Beam management method and apparatus

Abstract

Embodiments of the present application provide a beam management method and apparatus. The method comprises: a terminal device receiving configuration information from a network device; and the terminal device, on the basis of the configuration information, measuring a reference signal so as to acquire training data and / or model inference data and / or performance monitoring data for an AI / ML functionality / model.

Description

Beam management method and device Technical Field

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

[0002] NR Release 18 investigates artificial intelligence / machine learning (AI / ML) over the air interface. AI / ML can be used for the following use cases: channel state information (CSI) feedback enhancement, beam management, and positioning enhancement. CSI feedback enhancement can include CSI prediction and CSI compression; beam management can include spatial beam prediction (BM case-1) and temporal beam prediction (BM case-2); and positioning enhancement can include direct positioning and AI / ML-assisted positioning.

[0003] In some sub-use cases, a two-sided model can be used, with the AI / ML model located on both the end device and the network equipment. In other sub-use cases, a one-sided model can be used, with the AI / ML model located on either the end device or the network equipment. For beam management, the AI / ML model can be located on the end device and / or the network equipment.

[0004] 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.

[0005] Summary of the Invention

[0006] The inventors have discovered that terminal devices and / or network devices can utilize AI / ML functionality / models to predict beams based on beam measurement results, but there is currently no clear solution for how to obtain the various data used for AI / ML.

[0007] To address at least one of the above problems, embodiments of the present application provide a beam management method and apparatus.

[0008] According to one aspect of an embodiment of the present application, a beam management method is provided, including:

[0009] The terminal device receives configuration information from the network device;

[0010] The terminal device measures the reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0011] According to another aspect of an embodiment of the present application, a beam management device is provided, including:

[0012] a receiving unit, configured to receive configuration information from a network device;

[0013] A processing unit measures a reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0014] According to another aspect of an embodiment of the present application, a beam management method is provided, including:

[0015] The network device sends configuration information to the terminal device;

[0016] The configuration information is used by the terminal device to measure a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0017] According to another aspect of an embodiment of the present application, a beam management device is provided, including:

[0018] a sending unit, configured to send configuration information to a terminal device;

[0019] The configuration information is used by the terminal device to measure a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0020] According to another aspect of an embodiment of the present application, a communication system is provided, including:

[0021] A network device that sends configuration information to a terminal device;

[0022] A terminal device measures a reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0023] One of the benefits of the embodiments of the present application is that, based on the configuration information, a terminal device measures a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / models. This improves the performance and efficiency of AI / ML.

[0024] 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.

[0025] 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.

[0026] 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

[0027] 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.

[0028] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application;

[0029] FIG2 is a schematic diagram of a beam management method according to an embodiment of the present application;

[0030] FIG3 is a schematic diagram of AI / ML for beam management according to an embodiment of the present application;

[0031] FIG4 is a schematic diagram of a beam management method according to an embodiment of the present application;

[0032] FIG5 is a schematic diagram of a beam management method according to an embodiment of the present application;

[0033] FIG6 is a schematic diagram of a beam management device according to an embodiment of the present application;

[0034] FIG7 is a schematic diagram of a beam management device according to an embodiment of the present application;

[0035] FIG8 is a schematic diagram of a terminal device according to an embodiment of the present application;

[0036] FIG9 is a schematic diagram of a network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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, New Radio (NR), future 6G, etc., and / or other communication protocols currently known or to be developed in the future.

[0042] 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.

[0043] 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), IAB host, etc., and may also include remote radio head (RRH, Remote Radio Head), remote radio unit (RRU, Remote Radio Unit), relay (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.

[0044] 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.

[0045] 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.

[0046] 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.

[0047] 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.

[0048] The following describes the scenarios of the embodiments of the present application through examples, but the present application is not limited thereto.

[0049] 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 terminal devices 102 and 103. For simplicity, FIG1 illustrates only two terminal devices and one network device as an example, but the embodiments of the present application are not limited thereto.

[0050] In the embodiment of the present application, existing services or future services can be transmitted between the network device 101 and the terminal devices 102 and 103. 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.

[0051] It is worth noting that FIG1 shows that both terminal devices 102 and 103 are within the coverage range of network device 101, but the present application is not limited thereto. Both terminal devices 102 and 103 may not be within the coverage range of network device 101, or one terminal device 102 may be within the coverage range of network device 101 while the other terminal device 103 is outside the coverage range of network device 101.

[0052] 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.

[0053] The performance of the AI / ML functions / models used for beam management needs to be monitored so that corresponding control of the AI / ML functions / models, such as activation / deactivation / selection / switching / fallback, can be performed.

[0054] For example, network-side monitoring can be performed, where the network monitors performance metrics and makes activation / deactivation / selection / switching / fallback decisions.

[0055] For another example, UE-side monitoring may be performed, where the UE monitors performance metrics and makes activation / deactivation / selection / switching / fallback decisions.

[0056] For another example, hybrid monitoring can be performed, where the terminal side monitors performance metrics, and the network side makes activation / deactivation / selection / switching / fallback decisions.

[0057] In embodiments of the present application, one or more AI / ML models may be configured and run in a network device and / or a terminal device. The AI / ML models may be used for various signal processing functions in wireless communications, such as CSI prediction, CSI compression, beamforming, positioning management, and the like; however, the present application is not limited thereto.

[0058] Embodiments of the first aspect

[0059] An embodiment of the present application provides a beam management method, which is described from the perspective of a terminal device.

[0060] FIG2 is a schematic diagram of a beam management method according to an embodiment of the present application. As shown in FIG2 , the method includes:

[0061] 201, the terminal device receives configuration information from the network device;

[0062] 202. The terminal device measures a reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0063] It is worth noting that FIG2 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 FIG2 above.

[0064] In some embodiments, functionality refers to an AI / ML feature / feature group enabled by a configuration, where the configuration is supported based on conditions indicated by UE capabilities.

[0065] For example, the AL / ML function may be one or more functions, or one or more logical models, or one or more sub-functions, or one or more features, or one or more feature groups.

[0066] For another example, the function can be to use AI / ML for spatial beam prediction, or to use AI / ML for time beam prediction, or to use AI / ML for CSI prediction, or to use AI / ML for direct positioning, or to use AI / ML for assisted positioning, and so on.

[0067] In some embodiments, an AI / ML functionality / model can be used for beam management. One or more reference signals are used for measurement, and the measurement results are input to the AI / ML functionality / model. Another one or more reference signals are used for inference at the output of the AI / ML functionality / model.

[0068] For ease of description, beam management based on AI / ML functionality / models is referred to as model inference, training data based on AI / ML functionality / models is referred to as training data collection (non-AI / ML methods can also be used for training data collection), and performance monitoring based on AI / ML functionality / models is referred to as performance monitoring.

[0069] In some embodiments, the configuration information may include configuration information of one or more reference signals, such as CSI-RS configuration information, etc. The present application is not limited thereto, and for specific configuration information, reference may be made to related technologies. The configuration information may include configuration information for training data collection, and / or configuration information for model inference, and / or configuration information for performance monitoring.

[0070] Figure 3 is a schematic diagram of AI / ML for beam management in an embodiment of the present application. As shown in Figure 3, one or more reference signals in the second reference signal resource set (set B) can be received and measured by the terminal device, and the measurement results can be used as input to the AI / ML. One or more reference signals in the first reference signal resource set (set A) can be used by the terminal device for the output of the AI / ML, for example, the measurement results can be used as label data or ground truth data for the AI / ML. For the specific content of AI / ML and set A and set B, please refer to the relevant technology and will not be repeated here.

[0071] FIG4 is another schematic diagram of the beam management method according to an embodiment of the present application, which is illustrated by taking a terminal device configured with AI / ML as an example. As shown in FIG4 , the method includes:

[0072] 401. A terminal device receives configuration information from a network device; for example, the configuration information includes a second reference signal resource set (set B) for beam measurement and a first reference signal resource set (set A) for beam prediction.

[0073] At 402 , the terminal device performs beam measurement and inputs the beam measurement results into an AI / ML functionality / model. For example, the measurement results of the reference signals in set B are used as input to the AI / ML, and the reference signals in set A are used for prediction (or inference).

[0074] 403. The terminal device sends the beam prediction result to the network device.

[0075] For example, the AI / ML function is located on the terminal device side. After the AI / ML function is enabled or activated, the terminal device performs beam measurement based on the reference signal from the network side, uses AI / ML to perform beam prediction based on the beam measurement results, and sends the prediction results to the network device.

[0076] It is worth noting that FIG4 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 FIG4 above.

[0077] The above schematically illustrates AI / ML-based beam management. The following describes data acquisition (collection).

[0078] In some embodiments, for performance monitoring of the AI / ML functionality / model, the method further includes: the terminal device reporting ground truth data for the measurement results.

[0079] For example, for performance monitoring of AI / ML functions / models, such as network-side monitoring or hybrid monitoring, the UE can report ground truth data of measurement results, such as L1-RSRP / L1-SINR, to the gNB. The input of the AI / ML function / model is the measurement results of the reference signals used for performance monitoring, and the output of the AI / ML function / model is the ground truth data.

[0080] In some embodiments, differential L1-RSRP is not applied in reporting ground truth data for performance monitoring.

[0081] For example, when reporting ground truth data for performance monitoring, differential L1-RSRP is not applied. The L1-RSRP step size (e.g., a step size less than 1 dB, or a step size greater than 1 dB) and the bit length of the L1-RSRP report can be predefined or configurable. For another example, a new UCI format can be introduced for non-differential L1-RSRP reporting.

[0082] In some embodiments, differential L1-RSRP is applied in reporting ground truth data for performance monitoring, and the granularity of the differential L1-RSRP is smaller than a predetermined value.

[0083] For example, when reporting ground truth data for performance monitoring, differential L1-RSRP is applied. Finer granularity (e.g., a step size less than 2 dB) (or greater granularity, e.g., a step size greater than 2 dB) can be applied to differential L1-RSRP reporting. The new step size can be predefined or configurable. For another example, a new UCI format can be introduced for differential L1-RSRP reporting with finer granularity.

[0084] In some embodiments, in reporting ground truth data for performance monitoring, the value range of L1-RSRP is greater than a predetermined value; wherein a first bit length (e.g., greater than an existing bit length or greater than a predetermined value) is used and / or a first UCI format (e.g., a new UCI format for the first bit length) is used and / or a first offset is used.

[0085] For example, when reporting ground truth data for performance monitoring, the existing value range of L1-RSRP ([-144-44] dBm) can be expanded. In one example, the bit length of the L1-RSRP value can be extended, and accordingly, the UCI format used for L1-RSRP reporting is changed. In another example, the existing bit length, e.g., 7 bits, is maintained, and an additional deviation (Δ) can be introduced. The L1-RSRP value in the report can be, for example, {the value of L1-RSRP using the existing bit length + Δ}.

[0086] For another example, when reporting ground truth data for performance monitoring, if differential L1-RSRP reporting is applied, the quantization step size (e.g., a step size less than 1 dB or a step size greater than 1 dB) and bit length of the L1-RSRP for the best beam (e.g., the beam with the highest L1-RSRP) may be predefined or configurable. Accordingly, the UCI format of the L1-RSRP report may be changed.

[0087] For another example, for ground truth data used for performance monitoring, floating point data may be used. The bit length may be predefined or configurable, such as 32 bits. In one example, differential L1-RSRP may be used, while in another example, differential L1-RSRP may not be used.

[0088] In some embodiments, a new reporting quantity, for example, called a first reporting quantity, may be introduced for reporting ground truth data for performance monitoring.

[0089] In some embodiments, the terminal device further determines whether the measurement result is used for the performance monitoring based on data quality.

[0090] For example, for performance monitoring of AI / ML functions / models, such as network-side monitoring or hybrid monitoring, it is possible to define whether measurement results should be considered valid for performance monitoring, i.e., taking data quality into account. For example, a threshold for data quality can be configured or predefined, such as for L1-RSRP or L1-SINR (or SNR) or hypothetical BLER.

[0091] For example, the UE also measures L1-RSRP or L1-SINR (or SNR) or assumed BLER to determine data quality. If the data quality is below a threshold, the measurement result is discarded and is not used for reporting or determining the performance of the AI / ML function / model.

[0092] The above schematically illustrates the ground truth data for performance monitoring. The following describes the ground truth data for training data collection.

[0093] In some embodiments, for the collection of training data for the AI / ML functionality / model, the method further includes: the terminal device reporting ground truth data for the measurement results.

[0094] For example, for training data collection for AI / ML functions / models, the UE can report ground truth data of measurement results, such as L1-RSRP / L1-SINR, to the gNB. The input to the AI / ML function / model is the measurement results of the reference signals used for training data collection, and the output of the AI / ML function / model is the ground truth data.

[0095] In some embodiments, differential L1-RSRP is not applied in reporting ground truth data for training data collection.

[0096] For example, when reporting ground truth data for training data collection, differential L1-RSRP is not applied. The L1-RSRP step size (e.g., a step size less than 1 dB, or a step size greater than 1 dB) and the bit length of the L1-RSRP report (e.g., greater than 7 bits) can be predefined or configurable. For another example, if ground truth data is carried via UCI, a new UCI format can be introduced for non-differential L1-RSRP reporting.

[0097] In some embodiments, differential L1-RSRP is applied in reporting of ground truth data for training data collection, and the granularity of the differential L1-RSRP is smaller than a predetermined value.

[0098] For example, when reporting ground truth data for training data collection, differential L1-RSRP is applied. Finer granularity (e.g., a step size less than 2 dB) (or greater granularity, e.g., a step size greater than 2 dB) can be applied to differential L1-RSRP reporting. The new step size can be predefined or configurable. For another example, if ground truth data is carried via UCI, a new UCI format can be introduced for differential L1-RSRP reporting with finer granularity.

[0099] In some embodiments, in the reporting of ground truth data for training data collection, the value range of L1-RSRP is greater than a predetermined value; wherein a first bit length is used and / or a first UCI format is used and / or a first offset is used.

[0100] For example, when reporting ground truth data for training data collection, the existing value range of L1-RSRP ([-144-44] dBm) can be expanded. In one example, the bit length of the L1-RSRP value can be extended. Accordingly, if the ground truth data is carried by UCI, the UCI format used for L1-RSRP reporting is changed. In another example, the existing bit length, such as 7 bits, is maintained, and an additional deviation (Δ) can be introduced. The L1-RSRP value in the report can be, for example, {the value of L1-RSRP using the existing bit length + Δ}.

[0101] For another example, when reporting ground truth data for training data collection, if differential L1-RSRP reporting is applied, the quantization step size (e.g., a step size less than 1 dB or a step size greater than 1 dB) and bit length (e.g., 7 bits) for the L1-RSRP of the best beam (e.g., the beam with the highest L1-RSRP) may be predefined or configurable. Accordingly, if the ground truth data is carried via UCI, the UCI format used for L1-RSRP reporting may be changed.

[0102] For another example, floating-point data can be used for ground truth data used for training data collection. The bit length can be predefined or configurable, such as 32 bits. In one example, differential L1-RSRP can be applied, while in another example, differential L1-RSRP may not be applied.

[0103] In some embodiments, a new report quantity (e.g., a first report quantity) and / or a new UCI (e.g., a first UCI, if the training data is reported via the UCI) may be introduced for reporting ground truth data for training data collection.

[0104] In some embodiments, for the collection of training data for AI / ML functions / models, the quantization scheme of the above embodiments can also be applied to the collection of model input data.

[0105] For example, if the training data is reported via UCI, a new reporting amount (e.g., called a second reporting amount) and / or a new UCI format (e.g., called a second UCI) may be introduced for the model input data. The reporting amount and / or the new UCI format of the model input data may be the same as the reporting amount and / or the new UCI format of the reference real data (ground truth data), or the reporting amount and / or the new UCI format of the model input data may be different from the reporting amount and / or the new UCI format of the reference real data (ground truth data).

[0106] In some embodiments, the quantization of the training data collection (including model input data and / or benchmark real data) can be the same as the quantization of the benchmark real data for performance monitoring. Alternatively, the quantization of the training data collection (including model input data and / or benchmark real data) can be different from the quantization of the benchmark real data for performance monitoring.

[0107] The above schematically illustrates data acquisition (collection). The following describes the life cycle management (LCM) operations of AI / ML functions / models.

[0108] In some embodiments, the AI / ML functionality / model for beam management is activated / deactivated via radio resource control (RRC) configuration, and / or the AI / ML functionality / model is switched via MAC CE and / or DCI.

[0109] In some embodiments, for the terminal side model, the terminal device reports model selection information via RRC and / or MAC CE and / or UCI. For example, if the reference signals of set A (set A) and set B (set B) for AI / ML inference are configured for the UE, the AI / ML function / model on the terminal side is activated.

[0110] In some embodiments, the AI / ML functionality / model is configured for the terminal device through radio resource control (RRC), and / or the AI / ML functionality / model is activated / deactivated through MAC CE and / or DCI, and / or the AI / ML functionality / model is switched through MAC CE and / or DCI.

[0111] For example, the AI / ML function / model may be configured to the UE via RRC configuration, and the MAC CE and / or DCI may be used to further activate / deactivate the AI / ML function / model. The MAC CE and / or DCI may also be used to perform switching of the AI / ML function / model. For another example, if the DCI is used to activate / deactivate the AI / ML function / model and / or switch the AI / ML function / model, the UE may feedback an acknowledgement of the DCI.

[0112] In one example, if the UE is configured with reference signals of set B and set A, the AI / ML function / model is configured. DCI can be used to further activate / deactivate the AI / ML function / model. The DCI can be a scheduling DCI or a non-scheduling DCI, such as DCI 0_1 / 0_2 / 1_1 / 2_2.

[0113] For example, some specific fields of DCI format 0_1 / 0_2 are set to predefined values, which can be used to activate the AI / ML function / model. After receiving the DCI for activation, the UE can perform AI / ML operations. Before receiving the DCI for activation or after receiving the DCI for deactivation, the AI / ML operation is disabled. In one example, the UE can fall back to the existing (legacy) operation, for example, the configuration related to set A can be ignored and only the configuration of set B can be considered.

[0114] For another example, the DCI format used to activate / deactivate an AI / ML function / model may be the same as the DCI format used to switch an AI / ML function / model. For example, in the DCI format, certain specific values ​​of one or more fields may indicate activation or deactivation, while another one or more fields may be used to indicate switching of the AI / ML function / model.

[0115] For another example, the DCI format used to activate / deactivate the AI / ML function / model may be different from the DCI format used to switch the AI / ML function / model. For example, a first DCI format is used to activate / deactivate the AI / ML function / model, and a second DCI format is used to switch the AI / ML function / model.

[0116] The above schematically illustrates the LCM operation of the AI / ML function / model. The following describes the reference signal configuration.

[0117] In some embodiments, for training data collection, the second reference signal used for model input data collection is configured with a 'repetition' parameter via RRC, and / or the first reference signal used for ground truth data collection is configured with a 'repetition' parameter via RRC.

[0118] For example, for training data collection, the reference signal for model input data collection (which can be the same as set B) can be configured by RRC to have a "repetition" factor (set to "off"). The reference signal for ground truth data collection (which can be the same as set A) can be configured by RRC to have a "repetition" factor (set to "off"). The reference signal for model input data collection and the reference signal for ground truth data collection are sent to the UE.

[0119] In some embodiments, for training data collection, the second reference signal used for model input data collection is configured or indicated in a TCI state, and / or, the first reference signal used for ground truth data collection is configured or indicated in a TCI state.

[0120] For example, a reference signal used for model input data collection can be configured to have a TCI state. A reference signal used for baseline real data collection can be configured to have a TCI state.

[0121] In some embodiments, the TCI state of the second reference signal used for collecting model input data is the same as the TCI state of the first reference signal used for collecting ground truth data.

[0122] For example, the TCI state of the reference signal used for model input data collection can be the same as the TCI state of the reference signal used for benchmark real data collection (e.g., QCLed with Type-D, i.e., the same UE Rx beam is applied).

[0123] For another example, the TCI state of the reference signal used for model input data collection may be different from the TCI state of the reference signal used for baseline real data collection.

[0124] In some embodiments, the time domain behavior of the reference signal used for training data collection can be aperiodic, periodic, or semi-persistent. The reference signal collected for model input data and the reference signal collected for baseline real data can be configured to have the same time domain behavior.

[0125] In some embodiments, for training data collection, the second reference signal for model input data collection and / or the first reference signal for ground truth data collection are configured through the CSI reporting setting (CSI-ReportConfig) or associated with the CSI reporting setting (CSI-ReportConfig).

[0126] In some embodiments, for training data collection, the second reference signal for model input data collection and / or the first reference signal for ground truth data collection are configured through CSI resource settings (CSI-ResourceConfig) or associated with CSI resource settings (CSI-ResourceConfig).

[0127] In some embodiments, for training data collection, a second reference signal for model input data collection and / or a first reference signal for ground truth data collection is configured via a CSI-RS / CSI-SSB resource set or associated with a CSI-RS / CSI-SSB resource set.

[0128] For example, new reporting metrics can be introduced for reporting model input data and / or ground truth data. Alternatively, existing reporting metrics can be reused, for example, if training data is not reported through UCI, the reporting metric can be set to 'none'.

[0129] For another example, one or more reference signals used for model input data collection and / or one or more reference signals used for ground truth data collection can be associated with the same (or different) CSI reporting settings / configurations (CSI-ReportConfig), or associated with the same (or different) CSI resource settings / configurations (CSI-ResourceConfig), or associated with the same (or different) CSI-RS resource sets / CSI-SSB resource sets.

[0130] For another example, if the training data is reported through UCI, the model input data and the ground truth data can be associated and reported together through UCI.

[0131] In some embodiments, for training data collection, a second reference signal for model input data collection and / or a first reference signal for ground truth data collection are configured via RRC IE.

[0132] For example, for training data collection, reference signals for model input data collection and / or reference signals for ground truth data collection are not configured / associated via CSI reporting settings / configurations (CSI ReportConfig), or CSI resource settings / configurations (CSI ResourceConfig), or CSI-RS resource sets / CSI-SSB resource sets. In this case, a new RRC IE can be introduced to configure reference signals for model input data collection and / or reference signals for ground truth data collection, for example, a new IE: CSI-RS-ResourceConfigDataCollection can be used.

[0133] In some embodiments, PDSCH rate matching is applied to the second reference signal used for model input data collection and / or the first reference signal used for ground truth data collection.

[0134] In some embodiments, PDSCH rate matching is not applied to the second reference signal used for model input data collection and / or the first reference signal used for ground truth data collection.

[0135] In some embodiments, the reference signal for model input data collection and / or the reference signal for baseline real data collection can be configured via the same RRC IE. In other embodiments, the reference signal for model input data collection and / or the reference signal for baseline real data collection can be configured via different RRC IEs.

[0136] The above schematically illustrates the situation of training data collection, and the following describes the situation of model inference.

[0137] In some embodiments, for model inference, the second reference signal set (set B) used for model input is configured with a 'repetition' parameter via RRC, and / or the first reference signal set (set A) used for model output is configured with a 'repetition' parameter via RRC.

[0138] For example, for model inference, the reference signal (set B) used for model input can be configured by RRC to have a "repetition" factor (set to "off"). The reference signal (set A) used for model output (or prediction) can be configured by RRC to not have a "repetition" factor, or the reference signal (set A) used for model output (or prediction) can be configured by RRC to have a "repetition" factor (set to "off"). The reference signal (set A) used for model output (or prediction) may not be sent to the UE.

[0139] In some embodiments, for model inference, a second reference signal set (set B) for model input is configured or indicated in a TCI state, and / or a first reference signal set (set A) for model output is configured or indicated in a TCI state.

[0140] In some embodiments, the TCI state of the second reference signal set (set B) for model input is the same as the TCI state of the first reference signal set (set A) for model output.

[0141] For example, the TCI state of the reference signal (set B) used for model input can be the same as the TCI state of the reference signal (set A) used for model output (or prediction) (e.g., QCLed with Type-D, i.e., the same UE Rx beam is applied).

[0142] As another example, the TCI state of the reference signal (set B) used for model input may be different from the TCI state of the reference signal (set A) used for model output (or prediction).

[0143] In some embodiments, for the network-side model (including BM case-1 and / or BM case-2), new reporting quantities and / or new UCI formats may be introduced. The new reporting quantities may be applied to the reference signal (set B) used for model input.

[0144] The above schematically illustrates the model inference process. The following describes the performance monitoring process.

[0145] In some embodiments, for performance monitoring, the second reference signal used for measurement is configured with a 'repetition' parameter via RRC, and / or the first reference signal used for monitoring is configured with a 'repetition' parameter via RRC.

[0146] For example, for performance monitoring, the reference signals used for measurement (which can be the same as set B) can be configured by RRC to have a "repetition" factor (set to "off"). The reference signals used for monitoring (which can be the same as set A or a subset of set A) can be configured by RRC to have a "repetition" factor (set to "off"). The reference signals used for measurement and the reference signals used for monitoring are sent to the UE.

[0147] In some embodiments, for performance monitoring, the second reference signal used for measurement is configured or indicates a TCI state, and / or the first reference signal used for monitoring is configured or indicates a TCI state.

[0148] For example, a reference signal for measurement may be configured to have a TCI state, and a reference signal for monitoring may be configured to have a TCI state.

[0149] In some embodiments, the TCI state of the second reference signal for measurement is the same as the TCI state of the first reference signal for monitoring.

[0150] For example, the TCI state of the reference signal used for measurement can be the same as the TCI state of the reference signal used for monitoring (e.g., QCLed with Type-D, i.e., the same UE Rx beam is applied).

[0151] For another example, the TCI state of the reference signal used for measurement may be different from the TCI state of the reference signal used for monitoring.

[0152] In some embodiments, for performance monitoring, the second reference signal for measurement and / or the first reference signal for monitoring is configured through a CSI reporting configuration (CSI-ReportConfig) or is associated with the CSI reporting configuration (CSI-ReportConfig).

[0153] In some embodiments, for performance monitoring, the second reference signal for measurement and / or the first reference signal for monitoring is configured through a CSI resource configuration (CSI-ResourceConfig) or is associated with the CSI resource configuration (CSI-ResourceConfig).

[0154] In some embodiments, for performance monitoring, the second reference signal for measurement and / or the first reference signal for monitoring is configured through a CSI-RS / CSI-SSB resource set or associated with a CSI-RS / CSI-SSB resource set.

[0155] In some embodiments, a new reporting amount may be introduced to report measurement results of the second reference signal for measurement and / or the first reference signal for monitoring. Alternatively, an existing reporting amount may be reused. For example, if measurement results do not need to be reported, the reporting amount may be set to 'none'.

[0156] For example, one or more reference signals for measurement and / or one or more reference signals for monitoring may be associated with the same (or different) CSI reporting setting / configuration (CSI-ReportConfig), or associated with the same (or different) CSI resource setting / configuration (CSI-ResourceConfig), or associated with the same (or different) CSI-RS resource set / CSI-SSB resource set.

[0157] In some embodiments, for performance monitoring, the second reference signal used for measurement and / or the first reference signal used for monitoring are not configured / associated via CSI reporting settings / configurations (CSI ReportConfig), or CSI resource settings / configurations (CSI ResourceConfig), or CSI-RS resource sets / CSI-SSB resource sets. In this case, a new RRC IE may be introduced to configure the second reference signal used for measurement and / or the first reference signal used for monitoring.

[0158] In some embodiments, for BM case-2 (time beam prediction), measurement reference signals can be configured to be transmitted at one or more time instances, and reference signals for performance monitoring can be configured to be transmitted at one or more time instances. In one example, different subsets of reference signals for performance monitoring can be transmitted at different time instances.

[0159] In some embodiments, new reporting quantities and / or new UCI formats may be introduced for network-side models and network-side monitoring (including BM Case-1 and / or BM Case-2). The new reporting quantities and / or new UCI formats may be applied to reference signals for measurement and / or reference signals for performance monitoring.

[0160] The following schematically illustrates the relationship between training data collection, model inference, and performance monitoring.

[0161] In some embodiments, at least some of the following reference signals are the same: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0162] In some embodiments, at least some of the following reference signals are different: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0163] In some embodiments, at least some of the following reference signals are configured by the same configuration information: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0164] In some embodiments, at least some of the following reference signals are configured by different configuration information: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0165] For example, the reference signals for all (or a subset) of the following operations may be the same or different; the reference signals for all (or a subset) of the following operations may be configured via the same configuration or via a separate configuration (e.g., CSI report setting / configuration (CSI ReportConfig), or CSI resource setting\configuration (CSI ResourceConfig), or CSI-RS resource set / CSI-SSB resource set); if the reporting amount is configured, the reporting amount may be the same or different for the reference signals for all (or a subset thereof) of the following operations.

[0166] - Reference signals used for model input data collection during training data collection;

[0167] -The reference signal used as input for the AI / ML model during model inference, i.e., set B;

[0168] - Reference signals used for measurement in performance monitoring (same as or different from set B, or a subset of set B, or same as or different from set A, or a subset of set A).

[0169] For another example, for performance monitoring, including terminal-side monitoring, hybrid monitoring, and network-side monitoring, new reporting quantities can be introduced. The reference signal used for measurement in performance monitoring can be aperiodic, periodic, or semi-persistent.

[0170] In some embodiments, at least some of the following reference signals are the same: a first reference signal used for collecting ground truth data in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0171] In some embodiments, at least some of the following reference signals are different: a first reference signal used for collecting ground truth data in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0172] In some embodiments, at least some of the following reference signals are configured by the same configuration information: a first reference signal used for collecting ground truth data in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0173] In some embodiments, at least some of the following reference signals are configured by different configuration information: a first reference signal used for collecting benchmark real data (ground truth data) in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0174] For example, the reference signals for all (or a subset) of the following operations may be the same or different; the reference signals for all (or a subset) of the following operations may be configured via the same configuration or via a separate configuration (e.g., CSI report setting / configuration (CSI ReportConfig), or CSI resource setting\configuration (CSI ResourceConfig), or CSI-RS resource set / CSI-SSB resource set); if the reporting amount is configured, the reporting amount may be the same or different for the reference signals for all (or a subset thereof) of the following operations.

[0175] - Reference signals used for ground truth data collection during training data collection;

[0176] -The reference signal used for AI / ML model output (or prediction) in model inference, i.e., set A;

[0177] - Reference signals used for monitoring in performance monitoring (the same as or different from set A, or a subset of set A; or the same as or different from set B, or a subset of set B).

[0178] For another example, for performance monitoring, including terminal-side monitoring, hybrid monitoring, and network-side monitoring, new reporting quantities can be introduced. The reference signal used for monitoring in performance monitoring can be aperiodic, periodic, or semi-persistent.

[0179] The embodiments of the present application can be applied to both the UE-side model and the gNB-side model, but the present application is not limited thereto. Furthermore, the AI / ML of the embodiments of the present application can be used for beam management, such as temporal beam prediction and / or spatial beam prediction, but the present application is not limited thereto. For example, non-AI / ML methods can also be used for data collection.

[0180] 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.

[0181] As can be seen from the above embodiments, the terminal device measures the reference signal based on the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / models. This can improve the performance and efficiency of AI / ML.

[0182] Embodiments of the second aspect

[0183] The embodiment of the present application provides a beam management method, which is described from the perspective of a network device. The embodiment of the second aspect can be combined with the embodiment of the first aspect, and the same contents as the embodiment of the first aspect will not be repeated.

[0184] FIG5 is a schematic diagram of a beam management method according to an embodiment of the present application. As shown in FIG16 , the method includes:

[0185] 501, the network device sends configuration information to the terminal device;

[0186] As shown in FIG5 , the method may further include:

[0187] 502. The terminal device measures a reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0188] 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.

[0189] 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.

[0190] As can be seen from the above embodiments, the terminal device measures the reference signal based on the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / models. This can improve the performance and efficiency of AI / ML.

[0191] Embodiments of the third aspect

[0192] The embodiment of the present application provides a beam management device. The device may be, for example, a terminal device, or one or more components or assemblies configured in the terminal device. The contents that are the same as those in the first and second aspects of the embodiment are not repeated here.

[0193] FIG6 is a schematic diagram of a beam management device according to an embodiment of the present application. As shown in FIG6 , the beam management device 600 according to the embodiment of the present application includes:

[0194] A receiving unit 601 receives configuration information from a network device;

[0195] The processing unit 602 measures the reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0196] In some embodiments, as shown in FIG6 , the apparatus may further include a sending unit 603 .

[0197] In some embodiments, for performance monitoring of the AI / ML functionality / model, the sending unit 603 reports ground truth data for the measurement results.

[0198] In some embodiments, differential L1-RSRP is not applied in reporting ground truth data for performance monitoring; or, differential L1-RSRP is applied in reporting ground truth data for performance monitoring, and the granularity of the differential L1-RSRP is less than a predetermined value.

[0199] In some embodiments, in reporting ground truth data for performance monitoring, a value range of L1-RSRP is greater than a predetermined value; wherein a first bit length is used and / or a first UCI format is used and / or a first offset is used.

[0200] In some embodiments, the processing unit 602 further determines whether the measurement result is used for the performance monitoring according to data quality.

[0201] In some embodiments, for the training data collection of the AI / ML functionality / model, the sending unit 603 reports the ground truth data for the measurement results.

[0202] In some embodiments, differential L1-RSRP is not applied in reporting of ground truth data for training data collection; or, differential L1-RSRP is applied in reporting of ground truth data for training data collection, and the granularity of the differential L1-RSRP is less than a predetermined value.

[0203] In some embodiments, in ground truth data reporting for training data collection, a value range of L1-RSRP is greater than a predetermined value; wherein a first bit length is used and / or a first UCI format is used and / or a first offset is used.

[0204] In some embodiments, the AI / ML functionality / model for beam management is activated / deactivated via radio resource control (RRC) configuration, and / or the AI / ML functionality / model is switched via MAC CE and / or DCI.

[0205] In some embodiments, for the terminal side model, the terminal device reports the model selection information via RRC and / or MAC CE and / or UCI.

[0206] In some embodiments, the AI / ML functionality / model is configured for the terminal device through radio resource control (RRC), and / or the AI / ML functionality / model is activated / deactivated through MAC CE and / or DCI, and / or the AI / ML functionality / model is switched through MAC CE and / or DCI.

[0207] In some embodiments, for training data collection, the second reference signal used for model input data collection is configured with a 'repetition' parameter via RRC, and / or the first reference signal used for ground truth data collection is configured with a 'repetition' parameter via RRC.

[0208] In some embodiments, for training data collection, the second reference signal used for model input data collection is configured or indicated in a TCI state, and / or, the first reference signal used for ground truth data collection is configured or indicated in a TCI state.

[0209] In some embodiments, the TCI state of the second reference signal used for collecting model input data is the same as the TCI state of the first reference signal used for collecting ground truth data.

[0210] In some embodiments, for training data collection, a second reference signal for model input data collection and / or a first reference signal for ground truth data collection are configured via a CSI reporting setting (CSI-ReportConfig).

[0211] In some embodiments, the second reference signal used for model input data collection and / or the first reference signal used for ground truth data collection are configured through CSI resource settings (CSI-ResourceConfig).

[0212] In some embodiments, the second reference signal used for model input data collection and / or the first reference signal used for ground truth data collection are configured via a CSI-RS / CSI-SSB resource set.

[0213] In some embodiments, for training data collection, a second reference signal for model input data collection and / or a first reference signal for ground truth data collection are configured via RRC IE.

[0214] In some embodiments, PDSCH rate matching is applied to the second reference signal used for model input data collection and / or the first reference signal used for ground truth data collection.

[0215] In some embodiments, PDSCH rate matching is not applied to the second reference signal used for model input data collection and / or the first reference signal used for ground truth data collection.

[0216] In some embodiments, for model inference, the second reference signal set (set B) used for model input is configured with a 'repetition' parameter via RRC, and / or the first reference signal set (set A) used for model output is configured with a 'repetition' parameter via RRC.

[0217] In some embodiments, for model inference, a second reference signal set (set B) for model input is configured or indicated in a TCI state, and / or a first reference signal set (set A) for model output is configured or indicated in a TCI state.

[0218] In some embodiments, the TCI state of the second reference signal set (set B) for model input is the same as the TCI state of the first reference signal set (set A) for model output.

[0219] In some embodiments, for performance monitoring, the second reference signal used for measurement is configured with a 'repetition' parameter via RRC, and / or the first reference signal used for monitoring is configured with a 'repetition' parameter via RRC.

[0220] In some embodiments, for performance monitoring, the second reference signal used for measurement is configured or indicates a TCI state, and / or the first reference signal used for monitoring is configured or indicates a TCI state.

[0221] In some embodiments, the TCI state of the second reference signal for measurement is the same as the TCI state of the first reference signal for monitoring.

[0222] In some embodiments, for performance monitoring, the second reference signal for measurement and / or the first reference signal for monitoring are configured through a CSI reporting configuration (CSI-ReportConfig).

[0223] In some embodiments, the second reference signal used for measurement and / or the first reference signal used for monitoring is configured through CSI resource configuration (CSI-ResourceConfig).

[0224] In some embodiments, the second reference signal for measurement and / or the first reference signal for monitoring are configured through a CSI-RS / CSI-SSB resource set.

[0225] In some embodiments, the second reference signal for measurement and / or the first reference signal for monitoring are configured through RRC IE.

[0226] In some embodiments, at least some of the following reference signals are the same: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0227] In some embodiments, at least some of the following reference signals are different: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0228] In some embodiments, at least some of the following reference signals are configured by the same configuration information: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0229] In some embodiments, at least some of the following reference signals are configured by different configuration information: a second reference signal used for model input data collection in training data collection, a second reference signal set (set B) used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

[0230] In some embodiments, at least some of the following reference signals are the same: a first reference signal used for collecting ground truth data in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0231] In some embodiments, at least some of the following reference signals are different: a first reference signal used for collecting ground truth data in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0232] In some embodiments, at least some of the following reference signals are configured by the same configuration information: a first reference signal used for collecting ground truth data in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0233] In some embodiments, at least some of the following reference signals are configured by different configuration information: a first reference signal used for collecting benchmark real data (ground truth data) in training data collection, a first reference signal set (set A) used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

[0234] 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.

[0235] It is worth noting that the above only describes the components or modules related to the present application, but the present application is not limited thereto. The beam management device 600 may also include other components or modules. For the specific contents of these components or modules, reference may be made to the relevant art.

[0236] In addition, for the sake of simplicity, FIG6 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.

[0237] As can be seen from the above embodiments, the terminal device measures the reference signal based on the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / models. This can improve the performance and efficiency of AI / ML.

[0238] Embodiments of the fourth aspect

[0239] The embodiment of the present application provides a beam management device. The device may be, for example, a network device, or one or more components or assemblies configured in the network device. The contents that are the same as those in the first to third aspects of the embodiment are not repeated here.

[0240] FIG7 is another schematic diagram of a beam management device according to an embodiment of the present application. As shown in FIG7 , the beam management device 700 includes:

[0241] A sending unit 701 sends configuration information to a terminal device; wherein the configuration information is used by the terminal device to measure a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0242] In some embodiments, as shown in FIG7 , the beam management device 700 may further include:

[0243] The receiving unit 702 receives data on the measurement result reported by the terminal device.

[0244] 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.

[0245] It is worth noting that the above only describes the components or modules related to the present application, but the present application is not limited thereto. The beam management device 700 may also include other components or modules. For the specific contents of these components or modules, reference may be made to the relevant art.

[0246] In addition, for the sake of simplicity, FIG7 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.

[0247] As can be seen from the above embodiments, the terminal device measures the reference signal based on the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / models. This can improve the performance and efficiency of AI / ML.

[0248] Embodiments of the fifth aspect

[0249] An embodiment of the present application also provides a communication system, and reference may be made to FIG1 . The contents that are the same as those in the first to fourth aspects of the embodiments will not be repeated.

[0250] In some embodiments, the communication system 100 may include at least:

[0251] A network device that sends configuration information to a terminal device;

[0252] A terminal device measures a reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0253] The embodiment of the present application also provides a terminal device, but the present application is not limited thereto and may also be other devices.

[0254] Figure 8 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in Figure 8 , terminal device 800 may include a processor 810 and a memory 820. Memory 820 stores data and programs and is coupled to processor 810. 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 or other functions.

[0255] For example, the processor 810 may be configured to execute a program to implement the beam management method as described in the embodiment of the first aspect. For example, the processor 810 may be configured to perform the following control: receiving configuration information from a network device; and measuring a reference signal based on the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0256] As shown in Figure 8 , the terminal device 800 may further include: a communication module 830, an input unit 840, a display 850, and a power supply 860. The functions of these components are similar to those in the prior art and are not described in detail here. It is worth noting that the terminal device 800 does not necessarily include all of the components shown in Figure 8 , and these components are not essential. Furthermore, the terminal device 800 may also include components not shown in Figure 8 , for which reference may be made to the prior art.

[0257] An embodiment of the present application further provides a network device, which may be, for example, a base station, but the present application is not limited thereto and may also be other network devices.

[0258] Figure 9 is a schematic diagram illustrating the structure of a network device according to an embodiment of the present application. As shown in Figure 9 , network device 900 may include a processor 910 (e.g., a central processing unit (CPU)) and a memory 920 ; the memory 920 is coupled to the processor 910 . The memory 920 may store various data and may also store an information processing program 930 , which is executed under the control of the processor 910 .

[0259] For example, the processor 910 may be configured to execute a program to implement the beam management method as described in the embodiment of the second aspect. For example, the processor 910 may be configured to perform the following control: sending configuration information to a terminal device; the configuration information is used by the terminal device to measure a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0260] In addition, as shown in Figure 9, network device 900 may further include: a transceiver 940 and an antenna 950; wherein, the functions of these components are similar to those in the prior art and are not further described here. It is worth noting that network device 900 does not necessarily include all the components shown in Figure 9; in addition, network device 900 may also include components not shown in Figure 9, and reference may be made to the prior art for details.

[0261] An embodiment of the present application also provides a computer program, wherein when the program is executed in a terminal device, the program enables the terminal device to perform the beam management method described in the embodiment of the first aspect.

[0262] An embodiment of the present application also provides a storage medium storing a computer program, wherein the computer program enables a terminal device to execute the beam management method described in the embodiment of the first aspect.

[0263] An embodiment of the present application also provides a computer program, wherein when the program is executed in a network device, the program enables the network device to perform the beam management method described in the embodiment of the second aspect.

[0264] An embodiment of the present application also provides a storage medium storing a computer program, wherein the computer program enables a network device to execute the beam management method described in the embodiment of the second aspect.

[0265] 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.

[0266] 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 the figure and / or one or more combinations of functional block diagrams can correspond to various software modules of the computer program flow or to various hardware modules. These software modules can respectively correspond to the various steps shown in the figure. These hardware modules can be implemented by solidifying these software modules, for example, using a field programmable gate array (FPGA).

[0267] 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.

[0268] 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.

[0269] 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.

[0270] Regarding the implementation methods including the above embodiments, the following additional notes are also disclosed:

[0271] 1. A beam management method, comprising:

[0272] The terminal device receives configuration information from the network device;

[0273] The terminal device measures the reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0274] 2. A beam management method, comprising:

[0275] The network device sends configuration information to the terminal device;

[0276] The configuration information is used by the terminal device to measure a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functionality / model.

[0277] 3. A terminal 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 beam management method as described in Note 1.

[0278] 4. 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 beam management method as described in Note 2.

[0279] 5. A computer program product, comprising at least a computer program, wherein when the computer program is executed by a processor, the terminal device executes the beam management method as described in Note 1.

[0280] 6. A computer program product, comprising at least a computer program, wherein when the computer program is executed by a processor, the network device executes the beam management method as described in Note 2.

Claims

1. A beam management device, comprising: a receiving unit, configured to receive configuration information from a network device; A processing unit measures a reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for the AI / ML function / model.

2. The device according to claim 1, wherein For performance monitoring of the AI / ML function / model, the apparatus further includes: A sending unit reports benchmark real data for the measurement result.

3. The device according to claim 2, wherein In reporting the benchmark real data for performance monitoring, differential L1-RSRP is not applied; or, in reporting the benchmark real data for performance monitoring, differential L1-RSRP is applied, and the granularity of the differential L1-RSRP is less than a predetermined value; And / or, in reporting of benchmark real data for performance monitoring, the value range of L1-RSRP is greater than a predetermined value; wherein, the first bit length is used and / or the first UCI format is used and / or the first offset is used.

4. The device according to claim 2, wherein The processing unit further determines whether the measurement result is used for the performance monitoring according to data quality.

5. The device according to claim 1, wherein For the collection of training data for the AI / ML function / model, the apparatus further comprises: A sending unit reports benchmark real data for the measurement result.

6. The device according to claim 5, wherein In reporting the baseline real data for training data collection, differential L1-RSRP is not applied; or, in reporting the baseline real data for training data collection, differential L1-RSRP is applied, and the granularity of the differential L1-RSRP is less than a predetermined value; And / or, in the benchmark real data reporting for training data collection, the value range of L1-RSRP is greater than a predetermined value; wherein the first bit length is used and / or the first UCI format is used and / or the first deviation is used.

7. The device according to claim 1, wherein Activate / deactivate the AI / ML function / model for beam management through radio resource control configuration, and / or switch the AI / ML function / model through MAC CE and / or DCI; For the terminal side model, the terminal device reports the model selection information through RRC and / or MAC CE and / or UCI; Alternatively, the AI / ML function / model is configured for the terminal device through radio resource control, and / or the AI / ML function / model is activated / deactivated through MAC CE and / or DCI, and / or the AI / ML function is switched through MAC CE and / or DCI. Features / Models.

8. The device according to claim 1, wherein For training data collection, the second reference signal used for model input data collection is configured with a 'repetition' parameter via RRC, and / or the first reference signal used for baseline real data collection is configured with a 'repetition' parameter via RRC.

9. The device according to claim 1, wherein For training data collection, the second reference signal used for model input data collection is configured or indicated in a TCI state, and / or the first reference signal used for benchmark real data collection is configured or indicated in a TCI state; The TCI state of the second reference signal used for collecting model input data is the same as the TCI state of the first reference signal used for collecting benchmark real data.

10. The device according to claim 1, wherein For training data collection, the second reference signal for model input data collection and / or the first reference signal for benchmark real data collection are configured via CSI reporting settings, Alternatively, the second reference signal for model input data collection and / or the first reference signal for reference real data collection are configured via CSI resource settings, Alternatively, the second reference signal used for collecting model input data and / or the first reference signal used for collecting benchmark real data are configured through a CSI-RS / CSI-SSB resource set.

11. The device according to claim 1, wherein For training data collection, a second reference signal for model input data collection and / or a first reference signal for benchmark real data collection are configured via RRC IE; PDSCH rate matching is applied to the second reference signal used for model input data collection and / or the first reference signal used for benchmark real data collection, or PDSCH rate matching is not applied to the second reference signal used for model input data collection and / or the first reference signal used for benchmark real data collection.

12. The device according to claim 1, wherein For model inference, the second reference signal set used for model input is configured with a 'repetition' parameter via RRC, and / or the first reference signal set used for model output is configured with a 'repetition' parameter via RRC.

13. The device according to claim 1, wherein For model inference, the second reference signal set for model input is configured or indicated as a TCI state, and / or the first reference signal set for model output is configured or indicated as a TCI state; The TCI state of the second reference signal set used for model input is the same as the TCI state of the first reference signal set used for model output.

14. The device according to claim 1, wherein For performance monitoring, the second reference signal used for measurement is configured with a 'repetition' parameter via RRC, and / or the first reference signal used for monitoring is configured with a 'repetition' parameter via RRC.

15. The device according to claim 1, wherein For performance monitoring, the second reference signal used for measurement is configured or indicates a TCI state, and / or the first reference signal used for monitoring is configured or indicates a TCI state; The TCI state of the second reference signal used for measurement is the same as the TCI state of the first reference signal used for monitoring.

16. The device according to claim 1, wherein For performance monitoring, the second reference signal for measurement and / or the first reference signal for monitoring is configured via CSI reporting settings, Alternatively, the second reference signal for measurement and / or the first reference signal for monitoring are configured through CSI resource setting, Alternatively, the second reference signal for measurement and / or the first reference signal for monitoring is configured through a CSI-RS / CSI-SSB resource set; Alternatively, the second reference signal for measurement and / or the first reference signal for monitoring are configured through RRC IE.

17. The device according to claim 1, wherein At least some of the following reference signals are the same: a second reference signal used for model input data collection in training data collection, a second reference signal set used for model input in model inference, and a second reference signal used for measurement in performance monitoring; Alternatively, at least some of the following reference signals are different: a second reference signal used for model input data collection in training data collection, a second reference signal set used for model input in model inference, and a second reference signal used for measurement in performance monitoring; Alternatively, at least some of the following reference signals are configured by the same configuration information: a second reference signal used for model input data collection in training data collection, a second reference signal set used for model input in model inference, and a second reference signal used for measurement in performance monitoring; Alternatively, at least some of the following reference signals are configured by different configuration information: a second reference signal used for model input data collection in training data collection, a second reference signal set used for model input in model inference, and a second reference signal used for measurement in performance monitoring.

18. The device according to claim 1, wherein At least some of the following reference signals are the same: a first reference signal used for benchmark real data collection in training data collection, a first reference signal set used for model output in model inference, and a first reference signal used for monitoring in performance monitoring; Alternatively, at least some of the following reference signals are different: A reference signal, a first reference signal set used for model output in model inference, and a first reference signal used for monitoring in performance monitoring; Alternatively, at least some of the following reference signals are configured by the same configuration information: a first reference signal used for benchmark real data collection in training data collection, a first reference signal set used for model output in model inference, and a first reference signal used for monitoring in performance monitoring; Alternatively, at least some of the following reference signals are configured by different configuration information: a first reference signal used for benchmark real data collection in training data collection, a first reference signal set used for model output in model inference, and a first reference signal used for monitoring in performance monitoring.

19. A beam management device, comprising: a sending unit, configured to send configuration information to a terminal device; The configuration information is used by the terminal device to measure a reference signal to obtain training data and / or model inference data and / or performance monitoring data for AI / ML functions / models.

20. A communication system comprising: A network device that sends configuration information to a terminal device; The terminal device measures the reference signal according to the configuration information to obtain training data and / or model inference data and / or performance monitoring data for the AI / ML function / model.

Images