3 is a schematic diagram of a device having an AI / ML functional module;

[0028] FIG4 is another diagram of interaction between a network device and a terminal device;

[0029] FIG5 is a schematic diagram of an information transmission method according to an embodiment of the present application;

[0030] FIG6 is a modular schematic diagram of a network device and a terminal device according to an embodiment of the present application;

[0031] FIG7 is an interaction diagram of a network device and a terminal device in application scenario 1 of an embodiment of the present application;

[0032] FIG8 is a comparison diagram of the interaction process between the method of the embodiment of the present application and the existing method;

[0033] FIG9 is an interaction diagram of a network device and a terminal device in application scenario 2 of an embodiment of the present application;

[0034] FIG10 is an interaction diagram of a network device and a terminal device in application scenario 3 of an embodiment of the present application;

[0035] FIG11 is an interaction diagram of a network device and a terminal device in application scenario 4 of an embodiment of the present application;

[0036] FIG12 is an interaction diagram of a network device and a terminal device in application scenario 5 of an embodiment of the present application;

[0037] FIG13 is another schematic diagram of the information transmission method according to an embodiment of the present application;

[0038] FIG14 is a schematic diagram of a method for verifying model performance according to an embodiment of the present application;

[0039] FIG15 is another schematic diagram of a method for verifying model performance according to an embodiment of the present application;

[0040] FIG16 is another schematic diagram of the method for verifying model performance according to an embodiment of the present application;

[0041] FIG17 is a schematic diagram of an AI / ML model identification method according to an embodiment of the present application;

[0042] FIG18 is another schematic diagram of the AI / ML model identification method according to an embodiment of the present application;

[0043] FIG19 is another schematic diagram of the AI / ML model identification method according to an embodiment of the present application;

[0044] FIG20 is a schematic diagram of an information transmission device according to an embodiment of the present application;

[0045] FIG21 is another schematic diagram of the information transmission device according to an embodiment of the present application;

[0046] FIG22 is a schematic block diagram of a system structure of a terminal device according to an embodiment of the present application;

[0047] Figure 23 is a schematic block diagram of the system structure of the network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] The above and other features of the present application will become apparent through the following description with reference to the accompanying drawings. In the description and the accompanying drawings, specific embodiments of the present application are disclosed in detail, which illustrate some embodiments in which the principles of the present application can be adopted. It should be understood that the present application is not limited to the described embodiments. On the contrary, the present application includes all modifications, variations and equivalents that fall within the scope of the appended claims.

[0049] In the embodiments of the present application, the terms "first", "second", etc. are used to distinguish different elements from the name, but do not indicate the spatial arrangement or temporal order of these elements, and these elements should not be limited by these terms. The term "and / or" includes any one and all combinations of one or more of the associated listed terms. The terms "comprising", "including", "having", etc. refer to the presence of the stated features, elements, components or components, but do not exclude the presence or addition of one or more other features, elements, components or components.

[0050] In the embodiments of this application, the singular forms "a," "the," etc. include plural forms and should be broadly understood to mean "a" or "a type" rather than being limited to "one." Furthermore, the term "said" should be understood to include both singular and plural forms, unless the context clearly indicates otherwise. Furthermore, the term "according to" should be understood to mean "at least in part based on...", and the term "based on" should be understood to mean "at least in part based on...", unless the context clearly indicates otherwise.

[0051] In the embodiments of the present application, the term "communication network" or "wireless communication network" may refer to a network that complies with any of the following communication standards, such as Long Term Evolution (LTE), enhanced Long Term Evolution (LTE-A, LTE-Advanced), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), etc.

[0052] Furthermore, communication between devices in the communication system may be carried out according to communication protocols of any stage, for example, including but not limited to the following communication protocols: 1G (generation), 2G, 2.5G, 2.75G, 3G, 4G, 4.5G and 5G, 5G-Advanced, New Radio (NR), future 6G, etc., and / or other communication protocols currently known or to be developed in the future.

[0053] In the embodiments of the present application, the term "network device" refers to, for example, a device in a communication system that connects a terminal device to the communication network and provides services to the terminal device. Network devices may include, but are not limited to, the following devices: base station (BS), access point (AP), transmission reception point (TRP), broadcast transmitter, mobile management entity (MME), gateway, server, radio network controller (RNC), base station controller (BSC), etc.

[0054] Among them, base stations may include but are not limited to: NodeB (NodeB or NB), evolved NodeB (eNodeB or eNB) and 5G base station (gNB), central processing unit (CU, Centralized Unit), distributed processing unit (DU, Distributed Unit), etc., and may also include remote radio head (RRH, Remote Radio Head), remote radio unit (RRU, Remote Radio Unit), relay or low-power node (such as femeto, pico, etc.). The term "base station" can include some or all of their functions. Each base station can provide communication coverage for a specific geographical area. The term "cell" can refer to a base station and / or its coverage area, depending on the context in which the term is used.

[0055] In the embodiments of the present application, the term "user equipment" (UE) or "terminal equipment" (TE) refers to, for example, a device that accesses a communication network through a network device and receives network services. A terminal device can be fixed or mobile and may also be referred to as a mobile station (MS), a terminal, a subscriber station (SS), an access terminal (AT), a station, and so on.

[0056] Among them, terminal devices may include but are not limited to the following devices: cellular phones, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, machine-type communication devices, laptop computers, cordless phones, smart phones, smart watches, digital cameras, etc.

[0057] For another example, in scenarios such as the Internet of Things (IoT), the terminal device can also be a machine or device for monitoring or measurement, including but not limited to: machine type communication (MTC) terminal, vehicle-mounted communication terminal, device-to-device (D2D) terminal, machine-to-machine (M2M) terminal, and so on.

[0058] In addition, the term "network side" or "network device side" refers to one side of the network, which can be a base station or one or more network devices as described above. The term "user side" or "terminal side" or "terminal device side" refers to the user or terminal side, which can be a UE or one or more terminal devices as described above. Unless otherwise specified herein, "device" can refer to either network equipment or terminal equipment.

[0059] The following describes the scenarios of the embodiments of the present application through examples, but the present application is not limited thereto.

[0060] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application, schematically illustrating a situation using a terminal device and a network device as an example. As shown in FIG1 , a communication system 100 may include a network device 101 and a terminal device 102. For simplicity, FIG1 illustrates only one terminal device and one network device as an example, but the embodiments of the present application are not limited thereto.

[0061] In the embodiment of the present application, existing services or future services can be transmitted between the network device 101 and the terminal device 102. For example, these services may include, but are not limited to, enhanced mobile broadband (eMBB), massive machine type communication (mMTC), and ultra-reliable and low-latency communication (URLLC), etc.

[0062] In the embodiments of the present application, the high-layer signaling may be, for example, radio resource control (RRC) signaling; for example, an RRC message, including, for example, an MIB, system information, or a dedicated RRC message; or an RRC information element (RRC IE). The high-layer signaling may also be, for example, MAC (Medium Access Control) signaling; or a MAC control element (MAC CE). However, the present application is not limited thereto.

[0063] In an embodiment of the present application, the network device understands the feature support status of the terminal device through UE capability query and reporting, so that the network device may understand the capabilities of the terminal device, providing a basis for its subsequent scheduling, configuration and communication with the terminal device.

[0064] Figure 2 is an interaction diagram between a network device and a terminal device. As shown in Figure 2, the network device sends a UE capability query request, such as UECapabilityEnquiry, to the terminal device (UE), and the terminal device reports its UE capability information, such as UECapabilityInformation, to the network device.

[0065] For devices with AI / ML functional modules, such as terminal devices, this feature may not be a mandatory feature for the terminal device. Based on current 5G-Advanced and 6G technology trends, a terminal device may support multiple AI / ML features. A feature can be more specifically described by the functional parameters of a predefined feature group.

[0066] In an embodiment of the present application, for a certain use case or communication function, such as CSI feedback, CSI prediction, beam management, beam prediction, positioning, mobility management, etc., each can be classified as a feature, such as CSI feedback AI / ML feature, CSI prediction AI / ML feature, beam management AI / ML feature, positioning AI / ML feature, etc.

[0067] In the embodiment of the present application, each feature can be further subdivided into feature groups. The features in each group can be regarded as a sub-feature, or a functionality.

[0068] For example, an AI / ML feature or a feature group contains one or more AI / ML functionalities.

[0069] The AI / ML function of the terminal device may also be reported by the network device based on the capabilities of the terminal device, and configured to the terminal device based on the capabilities and judgment of the network device, thereby configuring one or more capabilities of the terminal device or activating one or more capabilities of the terminal device.

[0070] The AI / ML function of the terminal device may also be configured to the terminal device by the network device according to the function report of the terminal device, combined with the capabilities and judgment of the network device, thereby configuring one or more capabilities of the terminal device, or activating one or more capabilities of the terminal device. The terminal function reporting here is different from the terminal capability reporting process, which corresponds to the short-term or temporary or current AI / ML function information reported to the base station by the terminal device based on its capability characteristics or function support in a short-term or temporary situation or in the current situation.

[0071] In the embodiment of the present application, a function may also be referred to as a sub-feature, and different sub-features are distinguished by corresponding parameters or application conditions. Different sub-features or functions may have an ID or a predefined corresponding index.

[0072] In this way, different features can be distinguished by different feature identifiers, and different sub-features or functions under a feature can be distinguished by different sub-feature IDs or functionality IDs. Alternatively, different functions can be reported by network devices and configured for terminal devices based on the capabilities of the terminal devices.

[0073] As mentioned above, an AI / ML feature or a feature group includes one or more AI / ML functionalities. For example, for a CSI prediction feature, this feature may further include a low-speed prediction function and a high-speed prediction function.

[0074] The list of these sub-features or functions and their related parameters can be predefined to ensure a common understanding between the network side and the terminal side when communicating capability information.

[0075] To achieve a certain function, the terminal device can be specifically implemented by one AI / ML model or multiple AI / ML models.

[0076] In some embodiments, multiple functions can be implemented by one AI / ML model, which depends on the specific implementation of the terminal device, and the relevant implementation methods are not necessarily informed to the network device.

[0077] In the embodiments of this application, from the perspective of a logical model, a model that logically corresponds to a function is generally defined as a logical model. A logical model can be specifically implemented by one or more physical models. For ease of description, the embodiments of this application do not further distinguish between logical models and physical models. In other words, the models in the embodiments of this application can be either logical models or physical models.

[0078] In an embodiment of the present application, when a function is implemented by multiple models, different models may correspond to different scenarios, different sites, different cells, different configurations, different application conditions, etc., which are determined by the training and development process of the model.

[0079] Figure 3 is a schematic diagram of a device with an AI / ML functional module, which may be a terminal device or a network device. The terminal device or network device has an AI / ML functional module. For example, as shown in Figure 3, the terminal device or network device has a unit of a certain AI / ML Feature / Feature Group, and within the unit of the AI / ML Feature / Feature Group, there is at least one functional unit, such as a first functional unit and a second functional unit. For example, as shown in Figure 3, the first function is implemented by at least one model among model a, model b, model c..., and the second function is implemented by at least one model among model x, model y, model z... In some embodiments, the above-mentioned functional unit may also include multiple sub-functions, in which case a functional unit can be regarded as a functional group.

[0080] In an embodiment of the present application, the network device queries the terminal device for AI / ML-related capabilities, and the terminal device reports the corresponding capabilities based on the query.

[0081] Figure 4 is another diagram of the interaction between a network device and a terminal device. As shown in Figure 4, the network device queries the terminal device for AI / ML-related capabilities, i.e., UE AI / ML capability query. Based on the query, the terminal device reports to the network device, i.e., UE AI / ML capability report.

[0082] Based on the UE AI / ML capability report, the network device further sends a control message to activate or enable a certain AI / ML feature or feature group, or a specific functionality, on the terminal side. As shown in Figure 4, the network device activates the UE AI / ML feature and / or functionality on the terminal device.

[0083] In some embodiments, activation can be performed by feature ID, sub-feature ID, function ID, or by configuring a specific function, for example, by configuring the AI / ML IE in the CSI reporting configuration to activate a specific function.

[0084] Alternatively, the ID of the configuration or resource (or resource set) configuration reported by CSI is combined with the AI / ML activation indication to indicate the activation of a certain function, or equivalently, to activate the AI / ML implementation corresponding to the configuration.

[0085] After receiving the relevant activation information, the terminal device enables the corresponding AI / ML model to perform measurement and reporting based on its implementation.

[0086] However, the inventors have discovered that there is a problem in that the specific model information used on the terminal side does not need to be notified to the network side.

[0087] For example, corresponding to the first functionality, the number of models used by the terminal device and which specific model or models are used, for example, whether one model, such as model a, is used, or model a and model b are used, or model a, model b and model c are used, etc., are specific implementation information of the terminal and generally do not need to be informed to the network side.

[0088] For an AI / ML feature or function enabled or activated on the network side, the network side needs to perform basic performance monitoring of the feature or function. At the same time, when the terminal side uses a specific model to implement this activated feature or function, model-level performance monitoring or input-output-related monitoring is also required.

[0089] Furthermore, when a certain model or a certain group of models used by the terminal device is not suitable for the current application scenario or application conditions, the terminal device may autonomously switch the model or model group.

[0090] The terminal can switch models without informing the network. However, due to a lack of information about the terminal model switch, the network's AI / ML function monitoring system may send commands to the terminal device to deactivate or stop AI / ML features or functions if it determines that AI / ML performance is poor. This is actually inappropriate because the terminal device may perform very well after the model switch.

[0091] Embodiments of the first aspect

[0092] An embodiment of the present application provides an information transmission method, which is described from the perspective of a terminal device.

[0093] FIG5 is a schematic diagram of an information transmission method according to an embodiment of the present application. As shown in FIG5 , the method includes:

[0094] 501: The terminal device sends indication information related to AI / ML features and / or functions to the network device.

[0095] It is worth noting that FIG5 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG5 above.

[0096] In some embodiments, the AI / ML feature includes at least one AI / ML functionality, which may also be referred to as an AI / ML sub-feature.

[0097] In some embodiments, the AI / ML feature includes at least one AI / ML feature group.

[0098] In some embodiments, the AI / ML feature set includes at least one AI / ML function.

[0099] For example, an AI / ML feature or a feature group may include one or more AI / ML functionalities. For example, a CSI prediction feature may further include a low-speed prediction function and a high-speed prediction function.

[0100] In some embodiments, the AI / ML feature has an AI / ML feature ID.

[0101] The AI / ML function has an AI / ML functionality ID, or the AI / ML sub-feature has an AI / ML sub-feature ID.

[0102] In some embodiments, the indication information includes information about AI / ML features and / or functions that the terminal device desires to use.

[0103] In some embodiments, the AI / ML features and / or functions that the terminal device expects to use refer to, for example, the AI / ML features and / or functions that the terminal device determines are suitable for operation based on monitoring and measuring the model that implements the AI / ML features and / or functions.

[0104] In this way, for the terminal-side model, the input and output signal analysis based on the model can make a more accurate assessment of the model performance than the network-side monitoring. In addition, the evaluation of the AI / ML characteristics and / or functional performance corresponding to the model can be further inferred. Through network configuration, the performance evaluation of the AI / ML functional level on the terminal side and the related reporting to the network equipment are allowed or enabled, which realizes accurate AI / ML function monitoring, function activation, deactivation, function selection and function switching, and fallback to non-AI / ML mode. The characteristics of AI / ML functions to improve communication quality are effectively utilized.

[0105] In some embodiments, the terminal device sends the indication information to the network device in one of the following situations:

[0106] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0107] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0108] The network device sends to the terminal device a measurement and judgment indication of the AI / ML feature and / or functional operation related to the indication information; and

[0109] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0110] In some embodiments, the AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0111] Figure 6 is a modular schematic diagram of a network device and a terminal device according to an embodiment of the present application. As shown in Figure 6, the network device has a monitoring unit for feature and / or functional performance, and the terminal device has a unit of a certain AI / ML Feature / Feature Group. Within the unit of the AI / ML Feature / Feature Group, there is at least one functional unit, such as a first functional unit and a second functional unit. For example, as shown in Figure 6, the first function is implemented by at least one model among model a, model b, model c..., and the second function is implemented by at least one model among model x, model y, model z... In addition, within the unit of the AI / ML Feature / Feature Group, there is also a model-level monitoring unit, that is, a model-related monitoring measurement unit.

[0112] Below, the method of the embodiment of the present application and the operation of the terminal device and the network device are described in detail in combination with different application scenarios.

[0113] Application Scenario 1: AI / ML Model Switching

[0114] In some embodiments, the indicative information includes model indicative information.

[0115] For example, the model indication information includes change information of a model or a model group that implements the AI / ML feature and / or function.

[0116] In some embodiments, the change information of the model or model group includes at least one of the following information:

[0117] Information about changes to a model or group of models;

[0118] Identification of the AI / ML feature and / or function and indication of changes to the model or model group; and

[0119] Identification information of a model or model group.

[0120] For example, the information that the model or model group has changed is represented by bit information, for example, by 1 bit indicating that the model or model group has changed.

[0121] For example, the model or models in the model group are bilateral models. The network devices have a consistent understanding of the model identifier sent by the terminal. This identifier indicates that a certain model or model group on the terminal side can work with a certain model or model group on the network side, sometimes referred to as a paired identifier.

[0122] In some embodiments, the model indication information is carried in the relevant reporting information.

[0123] For example, the reporting information is CSI reporting information, and in the AI / ML-related IE of the CSI reporting configuration, reporting of indication information indicating model changes is predefined.

[0124] In some embodiments, the terminal device sends the model indication information to the network device via uplink signaling.

[0125] For example, the model indication information is carried through UCI, MAC CE, or RRC signaling.

[0126] In some embodiments, on the physical channel, the model indication information is sent via PUCCH or PUSCH.

[0127] The application scenario is described below with reference to the accompanying drawings.

[0128] Figure 7 is an interaction diagram of a network device and a terminal device in application scenario 1 of an embodiment of the present application. As shown in Figure 7, the network device activates the UE AI / ML features and / or functions of the terminal device. When the AI / ML features and / or functions are activated or enabled, the terminal device correspondingly turns on model a for related signal processing and simultaneously measures the model usage, such as the signal characteristics of the input and output of model a. If the terminal device decides to switch to model b based on its measurements or other reasons, the terminal device will send relevant indication information to the network device accordingly, i.e., model-related indication information.

[0129] In addition, it should be noted that the above description also applies to switching of model groups, for example, switching from model group a to model group b. Here, each model group may correspond to multiple models of multiple sub-conditions under the conditions to which it is applied.

[0130] In addition, the above description is also applicable to the stopping of a model or a model group, in which case the indication information of the model corresponds to the stopping of the model.

[0131] In addition, when the model indication information corresponds to different model operations, such as model switching, model stopping, model starting, model updating, etc., different model operations can be represented by multiple predefined bit information as model indication information.

[0132] The network monitors the performance of activated AI / M features and / or functions. Upon receiving model-related indications from the terminal device, the network device can adjust its monitoring strategy accordingly, for example, resetting monitoring-related event counters, timers, and model monitoring calculation methods. This means monitoring and determining the corresponding signal transceiver performance after the terminal switches models. This allows for more accurate model monitoring and determination of activation performance.

[0133] Figure 8 is a comparison diagram of the interaction process between the method of the embodiment of the present application and the existing method. As shown in (a) of Figure 8, in the existing method, when the network side monitors that the performance of the AI / ML feature / function is poor, it decides to stop the AI / ML feature / function and sends a deactivation command to the terminal side. The terminal side can only stop using the model that the terminal side just switched to and believes can have good performance. This leads to a decline in subsequent communication performance. In contrast, as shown in (b) of Figure 8, in the embodiment of the present application, the network side readjusts its monitoring and decision strategy after receiving the model-related indication information. Even if the performance of the monitored AI / ML feature / function is poor before receiving the information, the network side can reset the monitoring indicator, continue to maintain the activation state of the AI / ML function, and further monitor the new model used by the terminal side. In this way, due to the continuous use of the AI / ML function, the communication process can benefit from the gains brought by the AI / ML function for a longer period of time.

[0134] Application Scenario 2: Deactivation of AI / ML features and / or functions and / or fallback to non-AI / ML mode

[0135] In some embodiments, the indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0136] In some embodiments, the deactivation request information includes at least one of the following information:

[0137] Identification information of activated AI / ML features and / or functions;

[0138] Deactivation instruction information;

[0139] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0140] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0141] Report configuration identification information; and

[0142] Identification information of a resource or resource set configuration.

[0143] The application scenario is described below with reference to the accompanying drawings.

[0144] Figure 9 is an interaction diagram between a network device and a terminal device in application scenario 2 of an embodiment of the present application. As shown in Figure 9, the network device activates the UE AI / ML features and / or functions on the terminal device. After the AI / ML features and / or functions are activated, the terminal side measures and monitors the models used by it. When it is found that the performance of one or more models it has fails to meet certain conditions, the terminal side sends a deactivation request message for the activated AI / ML function, that is, a UE AI / ML feature and / or function deactivation request message, and / or a request message to fall back to a non-AI / ML mode.

[0145] The deactivation request information may include instruction information for activating the AI / ML feature and / or function, such as function ID information and deactivation instruction information.

[0146] The deactivation request information may also include information on the judgment level of the deactivation information and quality information of the judgment signal, such as SINR and RSRP information.

[0147] The deactivation request information may also include the ID of the CSI reporting configuration or resource (or resource set) configuration, combined with the AI / ML deactivation indication to indicate the desire to deactivate a certain function, or equivalently, the method of deactivating the AI / ML corresponding to the configuration.

[0148] In some embodiments, the request information for falling back to the non-AI / ML mode may include recommended configuration information, parameter information, etc. of the relevant non-AI / ML mode.

[0149] After receiving the deactivation request information, the network device determines that the AI / ML feature and / or function can be deactivated and / or can be rolled back to the non-AI / ML mode, and then sends an indication message of deactivating the UE AI / ML feature and / or function and / or rolling back to the non-AI / ML mode to the terminal device.

[0150] Application Scenario 3: Activation of AI / ML Features and / or Functions

[0151] The indication information includes activation request information for inactive AI / ML features and / or functions.

[0152] In some embodiments, the activation request information includes at least one of the following information:

[0153] Identification information for inactive AI / ML features and / or functions;

[0154] Request information or instructions for activating AI / ML;

[0155] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0156] Quality information about the signals used to activate AI / ML features and / or functions;

[0157] Report configuration identification information; and

[0158] Identification information of a resource or resource set configuration.

[0159] The application scenario is described below with reference to the accompanying drawings.

[0160] Figure 10 is an interaction diagram of a network device and a terminal device in application scenario 3 of an embodiment of the present application. As shown in Figure 10, for a certain AI / ML feature and / or function in an inactive state, after the terminal device measures or monitors the input or output related to the model, if it is determined that it is suitable to run the AI / ML feature and / or function, then the terminal device sends an activation request to start AI / ML to the network device. The request information may include function indication information, or configuration indication information, etc. That is, UE AI / ML feature and / or function activation request information is sent. After receiving the deactivation request information, the network device determines that the AI / ML feature and / or function can be activated, and then sends UE AI / ML feature and / or function activation indication information to the terminal device.

[0161] The activation request information may include information indicating the desired activation of the AI / ML feature and / or function, such as a feature ID and / or function ID, and / or information requesting or indicating activation of AI / ML. This may indicate, for example, a request to activate the AI / ML method for CSI prediction, a request to activate the AI / ML method for beamformation temporal or spatial prediction, and so on.

[0162] In addition, the activation request information may include information on the degree of judgment of the activation information and quality information of the signal used to judge the activation, such as SINR and RSRP information.

[0163] In addition, the activation request information may also include the ID of the CSI reporting configuration or resource (or resource set) configuration, combined with the AI / ML activation indication to indicate the desired activation of a certain function, or equivalently, the method of deactivating the AI / ML corresponding to the configuration.

[0164] Application Scenario 4: Selection of AI / ML Features and / or Functions

[0165] In some embodiments, the indication information includes indication information of AI / ML features and / or functions selected by the terminal device.

[0166] In some embodiments, the indication information includes at least one of the following information:

[0167] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0168] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0169] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0170] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0171] In some embodiments, the method further includes: the terminal device receiving measurement configuration information and / or reporting configuration information sent by the network device.

[0172] In some embodiments, the measurement configuration information includes at least one of a configuration of a CSI resource or resource set, and a configuration of an RS.

[0173] In some embodiments, one configuration in the measurement configuration information and / or the reporting configuration information corresponds to the implementation of at least one AI / ML feature and / or function.

[0174] In some embodiments, the measurement configuration information and / or reporting configuration information includes evaluation indication information of AI / ML features and / or functions, or selection indication information of AI / ML features and / or functions.

[0175] In some embodiments, the measurement configuration information and / or reporting configuration information is configured by identifying an AI / ML feature and / or function.

[0176] For example, the AI / ML features and / or functions corresponding to the measurement configuration information and / or reporting configuration information are determined by the network device based on the AI / ML features and / or functions supported by the terminal device reported by the terminal device.

[0177] In some embodiments, the measurement configuration information and / or reporting configuration information is configured via RRC, or,

[0178] After the measurement configuration information and / or reporting configuration information is configured through RRC, at least one measurement configuration and / or reporting configuration is activated by MAC CE.

[0179] In some embodiments, the measurement configuration information and / or reporting configuration information includes AI / ML evaluation instruction information.

[0180] The application scenario is described below with reference to the accompanying drawings.

[0181] Figure 11 is an interaction diagram between a network device and a terminal device in application scenario 4 of an embodiment of the present application. As shown in Figure 11, the network device sends measurement configuration information related to UE AI / ML features and / or functions to the terminal device. That is, the network side configures and sends measurement configuration information suitable for the terminal side to perform one or more AI / ML features and / or functions, such as the configuration of CSI resources or resource sets, RS configuration, etc.

[0182] For example, a configuration may correspond to the implementation of one or more functions, including AI / ML features and / or function evaluation instructions or function selection instructions. Alternatively, configuration can be performed using function IDs.

[0183] In some embodiments, the measurement configuration information may be configured via RRC, or after RRC configuration is completed, one or more measurement configurations may be activated by MAC CE, such as configurations of resources or resource sets corresponding to CSI or RS.

[0184] In some embodiments, the measurement configuration information may be configured via RRC, or after the RRC configuration is completed, one or more reporting configurations may be activated by MAC CE, such as reporting configuration corresponding to CSI.

[0185] In some embodiments, the measurement configuration information includes an instruction to perform AI / ML evaluation.

[0186] In some embodiments, the network side selects one or more AI / ML features and / or functions that the network device also supports or expects the terminal to use based on the AI / ML features and / or functions reported by the terminal device, and configures them to the terminal device. As described above, the configuration method may include a function ID, and may also include measurement configuration information corresponding to the AI / ML feature and / or function, such as the configuration of CSI resources or resource sets, or include reporting configuration corresponding to the feature and / or function.

[0187] The network side sends configuration-related signals, such as CSI-RS, SSB, etc. The transmitted signals may also be beamformed or precoded according to the functional configuration before being sent.

[0188] The terminal side performs relevant measurements, monitoring, and judgments on different models belonging to the one or more AI / ML features and / or functions based on the received signal configuration and the AI / ML feature and / or function evaluation indication information or function selection indication information, selects a suitable model or a suitable AI / ML feature and / or function, or makes a judgment on whether it is appropriate to activate the AI / ML feature and / or function.

[0189] The terminal device then sends AI / ML feature and / or function indication information.

[0190] In some embodiments, the indication information may include information of the selected feature ID or function ID.

[0191] Or configuration indication information for resource configuration or reported configuration of the network, for example, an ID is selected from configuration IDs configured on the network side.

[0192] In some embodiments, the indication information may include information on the degree of determination of the activation information, and quality information of the signal used to determine the activation, such as SINR and RSRP information.

[0193] After receiving the AI / ML feature and / or function indication information, the network device determines that the AI / ML feature and / or function can be activated, and then sends the UE AI / ML feature and / or function activation indication information to the terminal device.

[0194] Application Scenario 5: Switching AI / ML Features and / or Functions

[0195] In some embodiments, the indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0196] In some embodiments, the indication information includes at least one of the following information:

[0197] identification information of the characteristics and / or functions to be switched;

[0198] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0199] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0200] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0201] In some embodiments, the method further includes: the terminal device determines whether to switch the AI / ML features and / or functions based on the evaluation index information of switching AI / ML features and / or functions configured on the network side.

[0202] In some embodiments, the evaluation index information includes at least one of a measurement index, a measurement threshold, a comparison index, and a comparison threshold.

[0203] In some embodiments, the indication information includes at least one of input configuration related information and output configuration related information corresponding to the AI / ML feature and / or function.

[0204] The application scenario is described below with reference to the accompanying drawings.

[0205] Figure 12 is an interaction diagram of a network device and a terminal device in application scenario 5 of an embodiment of the present application. As shown in Figure 12, the network device activates a certain AI / ML feature and / or function of the terminal, such as the first AI / ML feature and / or function. In addition, the network device is also configured with permission for the terminal side to request switching of AI / ML features and / or functions, or the network side is configured with resource configurations of other AI / ML features and / or functions to the terminal device, but in the reporting configuration, the terminal device is not required to report. For example, through RRC, and or MAC CE configuration methods, etc. In addition, the network device configures the judgment basis for switching AI / ML features and / or functions to the terminal, such as measurement thresholds, comparison indicators, comparison thresholds, etc. The network side sends configuration-related signals, such as CSI-RS, SSB, etc.

[0206] The terminal device can then assess the potential performance of inactive AI / ML features and / or functions. The terminal device can analyze the input and output data of the inactive model that implements the AI / ML feature and / or function and compare it with the performance of the active model.

[0207] The terminal device determines whether the inactive model is better than the active model based on measurement indicators, comparison indicators, comparison thresholds, etc. configured on the network side, and thereby selects an inactive AI / ML feature and / or function that is better than the existing AI / ML feature and / or function, such as a second AI / ML feature and / or function.

[0208] The terminal device sends AI / ML feature and / or function switching request information, or second AI / ML feature and / or function activation request information.

[0209] In some embodiments, the request information may include information of the selected feature ID and / or function ID.

[0210] Or configuration indication information corresponding to the second function, such as a configuration ID.

[0211] In some embodiments, the indication information may include information on the degree of determination of the activation information, and quality information of the signal used to determine the activation, such as SINR and RSRP information.

[0212] For example, for the above application scenarios 4 and 5, taking the beam management feature as an example, it has the following two functions:

[0213] First function: The model to which this function belongs corresponds to CSI resource configuration or resource configuration set 1 (including CSI-RS and / or SSB configuration information), and the situation where the number of spatial beams sent by the corresponding network device is large;

[0214] Second function: The model to which this function belongs corresponds to CSI resource configuration or resource configuration set 2 (including CSI-RS and / or SSB configuration information), and the number of spatial beams sent by the corresponding network device is small.

[0215] The terminal device determines which function is more suitable based on the measured SSB or CSI-RS and the model output, and sends the function indication information accordingly.

[0216] For example, the function indication information may be CSI configuration-related indication information, such as a configuration ID, or SSB, CSI-RS-related parameter information.

[0217] For example, the function indication information may be function ID information, or other predefined function parameter information.

[0218] For example, the function indication information may be other information that enables the base station side to understand the function corresponding to the information.

[0219] In some embodiments, the above-mentioned function indication information and its corresponding function are predefined.

[0220] The other is related to CSI prediction, where the configuration corresponds to the time period or frequency domain density of the CSI-RS, or the configuration of its measurement window parameters.

[0221] That is to say, the terminal side sends information related to the input configuration corresponding to the function.

[0222] In addition, the terminal device can also send information related to the function output configuration, such as the time domain prediction of the corresponding beam, the number of predictions, the configuration information of the prediction window, etc.

[0223] That is, the terminal side sends information related to the output configuration corresponding to the function, that is, function-related reporting configuration information, such as a reporting configuration ID.

[0224] The embodiments of the above application scenarios are merely exemplary of the embodiments of the present application, but the present application is not limited thereto. Appropriate modifications may be made based on the embodiments of the above application scenarios. For example, the embodiments of the above application scenarios may be used individually, or one or more of the embodiments of the above application scenarios may be combined.

[0225] It can be seen from the above embodiments that the terminal device sends indication information related to AI / ML features and / or functions to the network device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of AI / ML features and / or functions, thereby making full use of AI / ML functions to improve communication quality and improve communication system performance.

[0226] Embodiments of the second aspect

[0227] An embodiment of the present application provides an information transmission method, which is described from the perspective of a network device. It corresponds to the embodiment of the first aspect, and the contents that are the same as those in the embodiment of the first aspect will not be repeated.

[0228] FIG13 is another schematic diagram of the information transmission method according to an embodiment of the present application. As shown in FIG13 , the method includes:

[0229] 1301: The network device receives indication information related to AI / ML features and / or functions sent by the terminal device.

[0230] It is worth noting that FIG13 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG13 above.

[0231] In some embodiments, the indication information includes information about AI / ML features and / or functions that the terminal device desires to use.

[0232] In some embodiments, the method further comprises one of the following steps:

[0233] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0234] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0235] The network device sends to the terminal device a measurement and judgment indication of the AI / ML feature and / or functional operation related to the indication information; and

[0236] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0237] In some embodiments, the AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0238] In some embodiments, the indication information includes indication information of AI / ML features and / or functions selected by the terminal device.

[0239] In some embodiments, the indication information includes at least one of the following information:

[0240] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0241] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0242] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0243] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0244] In some embodiments, the method further comprises:

[0245] The network device sends measurement configuration information and / or reporting configuration information to the terminal device.

[0246] In some embodiments, the measurement configuration information includes at least one of a configuration of a CSI resource or resource set, and a configuration of an RS.

[0247] In some embodiments, one configuration in the measurement configuration information and / or the reporting configuration information corresponds to the implementation of at least one AI / ML feature and / or function.

[0248] In some embodiments, the measurement configuration information and / or reporting configuration information includes evaluation indication information of AI / ML features and / or functions, or selection indication information of AI / ML features and / or functions.

[0249] In some embodiments, the measurement configuration information and / or reporting configuration information is configured by identifying an AI / ML feature and / or function.

[0250] In some embodiments, the AI / ML features and / or functions corresponding to the measurement configuration information and / or reporting configuration information are determined by the network device based on the AI / ML features and / or functions supported by the terminal device reported by the terminal device.

[0251] In some embodiments, the measurement configuration information and / or reporting configuration information is configured via RRC, or,

[0252] After the measurement configuration information and / or reporting configuration information is configured through RRC, at least one measurement configuration and / or reporting configuration is activated by MAC CE.

[0253] In some embodiments, the measurement configuration information and / or reporting configuration information includes AI / ML evaluation instruction information.

[0254] In some embodiments, the indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0255] In some embodiments, the indication information includes at least one of the following information:

[0256] identification information of the characteristics and / or functions to be switched;

[0257] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0258] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0259] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0260] In some embodiments, the method further comprises:

[0261] The network device configures evaluation indicator information for switching AI / ML features and / or functions to the terminal device.

[0262] In some embodiments, the evaluation index information includes at least one of a measurement index, a measurement threshold, a comparison index, and a comparison threshold.

[0263] In some embodiments, the indication information includes at least one of input configuration related information and output configuration related information corresponding to the AI / ML feature and / or function.

[0264] In some embodiments, the indicative information includes model indicative information.

[0265] In some embodiments, the model indication information includes change information of a model or a group of models that implement the AI / ML feature and / or function.

[0266] In some embodiments, the change information of the model or model group includes at least one of the following information:

[0267] Information about changes to a model or group of models;

[0268] Identification of the AI / ML feature and / or function and indication of changes to the model or model group; and

[0269] Identification information of a model or model group.

[0270] In some embodiments, the information that the model or model group has changed is represented by bit information.

[0271] In some embodiments, a change in a model or model group is indicated by 1 bit.

[0272] In some embodiments, the model or models in the model group are bilateral model structures.

[0273] In some embodiments, the model indication information is carried in the relevant reporting information.

[0274] In some embodiments, the reported information is CSI reporting information, and in the AI / ML-related IE of the CSI reporting configuration, reporting of indication information indicating model changes is predefined.

[0275] In some embodiments, the model indication information is sent via uplink signaling.

[0276] In some embodiments, the model indication information is carried via UCI, MAC CE, or RRC signaling.

[0277] In some embodiments, on the physical channel, the model indication information is sent via PUCCH or PUSCH.

[0278] In some embodiments, the indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0279] In some embodiments, the deactivation request information includes at least one of the following information:

[0280] Identification information of activated AI / ML features and / or functions;

[0281] Deactivation instruction information;

[0282] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0283] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0284] Report configuration identification information; and

[0285] Identification information of a resource or resource set configuration.

[0286] In some embodiments, the indication information includes activation request information for inactive AI / ML features and / or functions.

[0287] In some embodiments, the activation request information includes at least one of the following information:

[0288] Identification information for inactive AI / ML features and / or functions;

[0289] Request information or instructions for activating AI / ML;

[0290] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0291] Quality information about the signals used to activate AI / ML features and / or functions;

[0292] Report configuration identification information; and

[0293] Identification information of a resource or resource set configuration.

[0294] In some embodiments, the AI / ML feature includes at least one AI / ML function.

[0295] In some embodiments, the AI / ML features include at least one AI / ML feature group.

[0296] In some embodiments, the AI / ML feature set includes at least one AI / ML function.

[0297] In some embodiments, the AI / ML functionality is an AI / ML sub-feature.

[0298] In some embodiments, the AI / ML feature has an AI / ML feature identifier.

[0299] In some embodiments, the AI / ML function has an AI / ML function identifier, or the AI / ML sub-feature has an AI / ML sub-feature identifier.

[0300] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0301] It can be seen from the above embodiments that the network device receives indication information related to AI / ML features and / or functions from the terminal device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of AI / ML features and / or functions, thereby making full use of AI / ML functions to improve communication quality and improve communication system performance.

[0302] Embodiments of the third aspect

[0303] An embodiment of the present application provides a method for verifying model performance.

[0304] FIG14 is a schematic diagram of a method for verifying model performance according to an embodiment of the present application, which corresponds to the terminal side. As shown in FIG14 , the method includes:

[0305] 1401: The terminal device receives the test vector set and reporting configuration from the network device;

[0306] 1402: The terminal device inputs the test vectors in the test vector set into the model, and sends the output of the model to the network device according to the reporting configuration.

[0307] In some embodiments, the test vector set is carried via RRC.

[0308] In some embodiments, the correspondence between the test vectors in the test vector set and the model is indicated by indication information sent by the network device, or the test vector set has an identifier corresponding to the model.

[0309] FIG15 is another schematic diagram of a method for verifying model performance according to an embodiment of the present application, which corresponds to the network side. As shown in FIG15 , the method includes:

[0310] 1501: The network device sends the test vector set to the terminal device; and

[0311] 1502: The network device receives the output of the model sent by the terminal device.

[0312] In some embodiments, the network device sends the test vector set to the terminal device via an RRC message.

[0313] In some embodiments, the correspondence between the test vectors in the test vector set and the model is indicated by indication information sent by the network device, or the test vector set has an identifier corresponding to the model.

[0314] In some embodiments, the method further includes: the network device notifies the terminal device of the purpose of the test vector set through a specific RRC message or other binding configuration, and sends relevant reporting configuration.

[0315] FIG16 is another schematic diagram of a method for verifying model performance according to an embodiment of the present application, which corresponds to the network side and the terminal side. As shown in FIG16 , the method includes:

[0316] 1601: The terminal device registers, identifies, or notifies the network device of the model. This may include an indication that the model has not passed the test. In addition, the terminal device reports the model input and output information parameters to the network device.

[0317] 1602: The network device further obtains test vector information of the corresponding model (eg, based on the model identification) based on the obtained model input and output information parameters, model identification and other information.

[0318] For this test vector, one part corresponds to the input of the model, and the other part corresponds to the output that a qualified model should have.

[0319] 1603: The network device sends the test vector set to the terminal side and configures related reporting. The test vector corresponds to the model input.

[0320] 1604: The terminal device receives the test vector set, provides it to the model as input, and sends the output of the model to the network device according to the reporting configuration.

[0321] 1605: The network device compares the output of the model with the model expectations (e.g., based on the model identifier) ​​of the network device, thereby determining whether the untested model can meet the performance requirements.

[0322] 1606: When the performance requirement is met, the network device sends an indication to the terminal device that the model has passed the test or has been tested.

[0323] In some embodiments, if it is assumed that the untested model does not have a global identifier, the network side will issue a global identifier to the model after the above test is passed.

[0324] In some embodiments, the test vector set may be carried by RRC and inform the terminal device of its corresponding model through indication information, such as through the model identifier, or the test vector data set has an identifier corresponding to the model.

[0325] In addition, the network device may inform the terminal device of the purpose of the test vector set through a specific RRC message or other binding configuration, and send relevant reporting configuration.

[0326] In some embodiments, the test includes at least one of an interoperability test, a RAN4 test, and a network entry test.

[0327] As can be seen from the above embodiments, the network device sends a data set to the terminal device, and the data set can be used as a test vector to verify the performance of the terminal model.

[0328] Embodiments of the fourth aspect

[0329] An embodiment of the present application provides a method for identifying an AI / ML model.

[0330] To adapt to different application scenarios, configurations, and conditions, terminal devices or network equipment may utilize multiple AI / ML models for a single function or module. These models may not have undergone, or may have undergone insufficient, RAN4 testing, network access testing, or other types of interoperability testing before being used in the network. Further performance verification of the models is required before they are deployed in a live network or activated. Furthermore, performance evaluation of an inactive model is also required before activation.

[0331] FIG17 is a schematic diagram of an AI / ML model identification method according to an embodiment of the present application. As shown in FIG17 , the method includes:

[0332] 1701: Indicate whether the AI / ML model has been tested by using partial bit information or digital information in the model identifier of the AI / ML model.

[0333] In some embodiments, the model identification is a global identification.

[0334] In some embodiments, whether the test is passed is indicated by a bit or a number in the bit string or number string identified by the model.

[0335] In other words, a display indicator is introduced to mark the tested and untested models. The indicator can be a partial bit information or digital information in the model identifier, for example, a 1 represents a tested model and a 0 represents an untested model in the bit string or digital string of the model identifier.

[0336] In some embodiments, the indication may be that an untested model has no model identifier, and only a tested model has an identifier.

[0337] The above model identifier is a global identifier, which can be given offline or when registering the model network.

[0338] In some embodiments, the test includes at least one of an interoperability test, a RAN4 test, and a network entry test.

[0339] On the other hand, models that have not undergone interoperability testing, RAN4 testing, or network access testing do not have an initial global model identifier and must undergo a network-wide test verification process. After testing and verification that they meet performance requirements, the network side will grant the global model through model registration or model identification.

[0340] Specifically, after the model test verifies that the conditions are met, a network device may send relevant model information to a network element responsible for managing the global model identifier to obtain a global identifier.

[0341] Alternatively, after the model test verifies that the conditions are met, the terminal device initiates a model registration or model identification request for the model to the network side, sends relevant model information to the relevant network element responsible for managing the global model identification, and obtains the global identification.

[0342] FIG18 is another schematic diagram of the AI / ML model identification method according to an embodiment of the present application. As shown in FIG18 , the method includes:

[0343] 1801: For AI / ML models that do not have a global identifier, the network side provides the global identifier of the AI / ML model after testing and verification that the preset conditions are met.

[0344] In some embodiments, after the test verifies that the preset conditions are met, the network side provides a global identifier of the AI / ML model through a model registration or model identification process.

[0345] In some embodiments, after testing and verifying that preset conditions are met, the network device or terminal device sends relevant model information to the network element responsible for managing the global model identification, and obtains the global identification of the AI / ML model from the network element.

[0346] In some embodiments, the test includes at least one of an interoperability test, a RAN4 test, and a network entry test.

[0347] Alternatively, for all models that are not registered on the network, their initial model identification includes an identification of whether they have passed the interoperability test, RAN4 test, network access test, and other tests. For models that have not undergone the above tests, they need to be registered on the network and undergo the required test verification process on the network before being used. After meeting the conditions, they are granted relevant permission or initial identification information. Based on this permission information or initial identification information, the terminal side, or the base station, initiates a model registration or model identification request for the model to the network side, sends the relevant model information to the relevant network element responsible for managing the global model identification, and obtains the global identification.

[0348] FIG19 is another schematic diagram of an AI / ML model identification method according to an embodiment of the present application. As shown in FIG19 , the method includes:

[0349] 1901: For an AI / ML model that is not registered on the network, the initial model identifier of the AI / ML model is used to indicate whether the AI / ML model has been tested.

[0350] In some embodiments, the test includes at least one of an interoperability test, a RAN4 test, and a network entry test.

[0351] Embodiments of the fifth aspect

[0352] The embodiment of the present application provides an information transmission device. The device may be, for example, a terminal device, or one or more components or assemblies configured in the terminal device, which corresponds to the embodiment of the first aspect, and the same contents as the embodiment of the first aspect are not repeated here.

[0353] FIG20 is a schematic diagram of an information transmission device according to an embodiment of the present application. As shown in FIG20 , the information transmission device 2000 includes:

[0354] The first sending unit 2001 sends indication information related to AI / ML features and / or functions to the network device.

[0355] In some embodiments, the indication information includes information about AI / ML features and / or functions that the terminal device desires to use.

[0356] In some embodiments, the first sending unit 2001 sends the indication information to the network device in one of the following situations:

[0357] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0358] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0359] The network device sends to the terminal device a measurement and judgment indication of the AI / ML feature and / or functional operation related to the indication information; and

[0360] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0361] In some embodiments, the AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0362] In some embodiments, the indication information includes indication information of AI / ML features and / or functions selected by the terminal device.

[0363] In some embodiments, the indication information includes at least one of the following information:

[0364] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0365] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0366] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0367] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0368] In some embodiments, as shown in FIG20 , the apparatus further comprises:

[0369] The first receiving unit 2002 receives measurement configuration information and / or reporting configuration information sent by the network device.

[0370] In some embodiments, the measurement configuration information includes at least one of a configuration of a CSI resource or resource set, and a configuration of an RS.

[0371] In some embodiments, the indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0372] In some embodiments, the indication information includes at least one of the following information:

[0373] identification information of the characteristics and / or functions to be switched;

[0374] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0375] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0376] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0377] In some embodiments, the indicative information includes model indicative information.

[0378] In some embodiments, the model indication information includes change information of a model or a group of models that implement the AI / ML feature and / or function.

[0379] In some embodiments, the change information of the model or model group includes at least one of the following information:

[0380] Information about changes to a model or group of models;

[0381] Identification of the AI / ML feature and / or function and indication of changes to the model or model group; and

[0382] Identification information of a model or model group.

[0383] In some embodiments, the model indication information is carried in the relevant reporting information.

[0384] In some embodiments, the first sending unit 2001 sends the model indication information to the network device via uplink signaling.

[0385] In some embodiments, the indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0386] In some embodiments, the deactivation request information includes at least one of the following information:

[0387] Identification information of activated AI / ML features and / or functions;

[0388] Deactivation instruction information;

[0389] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0390] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0391] Report configuration identification information; and

[0392] Identification information of a resource or resource set configuration.

[0393] In some embodiments, the indication information includes activation request information for inactive AI / ML features and / or functions.

[0394] In some embodiments, the activation request information includes at least one of the following information:

[0395] Identification information for inactive AI / ML features and / or functions;

[0396] Request information or instructions for activating AI / ML;

[0397] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0398] Quality information about the signals used to activate AI / ML features and / or functions;

[0399] Report configuration identification information; and

[0400] Identification information of a resource or resource set configuration.

[0401] In addition, for the sake of simplicity, FIG20 only illustrates the connection relationship or signal direction between various components or modules. However, it should be clear to those skilled in the art that various related technologies such as bus connection can be used. The above-mentioned components or modules can be implemented by hardware facilities such as processors, memories, transmitters, and receivers; the implementation of this application is not limited to this.

[0402] It can be seen from the above embodiments that the terminal device sends indication information related to AI / ML features and / or functions to the network device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of AI / ML features and / or functions, thereby making full use of AI / ML functions to improve communication quality and improve communication system performance.

[0403] Embodiments of the seventh aspect

[0404] The present application provides an information transmission device. The device may be, for example, a network device, or one or more components or assemblies configured on the network device, which corresponds to the embodiment of the second aspect, and the contents identical to those of the embodiments of the first and second aspects are not repeated here.

[0405] FIG21 is another schematic diagram of an information transmission device according to an embodiment of the present application. As shown in FIG21 , the information transmission device 2100 includes:

[0406] The second receiving unit 2101 receives indication information related to AI / ML features and / or functions sent by the terminal device.

[0407] It is worth noting that the above description only describes the components or modules related to the present application, but the present application is not limited thereto. The information transmission device 2100 may also include other components or modules. For the specific contents of these components or modules, reference may be made to the relevant art.

[0408] In addition, for the sake of simplicity, FIG21 only illustrates the connection relationship or signal direction between various components or modules. However, it should be clear to those skilled in the art that various related technologies such as bus connection can be used. The above-mentioned components or modules can be implemented by hardware facilities such as processors, memories, transmitters, and receivers; the implementation of this application is not limited to this.

[0409] It can be seen from the above embodiments that the network device receives indication information related to AI / ML features and / or functions from the terminal device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of AI / ML features and / or functions, thereby making full use of AI / ML functions to improve communication quality and improve communication system performance.

[0410] Embodiments of the eighth aspect

[0411] An embodiment of the present application provides a terminal device, which includes the information transmission device as described in the embodiment of the sixth aspect.

[0412] Figure 22 is a schematic block diagram of the system architecture of a terminal device according to an embodiment of the present application. As shown in Figure 22, terminal device 2200 may include a processor 2210 and a memory 2220; memory 2220 is coupled to processor 2210. It should be noted that this diagram is exemplary; other types of structures may be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0413] In one embodiment, the functionality of the information transmission device may be integrated into the processor 2210 .

[0414] The processor 2210 is configured to: the terminal device sends indication information related to AI / ML features and / or functions to the network device.

[0415] In some embodiments, the indication information includes information about AI / ML features and / or functions that the terminal device desires to use.

[0416] In some embodiments, the terminal device sends the indication information to the network device in one of the following situations:

[0417] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0418] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0419] The network device sends to the terminal device a measurement and judgment indication of the AI / ML feature and / or functional operation related to the indication information; and

[0420] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0421] In some embodiments, the AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0422] In some embodiments, the indication information includes indication information of AI / ML features and / or functions selected by the terminal device.

[0423] In some embodiments, the indication information includes at least one of the following information:

[0424] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0425] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0426] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0427] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0428] In some embodiments, the processor 2210 is further configured to:

[0429] The terminal device receives the measurement configuration information and / or reporting configuration information sent by the network device.

[0430] In some embodiments, the measurement configuration information includes at least one of a configuration of a CSI resource or resource set, and a configuration of an RS.

[0431] In some embodiments, one configuration in the measurement configuration information and / or the reporting configuration information corresponds to the implementation of at least one AI / ML feature and / or function.

[0432] In some embodiments, the measurement configuration information and / or reporting configuration information includes evaluation indication information of AI / ML features and / or functions, or selection indication information of AI / ML features and / or functions.

[0433] In some embodiments, the measurement configuration information and / or reporting configuration information is configured by identifying an AI / ML feature and / or function.

[0434] In some embodiments, the AI / ML features and / or functions corresponding to the measurement configuration information and / or reporting configuration information are determined by the network device based on the AI / ML features and / or functions supported by the terminal device reported by the terminal device.

[0435] In some embodiments, the measurement configuration information and / or reporting configuration information is configured via RRC, or,

[0436] After the measurement configuration information and / or reporting configuration information is configured through RRC, at least one measurement configuration and / or reporting configuration is activated by MAC CE.

[0437] In some embodiments, the measurement configuration information and / or reporting configuration information includes AI / ML evaluation instruction information.

[0438] In some embodiments, the indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0439] In some embodiments, the indication information includes at least one of the following information:

[0440] identification information of the characteristics and / or functions to be switched;

[0441] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0442] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0443] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0444] In some embodiments, the processor 2210 is further configured to:

[0445] The terminal device determines whether to switch the AI / ML features and / or functions based on the evaluation indicator information of switching AI / ML features and / or functions configured on the network side.

[0446] In some embodiments, the evaluation index information includes at least one of a measurement index, a measurement threshold, a comparison index, and a comparison threshold.

[0447] In some embodiments, the indication information includes at least one of input configuration related information and output configuration related information corresponding to the AI / ML feature and / or function.

[0448] In some embodiments, the indicative information includes model indicative information.

[0449] In some embodiments, the model indication information includes change information of a model or a group of models that implement the AI / ML feature and / or function.

[0450] In some embodiments, the change information of the model or model group includes at least one of the following information:

[0451] Information about changes to a model or group of models;

[0452] Identification of the AI / ML feature and / or function and indication of changes to the model or model group; and

[0453] Identification information of a model or model group.

[0454] In some embodiments, the information that the model or model group has changed is represented by bit information.

[0455] In some embodiments, a change in a model or model group is indicated by 1 bit.

[0456] In some embodiments, the model or models in the model group are bilateral model structures.

[0457] In some embodiments, the model indication information is carried in the relevant reporting information.

[0458] In some embodiments, the reporting information is CSI reporting information,

[0459] In the AI / ML related IE of the CSI reporting configuration, the reporting of indication information indicating model changes is predefined.

[0460] In some embodiments, the terminal device sends the model indication information to the network device via uplink signaling.

[0461] In some embodiments, the model indication information is carried via UCI, MAC CE, or RRC signaling.

[0462] In some embodiments, on the physical channel, the model indication information is sent via PUCCH or PUSCH.

[0463] In some embodiments, the indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0464] In some embodiments, the deactivation request information includes at least one of the following information:

[0465] Identification information of activated AI / ML features and / or functions;

[0466] Deactivation instruction information;

[0467] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0468] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0469] Report configuration identification information; and

[0470] Identification information of a resource or resource set configuration.

[0471] In some embodiments, the indication information includes activation request information for inactive AI / ML features and / or functions.

[0472] In some embodiments, the activation request information includes at least one of the following information:

[0473] Identification information for inactive AI / ML features and / or functions;

[0474] Request information or instructions for activating AI / ML;

[0475] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0476] Quality information about the signals used to activate AI / ML features and / or functions;

[0477] Report configuration identification information; and

[0478] Identification information of a resource or resource set configuration.

[0479] In some embodiments, the AI / ML feature includes at least one AI / ML function.

[0480] In some embodiments, the AI / ML features include at least one AI / ML feature group.

[0481] In some embodiments, the AI / ML feature set includes at least one AI / ML function.

[0482] In some embodiments, the AI / ML functionality is an AI / ML sub-feature.

[0483] In some embodiments, the AI / ML feature has an AI / ML feature identifier.

[0484] In some embodiments, the AI / ML function has an AI / ML function identifier, or the AI / ML sub-feature has an AI / ML sub-feature identifier.

[0485] In another embodiment, the information transmission device can be configured separately from the processor 2210. For example, the transmission device of the physical random access channel can be configured as a chip connected to the processor 2210, and the function of the transmission device of the physical random access channel is realized through the control of the processor 2210.

[0486] As shown in FIG22 , the terminal device 2200 may further include: a communication module 2230, an input unit 2240, a display 2250, and a power supply 2260. It is worth noting that the terminal device 2200 does not necessarily include all the components shown in FIG22 ; in addition, the terminal device 2200 may also include components not shown in FIG22 , and reference may be made to related art for details.

[0487] As shown in FIG. 22 , the processor 2210 , sometimes also referred to as a controller or an operation control, may include a microprocessor or other processor device and / or logic device. The processor 2210 receives input and controls the operation of various components of the terminal device 2200 .

[0488] Memory 2220 can be, for example, one or more of a cache, flash memory, a hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It can store various data and programs for executing related information. Processor 2210 can execute the programs stored in memory 2220 to implement information storage or processing. The functions of other components are similar to those of existing devices and are not further described here. Each component of terminal device 2200 can be implemented using dedicated hardware, firmware, software, or a combination thereof without departing from the scope of the present invention.

[0489] It can be seen from the above embodiments that the terminal device sends indication information related to AI / ML features and / or functions to the network device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of AI / ML features and / or functions, thereby making full use of AI / ML functions to improve communication quality and improve communication system performance.

[0490] Embodiments of the ninth aspect

[0491] An embodiment of the present application provides a network device, which includes the information transmission device as described in the embodiment of the seventh aspect.

[0492] Figure 23 is a schematic block diagram of the system configuration of a network device according to an embodiment of the present application. As shown in Figure 23 , network device 2300 may include a processor 2310 and a memory 2320 ; the memory 2320 is coupled to the processor 2310 . The memory 2320 may store various data and may also store an information processing program 2330 , which is executed under the control of the processor 2310 .

[0493] In one embodiment, the functionality of the information transmission device may be integrated into the processor 2310 .

[0494] The processor 2310 may be configured to: receive, by the network device, indication information related to AI / ML features and / or functions sent by the terminal device.

[0495] In some embodiments, the indication information includes information about AI / ML features and / or functions that the terminal device desires to use.

[0496] In some embodiments, the processor 2310 may be further configured to perform one of the following steps:

[0497] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0498] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0499] The network device sends to the terminal device a measurement and judgment indication of the AI / ML feature and / or functional operation related to the indication information; and

[0500] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0501] In some embodiments, the AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0502] In some embodiments, the indication information includes indication information of AI / ML features and / or functions selected by the terminal device.

[0503] In some embodiments, the indication information includes at least one of the following information:

[0504] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0505] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0506] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0507] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0508] In some embodiments, the processor 2310 may also be configured to:

[0509] The network device sends measurement configuration information and / or reporting configuration information to the terminal device.

[0510] In some embodiments, the measurement configuration information includes at least one of a configuration of a CSI resource or resource set, and a configuration of an RS.

[0511] In some embodiments, one configuration in the measurement configuration information and / or the reporting configuration information corresponds to the implementation of at least one AI / ML feature and / or function.

[0512] In some embodiments, the measurement configuration information and / or reporting configuration information includes evaluation indication information of AI / ML features and / or functions, or selection indication information of AI / ML features and / or functions.

[0513] In some embodiments, the measurement configuration information and / or reporting configuration information is configured by identifying an AI / ML feature and / or function.

[0514] In some embodiments, the AI / ML features and / or functions corresponding to the measurement configuration information and / or reporting configuration information are determined by the network device based on the AI / ML features and / or functions supported by the terminal device reported by the terminal device.

[0515] In some embodiments, the measurement configuration information and / or reporting configuration information is configured via RRC, or,

[0516] After the measurement configuration information and / or reporting configuration information is configured through RRC, at least one measurement configuration and / or reporting configuration is activated by MAC CE.

[0517] In some embodiments, the measurement configuration information and / or reporting configuration information includes AI / ML evaluation instruction information.

[0518] In some embodiments, the indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0519] In some embodiments, the indication information includes at least one of the following information:

[0520] identification information of the characteristics and / or functions to be switched;

[0521] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0522] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0523] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0524] In some embodiments, the processor 2310 may also be configured to:

[0525] The network device configures evaluation indicator information for switching AI / ML features and / or functions to the terminal device.

[0526] In some embodiments, the evaluation index information includes at least one of a measurement index, a measurement threshold, a comparison index, and a comparison threshold.

[0527] In some embodiments, the indication information includes at least one of input configuration related information and output configuration related information corresponding to the AI / ML feature and / or function.

[0528] In some embodiments, the indicative information includes model indicative information.

[0529] In some embodiments, the model indication information includes change information of a model or a group of models that implement the AI / ML feature and / or function.

[0530] In some embodiments, the change information of the model or model group includes at least one of the following information:

[0531] Information about changes to a model or group of models;

[0532] Identification of the AI / ML feature and / or function and indication of changes to the model or model group; and

[0533] Identification information of a model or model group.

[0534] In some embodiments, the information that the model or model group has changed is represented by bit information.

[0535] In some embodiments, a change in a model or model group is indicated by 1 bit.

[0536] In some embodiments, the model or models in the model group are bilateral model structures.

[0537] In some embodiments, the model indication information is carried in the relevant reporting information.

[0538] In some embodiments, the reporting information is CSI reporting information,

[0539] In the AI / ML related IE of the CSI reporting configuration, the reporting of indication information indicating model changes is predefined.

[0540] In some embodiments, the model indication information is sent via uplink signaling.

[0541] In some embodiments, the model indication information is carried via UCI, MAC CE, or RRC signaling.

[0542] In some embodiments, on the physical channel, the model indication information is sent via PUCCH or PUSCH.

[0543] In some embodiments, the indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0544] In some embodiments, the deactivation request information includes at least one of the following information:

[0545] Identification information of activated AI / ML features and / or functions;

[0546] Deactivation instruction information;

[0547] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0548] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0549] Report configuration identification information; and

[0550] Identification information of a resource or resource set configuration.

[0551] In some embodiments, the indication information includes activation request information for inactive AI / ML features and / or functions.

[0552] In some embodiments, the activation request information includes at least one of the following information:

[0553] Identification information for inactive AI / ML features and / or functions;

[0554] Request information or instructions for activating AI / ML;

[0555] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0556] Quality information about the signals used to activate AI / ML features and / or functions;

[0557] Report configuration identification information; and

[0558] Identification information of a resource or resource set configuration.

[0559] In some embodiments, the AI / ML feature includes at least one AI / ML function.

[0560] In some embodiments, the AI / ML features include at least one AI / ML feature group.

[0561] In some embodiments, the AI / ML feature set includes at least one AI / ML function.

[0562] In some embodiments, the AI / ML functionality is an AI / ML sub-feature.

[0563] In some embodiments, the AI / ML feature has an AI / ML feature identifier.

[0564] In some embodiments, the AI / ML function has an AI / ML function identifier, or the AI / ML sub-feature has an AI / ML sub-feature identifier.

[0565] In another embodiment, the information transmission device can be configured separately from the processor 2310. For example, the information transmission device can be configured as a chip connected to the processor 2310, and the function of the transmission device of the physical random access channel is realized through the control of the processor 2310.

[0566] In addition, as shown in Figure 23, network device 2300 may also include: a transceiver 2340 and an antenna 2350; wherein, the functions of the above components are similar to those in the prior art and are not described here in detail. It is worth noting that network device 2300 does not necessarily include all the components shown in Figure 23; in addition, network device 2300 may also include components not shown in Figure 23, and reference may be made to the prior art for details.

[0567] It can be seen from the above embodiments that the network device receives indication information related to AI / ML features and / or functions from the terminal device, so that the network device can timely and accurately grasp the actual performance of the AI / ML features and / or functions, and realize precise operation of AI / ML features and / or functions, thereby making full use of AI / ML functions to improve communication quality and improve communication system performance.

[0568] Embodiments of the tenth aspect

[0569] An embodiment of the present application provides a communication system, including the terminal device as described in the embodiment of the eighth aspect and / or the network device as described in the embodiment of the ninth aspect.

[0570] For example, the structure of the communication system can be referred to FIG1 .

[0571] As shown in Figure 1, the communication system 100 includes a network device 101 and a terminal device 102. The network device 101 is the same as the network device recorded in the embodiment of the ninth aspect, and the terminal device 102 is the same as the terminal device recorded in the embodiment of the eighth aspect. The repeated content will not be repeated.

[0572] The above devices and methods of the present application can be implemented by hardware or by a combination of hardware and software. The present application relates to such a computer-readable program that, when executed by a logic component, enables the logic component to implement the devices or components described above, or enables the logic component to implement the various methods or steps described above. The present application also relates to a storage medium for storing the above program, such as a hard disk, a magnetic disk, an optical disk, a DVD, a flash memory, etc.

[0573] The method / device described in conjunction with the embodiments of the present application can be directly embodied as hardware, a software module executed by a processor, or a combination of the two. For example, one or more of the functional block diagrams shown in Figure 20 and / or one or more combinations of functional block diagrams can correspond to various software modules of a computer program flow or to various hardware modules. These software modules can correspond to the various steps shown in Figure 5, respectively. These hardware modules can be implemented by solidifying these software modules, for example, using a field programmable gate array (FPGA).

[0574] The software module may be located in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. A storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium; or the storage medium may be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The software module may be stored in the memory of the mobile terminal or in a memory card that can be inserted into the mobile terminal. For example, if the device (such as a mobile terminal) uses a large-capacity MEGA-SIM card or a large-capacity flash memory device, the software module may be stored in the MEGA-SIM card or the large-capacity flash memory device.

[0575] One or more of the functional blocks and / or one or more combinations of functional blocks described in the accompanying drawings may be implemented as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or any appropriate combination thereof for performing the functions described in this application. One or more of the functional blocks and / or one or more combinations of functional blocks described in the accompanying drawings may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in communication with a DSP, or any other such configuration.

[0576] The present application has been described above in conjunction with specific embodiments. However, those skilled in the art should understand that these descriptions are merely illustrative and are not intended to limit the scope of protection of the present application. Those skilled in the art may make various modifications and variations to the present application based on the spirit and principles of the present application, and such modifications and variations are also within the scope of the present application.

[0577] Regarding the implementation methods including the above embodiments, the following additional notes are also disclosed:

[0578] 1. A method for transmitting information, the method comprising:

[0579] The terminal device sends indication information related to AI / ML features and / or functions to the network device.

[0580] 2. The method according to Note 1, wherein:

[0581] The indication information includes information about the AI / ML features and / or functions that the terminal device desires to use.

[0582] 3. The method according to Note 1 or 2, wherein:

[0583] The terminal device sends the indication information to the network device in one of the following situations:

[0584] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0585] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0586] The network device sends to the terminal device a measurement and judgment indication of the AI / ML features and / or functional operations related to the indication information; and

[0587] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0588] 4. The method according to Note 3, wherein:

[0589] The AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0590] 5. The method according to any one of Notes 1 to 4, wherein:

[0591] The indication information includes indication information of the AI / ML features and / or functions selected by the terminal device.

[0592] 6. The method according to Note 5, wherein:

[0593] The indication information includes at least one of the following information:

[0594] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0595] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0596] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0597] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0598] 7. The method according to any one of Notes 1 to 6, further comprising:

[0599] The terminal device receives measurement configuration information and / or reporting configuration information sent by the network device.

[0600] 8. The method according to Note 7, wherein:

[0601] The measurement configuration information includes at least one of a configuration of a CSI resource or resource set and a configuration of an RS.

[0602] 9. The method according to Note 7 or 8, wherein:

[0603] One configuration in the measurement configuration information and / or the reporting configuration information corresponds to the implementation of at least one AI / ML feature and / or function.

[0604] 10. The method according to any one of Notes 7 to 9, wherein:

[0605] The measurement configuration information and / or reporting configuration information includes evaluation indication information of AI / ML features and / or functions, or selection indication information of AI / ML features and / or functions.

[0606] 11. The method according to any one of Notes 7 to 10, wherein:

[0607] The measurement configuration information and / or reporting configuration information is configured by identifying the AI / ML feature and / or function.

[0608] 12. The method according to any one of Notes 7 to 11, wherein:

[0609] The AI / ML features and / or functions corresponding to the measurement configuration information and / or reporting configuration information are determined by the network device based on the AI / ML features and / or functions supported by the terminal device reported by the terminal device.

[0610] 13. The method according to any one of Notes 7 to 12, wherein:

[0611] The measurement configuration information and / or reporting configuration information is configured via RRC, or,

[0612] After the measurement configuration information and / or reporting configuration information is configured through RRC, at least one measurement configuration and / or reporting configuration is activated by MAC CE.

[0613] 14. The method according to any one of Notes 7 to 13, wherein:

[0614] The measurement configuration information and / or reporting configuration information includes AI / ML evaluation instruction information.

[0615] 15. The method according to any one of Notes 1 to 14, wherein:

[0616] The indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0617] 16. The method according to Note 15, wherein:

[0618] The indication information includes at least one of the following information:

[0619] identification information of the characteristics and / or functions to be switched;

[0620] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0621] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0622] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0623] 17. The method according to Note 15 or 16, wherein the method further comprises:

[0624] The terminal device determines whether to switch the AI / ML features and / or functions based on the evaluation index information of switching AI / ML features and / or functions configured on the network side.

[0625] 18. The method according to Note 17, wherein:

[0626] The evaluation index information includes at least one of a measurement index, a measurement threshold, a comparison index, and a comparison threshold.

[0627] 19. The method according to any one of Notes 1 to 18, wherein:

[0628] The indication information includes at least one of input configuration related information and output configuration related information corresponding to the AI / ML feature and / or function.

[0629] 20. The method according to any one of Notes 1 to 19, wherein:

[0630] The indication information includes model indication information.

[0631] 21. The method according to Note 20, wherein:

[0632] The model indication information includes change information of a model or a model group that implements the AI / ML feature and / or function.

[0633] 22. The method according to Note 21, wherein:

[0634] The change information of the model or model group includes at least one of the following information:

[0635] Information about changes to a model or group of models;

[0636] Identification information of the AI / ML features and / or functions and information indicating changes to the model or model group; and

[0637] Identification information of a model or model group.

[0638] 23. The method according to Note 22, wherein:

[0639] The information that the model or model group has changed is represented by bit information.

[0640] 24. The method according to Note 23, wherein:

[0641] A 1-bit value indicates that a model or model group has changed.

[0642] 25. The method according to Note 22, wherein:

[0643] The model or the models in the model group are bilateral model structures.

[0644] 26. The method according to any one of Notes 20 to 25, wherein:

[0645] The model indication information is carried in the relevant reporting information.

[0646] 27. The method according to Note 26, wherein:

[0647] The reported information is CSI reporting information,

[0648] In the AI / ML related IE of the CSI reporting configuration, the reporting of indication information indicating model changes is predefined.

[0649] 28. The method according to any one of Notes 20 to 25, wherein:

[0650] The terminal device sends the model indication information to the network device through uplink signaling.

[0651] 29. The method according to Note 28, wherein:

[0652] The model indication information is carried through UCI or MAC CE or RRC signaling.

[0653] 30. The method according to Note 28 or 29, wherein:

[0654] On the physical channel, the model indication information is sent through the PUCCH or PUSCH.

[0655] 31. The method according to any one of Notes 1 to 30, wherein:

[0656] The indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0657] 32. The method according to Note 31, wherein:

[0658] The deactivation request information includes at least one of the following information:

[0659] Identification information of activated AI / ML features and / or functions;

[0660] Deactivation instruction information;

[0661] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0662] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0663] Report configuration identification information; and

[0664] Identification information of a resource or resource set configuration.

[0665] 33. The method according to any one of Notes 1 to 32, wherein:

[0666] The indication information includes activation request information for inactive AI / ML features and / or functions.

[0667] 34. The method according to Note 33, wherein:

[0668] The activation request information includes at least one of the following information:

[0669] Identification information for inactive AI / ML features and / or functions;

[0670] Request information or instructions for activating AI / ML;

[0671] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0672] Quality information about the signals used to activate AI / ML features and / or functions;

[0673] Report configuration identification information; and

[0674] Identification information of a resource or resource set configuration.

[0675] 35. The method according to any one of Notes 1 to 34, wherein:

[0676] The AI / ML feature includes at least one AI / ML function.

[0677] 36. The method according to any one of Notes 1 to 35, wherein:

[0678] The AI / ML features include at least one AI / ML feature group.

[0679] 37. The method according to Note 36, wherein:

[0680] The AI / ML feature group includes at least one AI / ML function.

[0681] 38. The method according to any one of Notes 1 to 37, wherein:

[0682] The AI / ML functionality is an AI / ML sub-feature.

[0683] 39. The method according to any one of Notes 1 to 38, wherein:

[0684] The AI / ML feature has an AI / ML feature identifier.

[0685] 40. The method according to any one of Notes 1 to 39, wherein:

[0686] The AI / ML function has an AI / ML function identifier, or the AI / ML sub-feature has an AI / ML sub-feature identifier.

[0687] 41. A method for transmitting information, the method comprising:

[0688] The network device receives indication information related to AI / ML features and / or functions sent by the terminal device.

[0689] 42. The method according to Note 41, wherein:

[0690] The indication information includes information about the AI / ML features and / or functions that the terminal device desires to use.

[0691] 43. The method according to Note 41 or 42, wherein the method further comprises one of the following steps:

[0692] The network device sends a configuration and / or instruction to the terminal device to send the instruction information;

[0693] The network device sends to the terminal device configuration and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0694] The network device sends to the terminal device a measurement and judgment indication of the AI / ML features and / or functional operations related to the indication information; and

[0695] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0696] 44. The method according to Note 43, wherein:

[0697] The AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

[0698] 45. The method according to any one of Notes 41 to 44, wherein:

[0699] The indication information includes indication information of the AI / ML features and / or functions selected by the terminal device.

[0700] 46. ​​The method according to Note 45, wherein:

[0701] The indication information includes at least one of the following information:

[0702] Identification information of the AI / ML features and / or functions selected by the terminal device;

[0703] Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device;

[0704] Information on the level of judgment used in selecting AI / ML features and / or functionality; and

[0705] Quality information of the signals used to determine the selection of AI / ML features and / or functions.

[0706] 47. The method according to any one of Notes 41 to 46, further comprising:

[0707] The network device sends measurement configuration information and / or reporting configuration information to the terminal device.

[0708] 48. The method according to Note 47, wherein:

[0709] The measurement configuration information includes at least one of a configuration of a CSI resource or resource set and a configuration of an RS.

[0710] 49. The method according to Note 47 or 48, wherein:

[0711] One configuration in the measurement configuration information and / or the reporting configuration information corresponds to the implementation of at least one AI / ML feature and / or function.

[0712] 50. The method according to any one of Notes 47 to 49, wherein:

[0713] The measurement configuration information and / or reporting configuration information includes evaluation indication information of AI / ML features and / or functions, or selection indication information of AI / ML features and / or functions.

[0714] 51. The method according to any one of Notes 47 to 50, wherein:

[0715] The measurement configuration information and / or reporting configuration information is configured by identifying the AI / ML feature and / or function.

[0716] 52. The method according to any one of Notes 47 to 51, wherein:

[0717] The AI / ML features and / or functions corresponding to the measurement configuration information and / or reporting configuration information are determined by the network device based on the AI / ML features and / or functions supported by the terminal device reported by the terminal device.

[0718] 53. The method according to any one of Notes 47 to 52, wherein:

[0719] The measurement configuration information and / or reporting configuration information is configured via RRC, or,

[0720] After the measurement configuration information and / or reporting configuration information is configured through RRC, at least one measurement configuration and / or reporting configuration is activated by MAC CE.

[0721] 54. The method according to any one of Notes 47 to 53, wherein:

[0722] The measurement configuration information and / or reporting configuration information includes AI / ML evaluation instruction information.

[0723] 55. The method according to any one of Notes 47 to 54, wherein:

[0724] The indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

[0725] 56. The method according to Note 55, wherein:

[0726] The indication information includes at least one of the following information:

[0727] identification information of the characteristics and / or functions to be switched;

[0728] Configuration instructions corresponding to the AI / ML features and / or functions that are the switching targets;

[0729] Information about the degree of discretion in switching AI / ML features and / or functions; and

[0730] Quality information of the signal used to determine switching AI / ML features and / or functions.

[0731] 57. The method according to Note 55 or 56, wherein the method further comprises:

[0732] The network device configures the terminal device to switch evaluation indicator information of AI / ML features and / or functions.

[0733] 58. The method according to Note 57, wherein:

[0734] The evaluation index information includes at least one of a measurement index, a measurement threshold, a comparison index, and a comparison threshold.

[0735] 59. The method according to any one of Notes 41 to 58, wherein:

[0736] The indication information includes at least one of input configuration related information and output configuration related information corresponding to the AI / ML feature and / or function.

[0737] 60. The method according to any one of Notes 41 to 59, wherein:

[0738] The indication information includes model indication information.

[0739] 61. The method according to Note 60, wherein:

[0740] The model indication information includes change information of a model or a model group that implements the AI / ML feature and / or function.

[0741] 62. The method according to Note 61, wherein:

[0742] The change information of the model or model group includes at least one of the following information:

[0743] Information about changes to a model or group of models;

[0744] Identification information of the AI / ML features and / or functions and information indicating changes to the model or model group; and

[0745] Identification information of a model or model group.

[0746] 63. The method according to Note 62, wherein:

[0747] The information that the model or model group has changed is represented by bit information.

[0748] 64. The method according to Note 63, wherein:

[0749] A 1-bit value indicates that a model or model group has changed.

[0750] 65. The method according to Note 62, wherein:

[0751] The model or the models in the model group are bilateral model structures.

[0752] 66. The method according to any one of Notes 60 to 65, wherein:

[0753] The model indication information is carried in the relevant reporting information.

[0754] 67. The method according to Note 66, wherein:

[0755] The reported information is CSI reporting information,

[0756] In the AI / ML related IE of the CSI reporting configuration, the reporting of indication information indicating model changes is predefined.

[0757] 68. The method according to any one of Notes 60 to 65, wherein:

[0758] The model indication information is sent via uplink signaling.

[0759] 69. The method according to Note 68, wherein:

[0760] The model indication information is carried through UCI or MAC CE or RRC signaling.

[0761] 70. The method according to Note 68 or 69, wherein:

[0762] On the physical channel, the model indication information is sent through the PUCCH or PUSCH.

[0763] 71. The method according to any one of Notes 41 to 70, wherein:

[0764] The indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

[0765] 72. The method according to Note 71, wherein:

[0766] The deactivation request information includes at least one of the following information:

[0767] Identification information of activated AI / ML features and / or functions;

[0768] Deactivation instruction information;

[0769] Information regarding the degree of judgment regarding deactivation of AI / ML features and / or functions;

[0770] Quality information of the signal used to determine deactivation of AI / ML features and / or functions;

[0771] Report configuration identification information; and

[0772] Identification information of a resource or resource set configuration.

[0773] 73. The method according to any one of Notes 41 to 72, wherein:

[0774] The indication information includes activation request information for inactive AI / ML features and / or functions.

[0775] 74. The method according to Note 73, wherein:

[0776] The activation request information includes at least one of the following information:

[0777] Identification information for inactive AI / ML features and / or functions;

[0778] Request information or instructions for activating AI / ML;

[0779] Information regarding the degree of judgment regarding activation of AI / ML features and / or functions;

[0780] Quality information about the signals used to activate AI / ML features and / or functions;

[0781] Report configuration identification information; and

[0782] Identification information of a resource or resource set configuration.

[0783] 75. The method according to any one of Notes 41 to 74, wherein:

[0784] The AI / ML feature includes at least one AI / ML function.

[0785] 76. The method according to any one of Notes 41 to 75, wherein:

[0786] The AI / ML features include at least one AI / ML feature group.

[0787] 77. The method according to Note 76, wherein:

[0788] The AI / ML feature group includes at least one AI / ML function.

[0789] 78. The method according to any one of Notes 41 to 77, wherein:

[0790] The AI / ML functionality is an AI / ML sub-feature.

[0791] 79. The method according to any one of Notes 41 to 78, wherein:

[0792] The AI / ML feature has an AI / ML feature identifier.

[0793] 80. The method according to any one of Notes 41 to 79, wherein:

[0794] The AI / ML function has an AI / ML function identifier, or the AI / ML sub-feature has an AI / ML sub-feature identifier.

[0795] 81. A method for transmitting information, the method comprising:

[0796] The terminal device sends indication information of the AI / ML features and / or functions selected by the terminal device to the network device.

[0797] 82. A method for transmitting information, the method comprising:

[0798] The terminal device sends switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target to the network device.

[0799] 83. A method for transmitting information, the method comprising:

[0800] The terminal device sends model indication information related to AI / ML features and / or functions to the network device.

[0801] 84. A method for transmitting information, the method comprising:

[0802] The terminal device sends a request message to the network device to deactivate the activated AI / ML features and / or functions and / or to fall back to a non-AI / ML mode.

[0803] 85. A method for transmitting information, the method comprising:

[0804] The terminal device sends activation request information for inactive AI / ML features and / or functions to the network device.

[0805] 86. A method for transmitting information, the method comprising one of the following steps:

[0806] Configuration and / or instructions for the network device to send to the terminal device instructions related to AI / ML features and / or functions;

[0807] The network device sends to the terminal device configurations and / or instructions related to the AI / ML features and / or functions of the instruction information;

[0808] The network device sends to the terminal device a measurement and judgment indication of the AI / ML features and / or functional operations related to the indication information; and

[0809] The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML feature and / or functional operation related to the indication information.

[0810] 87. A method for verifying model performance, the method comprising:

[0811] The terminal device receives the test vector set and reporting configuration from the network device;

[0812] The terminal device inputs the test vectors in the test vector set into the model, and sends the output of the model to the network device according to the reporting configuration.

[0813] 88. The method according to Note 87, wherein:

[0814] The test vector set is carried by RRC.

[0815] 89. The method according to Note 87 or 88, wherein:

[0816] The correspondence between the test vectors in the test vector set and the model is indicated by the indication information sent by the network device, or,

[0817] The test vector set has an identification corresponding to the model.

[0818] 90. A method for verifying model performance, the method comprising:

[0819] The network device sends the test vector set to the terminal device; and

[0820] The network device receives the output of the model sent by the terminal device.

[0821] 91. The method according to Note 90, wherein:

[0822] The network device sends the test vector set to the terminal device via an RRC message.

[0823] 92. The method according to Note 90 or 91, wherein:

[0824] The correspondence between the test vectors in the test vector set and the model is indicated by the indication information sent by the network device, or,

[0825] The test vector set has an identification corresponding to the model.

[0826] 93. The method according to any one of Notes 90 to 92, further comprising:

[0827] The network device informs the terminal device of the purpose of the test vector set through a specific RRC message or other binding configuration, and sends relevant reporting configuration.

[0828] 94. A terminal device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the method as described in any one of Notes 1-40, 81-85, and 87-89.

[0829] 95. A network device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the method as described in any one of Notes 41-80, 86, 90-93.

[0830] 96. A communication system, comprising the terminal device described in Note 94 and / or the network device described in Note 95.

[0831] 97. A method for identifying an AI / ML model, the method comprising:

[0832] Whether the AI / ML model has been tested is indicated by partial bit information or digital information in the model identifier of the AI / ML model.

[0833] 98. The method according to Note 97, wherein:

[0834] The model identifier is a global identifier.

[0835] 99. The method according to Note 97 or 98, wherein:

[0836] Whether the test has been passed is indicated by a bit or a number in the bit string or number string of the model identifier.

[0837] 100. A method for identifying an AI / ML model, the method comprising:

[0838] For AI / ML models that do not have a global identifier, the network side will provide the global identifier of the AI / ML model after testing and verification that the preset conditions are met.

[0839] 101. The method according to note 100, wherein:

[0840] After the test verifies that the preset conditions are met, the network side provides the global identification of the AI / ML model through the model registration or model identification process.

[0841] 102. The method according to Note 100 or 101, wherein:

[0842] After testing and verifying that the preset conditions are met, the network device or terminal device sends relevant model information to the network element responsible for managing the global model identification, and obtains the global identification of the AI / ML model from the network element.

[0843] 103. A method for identifying an AI / ML model, the method comprising:

[0844] For an AI / ML model that is not registered on the network, whether the AI / ML model has been tested is indicated by the initial model identifier of the AI / ML model.

[0845] 104. The method according to any one of Notes 97 to 103, wherein:

[0846] The test includes at least one of an interoperability test, a RAN4 test, and a network access test.

Claims

1. An information transmission device, the device is arranged in a terminal device, the device include: A first sending unit is configured to send indication information related to AI / ML features and / or functions to a network device.

2. The device according to claim 1, in, The indication information includes information about the AI / ML features and / or functions that the terminal device expects to use.

3. The device according to claim 1, in, The first sending unit sends the indication information to the network device in one of the following situations: The network device sends to the terminal device a configuration and / or an instruction for sending the instruction information; The network device sends to the terminal device a configuration and / or indication related to the AI / ML feature and / or function of sending the indication information; The network device sends to the terminal device a measurement and judgment indication of the AI / ML features and / or functional operations related to the indication information; and The network device sends to the terminal device at least one of an indicator, a condition, and a threshold for measuring and judging the AI / ML features and / or functional operations related to the indication information.

4. The device according to claim 3, in, The AI / ML feature and / or function operation includes at least one of monitoring, selecting, switching, activating, deactivating, and falling back to a non-AI / ML mode of the AI / ML feature and / or function.

5. The device according to claim 1, in, The indication information includes indication information of the AI / ML features and / or functions selected by the terminal device.

6. The device according to claim 5, in, The indication information includes at least one of the following information: Identification information of the AI / ML features and / or functions selected by the terminal device; Configuration indication information corresponding to the selected AI / ML features and / or functions for the measurement configuration information and / or reporting configuration information sent by the network device; Information about the level of judgment used in selecting AI / ML features and / or functionality; and Quality information of the signals used to determine the selection of AI / ML features and / or functions.

7. The device according to claim 1, in, The device also includes: A first receiving unit is configured to receive measurement configuration information and / or reporting configuration information sent by the network device.

8. The device according to claim 7, in, The measurement configuration information includes at least one of a configuration of a CSI resource or a resource set and a configuration of an RS.

9. The device according to claim 1, in, The indication information includes switching request information of the AI / ML feature and / or function or activation request information of the AI / ML feature and / or function as the switching target.

10. The device according to claim 9, in, The indication information includes at least one of the following information: identification information of the characteristics and / or functions to be switched; Configuration indication information corresponding to the AI / ML features and / or functions that are the switching targets; Information about the level of discretion in switching AI / ML features and / or functions; and Quality information of the signal used to determine switching AI / ML features and / or functions.

11. The device according to claim 1, in, The indication information includes model indication information.

12. The device according to claim 11, in, The model indication information includes change information of a model or a model group that implements the AI / ML feature and / or function.

13. The device according to claim 12, in, The change information of the model or model group includes at least one of the following information: Information about changes to a model or group of models; Identification information of the AI / ML features and / or functions and indication information of changes to the model or model group; and Identification information for a model or model group.

14. The device according to claim 11, in, The model indication information is carried in the relevant reporting information.

15. The device according to claim 11, in, The first sending unit sends the model indication information to the network device through uplink signaling.

16. The device according to claim 11, in, The indication information includes deactivation request information for activated AI / ML features and / or functions and / or request information for falling back to a non-AI / ML mode.

17. The device according to claim 16, in, The deactivation request information includes at least one of the following information: Identification information of activated AI / ML features and / or functions; Deactivation instruction information; Information on the degree of judgment to deactivate AI / ML features and / or functions; Quality information of the signal used to determine deactivation of AI / ML features and / or functions; Report configuration identification information; and Identification information of a resource or resource set configuration.

18. The device according to claim 1, in, The indication information includes activation request information for inactive AI / ML features and / or functions.

19. The device according to claim 18, in, The activation request information includes at least one of the following information: Identification information of inactive AI / ML features and / or functions; Request information or instructions to activate AI / ML; Information about the degree of judgment regarding activation of AI / ML features and / or functions; Quality information of signals used to determine activation of AI / ML features and / or functions; Report configuration identification information; and Identification information of a resource or resource set configuration.

20. An information transmission device, the device being arranged in a network device, the device include: The second receiving unit receives indication information related to AI / ML features and / or functions sent by the terminal device.

Images