Beam management method and device, user equipment and base station
The user equipment sends trigger requests and AI model input information to the base station. The base station instructs the AI model configuration, which solves the problem that the AI model cannot be determined when the AI model is located on the user equipment side, and achieves the enhancement of the requirements and efficiency improvement of AI-based beam management.
Patent Information
- Application Number
- CN202310518769.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-09
- Publication Date
- 2024-11-19
AI Technical Summary
When the AI model is on the user equipment side, the base station cannot determine which resources the user equipment needs to measure for beam management, resulting in the inability to effectively meet the AI-based beam management needs.
The user equipment sends the trigger request and input information of the AI model to the base station, informing the base station which resources it needs to measure. The base station indicates the AI model configuration to the user equipment based on this information, enhancing the UE's reporting content and the configuration content of the base station.
It has achieved enhanced configuration content of user equipment and base stations, met the needs of AI-based beam management, and improved the efficiency and effectiveness of beam management.
Smart Images

Figure CN118984490A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wireless communication technologies, and in particular to a beam management method, device, user equipment and base station. Background Art
[0002] The 3GPP (3rd Generation Partnership Project) R18 standard introduces AI (Artificial Intelligence)-based beam management. During the training and inference process of the AI model, the beam measurement results need to be used as the input of the AI model. When the AI model is located on the UE (User Equipment) side, the base station sends the reference signal to be measured, and the UE measures the beam quality as the input of the AI model, but at this time the base station does not know which resources the UE needs to measure. Summary of the invention
[0003] The beam management method, device, user equipment and base station provided by the present invention can enhance the reporting content of the UE and the configuration content of the base station to meet the actual needs of AI-based beam management.
[0004] In a first aspect, the present invention provides a beam management method, which is applied to a user equipment, and the method includes:
[0005] Sending a trigger request and / or input information of an AI model to a base station, wherein the trigger request indicates that the UE will perform an AI operation;
[0006] Receive the AI model configuration indicated by the base station.
[0007] Optionally, the trigger request includes an AI model operation type, and the operation type includes: model training, model inference, and model monitoring.
[0008] Optionally, the AI model configuration includes an output configuration of the AI model.
[0009] Optionally, the output configuration of the AI model includes: at least one output type or configuration identifier, wherein the identifier is used to indicate the output type or configuration.
[0010] Optionally, the output configuration of the AI model also includes:
[0011] Adjust the output type or configuration of a predefined AI model.
[0012] Optionally, the output configuration of the AI model includes at least one of the following information: an output type or configuration identifier;
[0013] The reference signal resource index to be adjusted;
[0014] The adjusted reference signal resource index;
[0015] The identifier is used to indicate the output type or configuration.
[0016] Optionally, the output configuration of the AI model also includes: reference signal resource configuration supported by the base station.
[0017] Optionally, the method further comprises:
[0018] An AI model output configuration operation type indication is received, where the indication is used to indicate the AI model output configuration being executed.
[0019] Optionally, the AI model configuration includes: an input-output association relationship of the AI model.
[0020] Optionally, the input-output association relationship of the AI model includes:
[0021] If the input beam is a subset of the output beam, the input-output association relationship of the AI model is indicated by the reference signal resource index during configuration, that is, reference signal resources with the same index indicate a corresponding relationship;
[0022] If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with at least one other resource, or by a configuration method in which one resource is associated with at least one resource group, and the resource group contains at least one resource.
[0023] Optionally, if the input beam is not a subset of the output beam, the input-output association relationship of the AI model is configured through DCI, MAC CE or RRC signaling.
[0024] Optionally, the input information includes at least one of the following information:
[0025] at least one pattern information;
[0026] The order of pattern transmission;
[0027] The number of resources required for configuration;
[0028] Alternative subset of resources.
[0029] Optionally, the input information further includes an input type indication, and the input type indication includes at least one of the following:
[0030] Fixed pattern transmission;
[0031] Transmit the patterns in the order in which they are transmitted;
[0032] Random transmission pattern;
[0033] Randomly transfer resources according to the required number of configured resources;
[0034] Transfer all resources in the candidate subset of resources;
[0035] Randomly transfer resources from the resource candidate subset according to the required number of configured resources.
[0036] In a second aspect, the present invention provides a beam management method, which is applied to a base station, and the method includes:
[0037] receiving a trigger request and / or input information of an AI model sent by a UE, wherein the trigger request indicates that the UE will perform an AI operation;
[0038] Send AI model configuration instructions to UE.
[0039] Optionally, the trigger request includes an AI model operation type, and the operation type includes: model training, model inference, and model monitoring.
[0040] Optionally, the AI model configuration includes an output configuration of the AI model.
[0041] Optionally, the output configuration of the AI model includes: at least one output type or configuration identifier, wherein the identifier is used to indicate the output type or configuration.
[0042] Optionally, the output configuration of the AI model also includes: adjusting the output type or configuration of a predefined AI model.
[0043] Optionally, the output configuration of the AI model includes at least one of the following:
[0044] Identification of the output type or configuration;
[0045] The reference signal resource index to be adjusted;
[0046] The adjusted reference signal resource index;
[0047] The identifier is used to indicate the output type or configuration.
[0048] Optionally, the output configuration of the AI model also includes: reference signal resource configuration supported by the base station.
[0049] Optionally, the method further comprises:
[0050] An AI model output configuration operation type indication is sent to the UE, where the indication is used to indicate the executed AI model output configuration.
[0051] Optionally, the AI model configuration includes: an input-output association relationship of the AI model.
[0052] Optionally, the input-output association relationship of the AI model includes:
[0053] If the input beam is a subset of the output beam, the input-output association relationship of the AI model is indicated by the reference signal resource index during configuration, that is, reference signal resources with the same index indicate a corresponding relationship;
[0054] If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with multiple other resources, or by a configuration method in which one resource is associated with at least one resource group, and the resource group contains at least one resource.
[0055] Optionally, if the input beam is not a subset of the output beam, the input-output association relationship of the AI model is configured through DCI, MAC CE or RRC signaling.
[0056] Optionally, the input information includes at least one of the following information:
[0057] at least one pattern information;
[0058] The order of pattern transmission;
[0059] The number of resources required for configuration;
[0060] Alternative subset of resources.
[0061] Optionally, the input information further includes an input type indication, and the input type indication includes at least one of the following:
[0062] Fixed pattern transmission;
[0063] Transmit the patterns in the order in which they are transmitted;
[0064] Random transmission pattern;
[0065] Randomly transfer resources according to the required number of configured resources;
[0066] Transfer all resources in the candidate subset of resources;
[0067] Randomly transfer resources from the resource candidate subset according to the required number of configured resources.
[0068] In a third aspect, the present invention provides a beam management device, the device being located in a user equipment, the device comprising:
[0069] A first sending unit, configured to send a trigger request and / or input information of an AI model to a base station, wherein the trigger request indicates that the UE will perform an AI operation;
[0070] The first receiving unit is used to receive the AI model configuration indicated by the base station.
[0071] In a fourth aspect, the present invention provides a beam management device, the device being located in a base station, and the device comprising:
[0072] A third receiving unit, configured to receive a trigger request sent by the UE and / or input information of the AI model, wherein the trigger request indicates that the UE will perform an AI operation;
[0073] The second sending unit is used to send an AI model configuration indication to the UE.
[0074] In a fifth aspect, the present invention provides a user equipment, the user equipment comprising:
[0075] at least one processor; and
[0076] a memory communicatively connected to the at least one processor; wherein,
[0077] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned beam management method applied to the user equipment.
[0078] In a sixth aspect, the present invention provides a base station, the base station comprising:
[0079] at least one processor; and
[0080] a memory communicatively connected to the at least one processor; wherein,
[0081] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned beam management method applied to the base station.
[0082] In a seventh aspect, the present invention provides a chip, comprising:
[0083] at least one processor; and
[0084] a memory communicatively connected to the at least one processor; wherein,
[0085] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned beam management method applied to the user equipment or the above-mentioned beam management method applied to the base station.
[0086] In an eighth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the above-mentioned beam management method applied to a user equipment or the above-mentioned beam management method applied to a base station is implemented.
[0087] The beam management method, device, user equipment and base station provided by the embodiments of the present invention, the user equipment informs the base station which resources the UE needs to measure by sending a trigger request and / or input information of the AI model to the base station. After the base station learns which resources the UE needs to measure, it instructs the UE to configure the AI model, thereby enhancing the reporting content of the UE and the configuration content of the base station to meet the actual needs of AI-based beam management. BRIEF DESCRIPTION OF THE DRAWINGS
[0088] Figure 1 This is a flow chart of a beam management method according to an embodiment of the present invention;
[0089] Figure 2 It is a structural schematic diagram of a beam management device according to an embodiment of the present invention;
[0090] Figure 3 It is a schematic diagram of the structure of a beam management device according to another embodiment of the present invention. DETAILED DESCRIPTION
[0091] Since analog beams can only send a limited number of shaped beams at the same time and the beam width is narrow, they can usually only cover a part of the cell. In order to achieve signal coverage of the entire cell, it is necessary to adopt a transmission method of joint scanning of multiple beams in the time domain to achieve complete coverage of the cell. When the transmit and receive beams between the base station and the UE are aligned, the maximum link gain can be obtained. The process of aligning the transmit and receive beams of the base station and the UE is called beam management.
[0092] The beam management process in the 3GPP R15 standard includes six processing processes: beam selection, beam measurement, beam reporting, beam switching, beam indication and beam recovery. Regarding beam reporting, the UE needs to detect and estimate the channel quality of multiple beams, but the UE does not need to report all measurement results to the base station. It only needs to select M beam measurement results for reporting. In the 3GPP R17 standard, in order to achieve beam enhancement in the Multi-TRP (multiple transmission point) scenario, the UE can report N groups of measurement results, and the beams in each group can be received by the UE at the same time.
[0093] During the training and inference process of the AI model, the beam measurement results need to be used as the input of the AI model. When the AI model is located on the UE side, the base station sends the reference signal that needs to be measured, and the UE measures and obtains the beam quality as the input of the AI model, but at this time the base station does not know which resources the UE needs to measure. Based on this, the present invention proposes a beam management method that can enhance the reporting content of the UE and the configuration content of the base station to meet the actual needs of AI-based beam management.
[0094] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0095] An embodiment of the present invention provides a beam management method, which is applied to a user equipment, such as Figure 1 As shown, the method includes:
[0096] S11. The UE sends a trigger request and / or input information of an AI model to the base station, where the trigger request indicates that the UE will perform an AI operation.
[0097] S12. The UE receives the AI model configuration indicated by the base station.
[0098] The beam management method provided in an embodiment of the present invention is that the user equipment informs the base station which resources the UE needs to measure by sending a trigger request and / or input information of the AI model to the base station. After the base station learns which resources the UE needs to measure, it instructs the UE to configure the AI model, thereby enhancing the reporting content of the UE and the configuration content of the base station to meet the actual needs of AI-based beam management.
[0099] The beam management method of the present invention is described in detail below with reference to specific embodiments.
[0100] (1) UE reports input type
[0101] The UE first sends a trigger request to the base station, where the trigger request indicates that the UE will perform an AI operation.
[0102] Among them, AI operations may include data collection (for model training), model inference, model monitoring, etc.
[0103] For example, the UE can use a 2-bit trigger request message to indicate the type of AI operation that the UE will perform, where 00 indicates data collection, 01 indicates model inference, and 10 indicates model detection. Alternatively, the trigger request sent by the UE to the base station can also be used only to notify the base station that an AI operation is to be performed. For example, a 1-bit trigger request message is used to indicate whether an AI operation is to be performed, 0 indicates no AI operation and the original beam management process is still used, and 1 indicates that an AI operation is to be performed and the base station side needs to send down relevant configuration information.
[0104] The UE may also send input information of the AI model to the base station. The input information of the AI model may include at least one of the following information: at least one pattern information, a pattern transmission order, a required number of configured resources, and a resource candidate subset.
[0105] Furthermore, the input information of the AI model may also include an input type indication, and the UE may support multiple input types, as follows:
[0106] Option 1: The input is a fixed pattern, that is, the UE's AI model input is fixed;
[0107] Option 2: The input is variable, which can be divided into four types:
[0108] Option 2-A: The input is transformed in a pre-configured pattern in a predetermined order;
[0109] Option 2-B: The input is randomly transformed in a pre-configured pattern:
[0110] Option 2-C: Randomly select resources output by the base station as model input;
[0111] Option 2-D: Randomly select from a subset of resource candidates.
[0112] If the UE can support the above multiple input types, it is necessary to inform the base station which input type currently needs to be configured.
[0113] Specifically, the UE can report the input type by reporting the input type ID, for example: ID = 1 represents option 1, ID = 2 represents option 2-A, ID = 3 represents option 2-B, ID = 4 represents option 2-C, and ID = 5 represents option 2-D. The specific contents corresponding to each input type are as follows:
[0114] Option 1: The UE reports pattern information and / or input type indication ID = 1, and the UE uses a bitmap to indicate the required configuration resources. Each bit in the bitmap indicates whether the corresponding resource needs to be configured;
[0115] Option 2-A: The UE reports the pattern transmission order, as well as the pattern information combination and / or input type indication.
[0116] ID=2
[0117] Among them, the pattern information combination UE indicates the resources required to be configured for each pattern by using multiple bitmaps, and the pattern transmission order can be a sequence of multiple pattern indexes. For example, the transmission order is 2451, and the base station configures the resources included in the pattern in the order of pattern 2, pattern 4, pattern 5, and pattern 1;
[0118] Option 2-B: The UE reports a pattern information combination (same as option 2-A) and / or an input type indication ID = 3. The base station randomly selects a pattern each time and configures the resources contained in the pattern for transmission;
[0119] Option 2-C: The UE reports the number of required configuration resources N and / or the input type indication ID = 4, and the base station randomly selects N from the supported beams for configuration;
[0120] Option 2-D: UE reports candidate subset and / or input type indication ID=5;
[0121] Furthermore, the UE may also indicate the number of configured resources N, and the base station randomly configures N resources in the candidate subset for transmission; or, the base station configures all resources in the candidate subset, and the UE selects the best N resources from the candidate subset as AI model input.
[0122] It should be noted that the UE may report the input type in the early stage, not necessarily before the model inference. Therefore, the above input information (at least one pattern information, pattern transmission order, required number of configured resources, resource candidate subset) and input type indication may be carried in different reporting information.
[0123] Therefore, the UE controls the AI model on the UE side by sending a trigger request and / or input information of the AI model to the base station. In addition, it can further inform the base station which resources the UE needs to measure, so that the base station can perform corresponding resource configuration according to the resources that the UE needs to measure, thereby meeting the actual needs of AI-based beam management.
[0124] (2) Base station indication model output
[0125] After receiving the trigger request reported by the UE and / or the input information of the AI model, the base station configures the model output, which may include the following methods:
[0126] Method 1: The base station predefines multiple model output types / configurations. For example, multiple input-output relationships are defined during the model training phase.
[0127] The base station can directly indicate the output configuration index, which may be the ID of the output type or the ID of a certain configuration;
[0128] The index is a combination of multiple configurations, for example, ID 1 and ID 2, where each ID may correspond to a different configuration type, for example, the number of beams supported horizontally and vertically, etc. (indicating index ID 1=2, ID 2=3).
[0129] Table 1 Number of horizontal beams
[0130] index quantity 1 8 2 16 3 32
[0131] Table 2 Number of vertical beams
[0132] index quantity 1 2 2 4 3 8
[0133] Method 2: The base station can fine-tune a certain output type. For example, in some scenarios, the beam formation on the base station side has changed and the same one cannot be found in the predefined output types. In this case, the predefined output type needs to be adjusted.
[0134] Specifically, the base station can output the configuration index, the beam index that needs to be adjusted, and the adjusted beam index, wherein the configuration index refers to the description in the above method 1 and will not be repeated here.
[0135] Method 3: The output type of the AI model does not belong to any of the previously trained ones, or the UE requests data collection operations for the AI model. At this time, the base station side directly configures the supported RS (Reference Signal) resources.
[0136] Among the above-mentioned methods, the base station and the UE may support only one of them, or may support multiple methods at the same time. The multiple methods here may be all of the above methods or some of them. The base station clearly indicates or implies through signaling which specific method the UE executes. The signaling here includes DCI (Downlink Control Information), MACCE (Medium Access Control Element) or RRC (Radio Resource Control), etc.
[0137] Therefore, the base station learns that the UE is going to perform an AI operation based on the trigger request sent by the UE, and thus indicates to the UE the AI model configuration supported by the base station side, aligning the requirements of the UE and the base station side for the AI model and meeting the actual needs of AI-based beam management.
[0138] (3) Base station configuration input and output association
[0139] During the AI model training phase, the base station can predefine the input and output association relationship of the AI model, which can be:
[0140] a. If the input beam is a subset of the output beam, the base station side indicates the association relationship between the UE input and output through the resource index (index) during configuration.
[0141] For example, the beams supported by the base station side include RS index = 1, RS index = 2, ... RS index = 8, and the input of the model is RS index = 1, RS index = 3, RS index = 5, RS index = 7. At this time, the same index indicates that there is a corresponding relationship between resources.
[0142] b. If the input beam is not a subset of the output beam, the base station needs to additionally configure the association between the input and output.
[0143] Optionally, the base station can indicate the association relationship between them by configuring one resource to associate with multiple other resources through DCI, MAC CE or RRC signaling; wherein, one resource refers to a resource that can be used as a model input; multiple other resources refer to certain resources that can be used as model outputs or supported by the base station configuration.
[0144] The specific association method can be:
[0145] Method 1: Divide the output resources into multiple groups. Each group contains one or more resources. When configuring the input resources, one or more groups will be associated.
[0146] Method 2: Directly associate one or more resources when configuring input resources.
[0147] Furthermore, if the association relationship changes, it can be modified through DCI, MAC CE or RRC signaling.
[0148] Furthermore, the base station may also receive input information of the AI model reported by the UE, and configure measurement resources for the UE according to the input information of the AI model.
[0149] Among them, the input information of the AI model may include at least one of the following information: at least one pattern information, the pattern transmission order, the required number of configuration resources, and a resource alternative subset.
[0150] Furthermore, the input information indication of the AI model may also include an input type indication of the AI model. When the input type indication is a fixed pattern, resources are configured for the UE according to the fixed pattern.
[0151] When the input type indicates that the pattern combination is configured in a pattern transmission order, the measurement resources corresponding to the patterns in the pattern combination are sent to the UE according to the pattern transmission order;
[0152] When the input type indicates a pattern combination, randomly selecting a measurement resource corresponding to a pattern from the pattern combination and sending it to the UE;
[0153] When the input type indicates the required number of configured resources, N beams are randomly selected from the supported beams and sent to the UE;
[0154] When the input type is indicated as a candidate resource subset, a predetermined number of resources are randomly selected from the candidate resource subset for configuration and sent to the UE, where the predetermined number is indicated by the UE; alternatively, all resources in the candidate resource subset are configured and sent to the UE, and the UE selects the optimal predetermined number of resources as the input of the AI model.
[0155] It should be noted that the UE can report the input type in the early stage, not necessarily before the model inference. Therefore, the above input information (at least one pattern information, pattern transmission order, required number of configured resources, resource candidate subset) and input type indication may be carried in different reporting information.
[0156] Since the base station pre-configures the input-output association relationship of the AI model, the UE can obtain the output configuration of the AI model based on the input information of the AI model and the input-output association relationship of the AI model, thereby meeting the actual needs of AI-based beam management.
[0157] The embodiment of the present invention also provides a beam management device, which is located in a user equipment, such as Figure 2 As shown, the device comprises:
[0158] A first sending unit 11 is configured to send a trigger request and / or input information of an AI model to a base station, where the trigger request indicates that the UE will perform an AI operation;
[0159] The first receiving unit 12 is used to receive the AI model configuration indicated by the base station.
[0160] The beam management device provided in an embodiment of the present invention is a user equipment that informs the base station which resources the UE needs to measure by sending a trigger request and / or input information of the AI model to the base station. After the base station learns which resources the UE needs to measure, it instructs the UE to configure the AI model, thereby enhancing the reporting content of the UE and the configuration content of the base station to meet the actual needs of AI-based beam management.
[0161] Optionally, the trigger request includes an AI model operation type, which includes: model training, model inference, and model monitoring.
[0162] Optionally, the AI model configuration includes an output configuration of the AI model.
[0163] Optionally, the AI model is used to perform beam management.
[0164] Optionally, the output configuration of the AI model includes: at least one output type or configuration identifier, which is used to indicate the output type or configuration.
[0165] Optionally, the output configuration of the AI model also includes:
[0166] Adjust the output type or configuration of a predefined AI model.
[0167] Optionally, the output configuration of the AI model includes at least one of the following information:
[0168] Identification of the output type or configuration;
[0169] The reference signal resource index to be adjusted;
[0170] The adjusted reference signal resource index;
[0171] This flag is used to indicate the output type or configuration.
[0172] Optionally, the output configuration of the AI model also includes: reference signal resource configuration supported by the base station.
[0173] Optionally, the device further comprises:
[0174] The second receiving unit is used to receive an AI model output configuration operation type indication, which is used to indicate the executed AI model output configuration.
[0175] Optionally, the AI model configuration includes: an input-output association relationship of the AI model.
[0176] Optionally, the input-output association relationship of the AI model includes:
[0177] If the input beam is a subset of the output beam, the input-output association relationship of the AI model is indicated by the reference signal resource index during configuration, that is, reference signal resources with the same index indicate a corresponding relationship;
[0178] If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with multiple other resources, or by a configuration method in which one resource is associated with at least one resource group, and the resource group contains at least one resource.
[0179] Optionally, if the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with at least one other resource when the downlink control message DCI, the media access control element MAC CE or the radio resource control RRC signaling is configured, or by a configuration method in which one resource is associated with at least one resource group.
[0180] Optionally, the input information includes at least one of the following information:
[0181] at least one pattern information;
[0182] The order of pattern transmission;
[0183] The number of resources required for configuration;
[0184] Alternative subset of resources.
[0185] Optionally, the input information further includes an input type indication, and the input type indication includes:
[0186] Fixed pattern transmission; or,
[0187] Transmit the patterns in the order in which they were transmitted; or,
[0188] Random transmission pattern; or,
[0189] Randomly transfer resources according to the number of resources required; or,
[0190] Transfer all resources in the resource candidate subset; or,
[0191] Randomly transfer resources from the resource candidate subset according to the required number of configured resources.
[0192] The device of this embodiment can be used to execute the technical solution of the above-mentioned method embodiment applied to user equipment. Its implementation principle and technical effect are similar and will not be repeated here.
[0193] The embodiment of the present invention also provides a beam management device, which is located in a base station, such as Figure 3 As shown, the device comprises:
[0194] A third receiving unit 21 is configured to receive a trigger request sent by a UE and / or input information of an AI model, where the trigger request indicates that the UE will perform an AI operation;
[0195] The second sending unit 22 is used to send an AI model configuration indication to the UE.
[0196] The beam management device provided by an embodiment of the present invention, the base station receives the trigger request sent by the UE and / or the input information of the AI model, learns which resources the UE needs to measure, and instructs the UE on the AI model configuration, thereby enhancing the reporting content of the UE and the configuration content of the base station to meet the actual needs of AI-based beam management.
[0197] Optionally, the trigger request includes an AI model operation type, which includes: model training, model inference, and model monitoring.
[0198] Optionally, the AI model configuration includes an output configuration of the AI model.
[0199] Optionally, the output configuration of the AI model includes: at least one output type or configuration identifier, which is used to indicate the output type or configuration.
[0200] Optionally, the output configuration of the AI model also includes: adjusting the output type or configuration of a predefined AI model.
[0201] Optionally, the output configuration of the AI model includes at least one of the following:
[0202] Identification of the output type or configuration;
[0203] The reference signal resource index to be adjusted;
[0204] The adjusted reference signal resource index;
[0205] This flag is used to indicate the output type or configuration.
[0206] Optionally, the output configuration of the AI model also includes: reference signal resource configuration supported by the base station.
[0207] Optionally, the device further comprises:
[0208] The third sending unit is used to send an AI model output configuration operation type indication to the UE, where the indication is used to indicate the executed AI model output configuration.
[0209] Optionally, the AI model configuration includes: an input-output association relationship of the AI model.
[0210] Optionally, the input-output association relationship of the AI model includes:
[0211] If the input beam is a subset of the output beam, the input-output association relationship of the AI model is indicated by the reference signal resource index during configuration, that is, reference signal resources with the same index indicate a corresponding relationship;
[0212] If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with at least one other resource, or by a configuration method in which one resource is associated with at least one resource group, and the resource group contains at least one resource.
[0213] Optionally, if the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with at least one other resource, or by a configuration method in which one resource is associated with at least one resource group, when DCI, MAC CE or RRC signaling is configured.
[0214] Optionally, the input information includes at least one of the following information:
[0215] at least one pattern information;
[0216] The order of pattern transmission;
[0217] The number of resources required for configuration;
[0218] Alternative subset of resources.
[0219] Optionally, the input information further includes an input type indication, and the input type indication includes:
[0220] Fixed pattern transmission; or,
[0221] Transmit the patterns in the order in which they were transmitted; or,
[0222] Random transmission pattern; or,
[0223] Randomly transfer resources according to the number of resources required; or,
[0224] Transfer all resources in the resource candidate subset; or,
[0225] Randomly transfer resources from the resource candidate subset according to the required number of configured resources.
[0226] The device of this embodiment can be used to execute the technical solution of the above-mentioned method embodiment applied to the base station. Its implementation principle and technical effect are similar and will not be repeated here.
[0227] An embodiment of the present invention further provides a user equipment, the user equipment comprising:
[0228] at least one processor; and
[0229] a memory communicatively connected to the at least one processor; wherein,
[0230] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned beam management method applied to the user equipment.
[0231] An embodiment of the present invention further provides a base station, the base station comprising:
[0232] at least one processor; and
[0233] a memory communicatively connected to the at least one processor; wherein,
[0234] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned beam management method applied to the base station.
[0235] An embodiment of the present invention further provides a chip, the chip comprising:
[0236] at least one processor; and
[0237] a memory communicatively connected to the at least one processor; wherein,
[0238] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned beam management method applied to the user equipment or the above-mentioned beam management method applied to the base station.
[0239] An embodiment of the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the above-mentioned beam management method applied to a user device or the above-mentioned beam management method applied to a base station is implemented.
[0240] Those skilled in the art can understand that all or part of the processes in the above method embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes in the above method embodiments. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.
[0241] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A beam management method, the method being applied to a user equipment UE, characterized in that: The method comprises: Sending a trigger request and / or input information of an artificial intelligence (AI) model to a base station, wherein the trigger request indicates that the UE will perform an AI operation; Receive the AI model configuration indicated by the base station.
2. The method according to claim 1, characterized in that The trigger request includes an AI model operation type, and the operation type includes: model training, model inference, and model monitoring.
3. The method according to claim 1 or 2, characterized in that: The AI model configuration includes the output configuration of the AI model.
4. The method according to claim 3, characterized in that The output configuration of the AI model includes: at least one output type or configuration identifier, where the identifier is used to indicate the output type or configuration.
5. The method according to claim 3, characterized in that: The output configuration of the AI model also includes: adjusting the output type or configuration of a predefined AI model.
6. The method according to claim 3 or 5, characterized in that: The output configuration of the AI model includes at least one of the following information: Identification of the output type or configuration; The reference signal resource index to be adjusted; The adjusted reference signal resource index; The identifier is used to indicate the output type or configuration.
7. The method according to claim 3, characterized in that The output configuration of the AI model also includes: reference signal resource configuration supported by the base station.
8. The method according to claim 1 or 2, further comprising: An AI model output configuration operation type indication is received, where the indication is used to indicate the AI model output configuration being executed.
9. The method according to claim 1 or 2, characterized in that: The AI model configuration includes: the input and output association relationship of the AI model.
10. The method according to claim 9, characterized in that The input-output association relationship of the AI model includes: If the input beam is a subset of the output beam, the input-output association relationship of the AI model is indicated by the reference signal resource index during configuration, that is, reference signal resources with the same index indicate a corresponding relationship; If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with at least one other resource, or by a configuration method in which one resource is associated with at least one resource group, and the resource group contains at least one resource.
11. The method according to claim 10, characterized in that If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is configured through a downlink control message DCI, a media access control element MAC CE or a radio resource control RRC signaling.
12. The method according to claim 1 or 2, characterized in that: The input information includes at least one of the following information: at least one pattern information; The order of pattern transmission; The number of resources required for configuration; Alternative subset of resources.
13. The method according to claim 12, characterized in that The input information further includes an input type indication, and the input type indication includes at least one of the following: Fixed pattern transmission; Transmit the patterns in the order in which they are transmitted; Random transmission pattern; Randomly transfer resources according to the required number of configured resources; Transfer all resources in the candidate subset of resources; The resources in the resource candidate subset are randomly transmitted according to the required number of configured resources.
14. A beam management method, the method being applied to a base station, characterized in that: The method comprises: receiving a trigger request and / or input information of an AI model sent by a UE, wherein the trigger request indicates that the UE will perform an AI operation; Send AI model configuration instructions to UE.
15. The method according to claim 14, characterized in that The trigger request includes an AI model operation type, and the operation type includes: model training, model inference, and model monitoring.
16. The method according to claim 14 or 15, characterized in that The AI model configuration includes the output configuration of the AI model.
17. The method according to claim 16, characterized in that The output configuration of the AI model includes: at least one output type or configuration identifier, where the identifier is used to indicate the output type or configuration.
18. The method according to claim 16, characterized in that The output configuration of the AI model also includes: adjusting the output type or configuration of a predefined AI model.
19. The method according to claim 14 or 16, characterized in that: The output configuration of the AI model includes at least one of the following: Identification of the output type or configuration; The reference signal resource index to be adjusted; The adjusted reference signal resource index; The identifier is used to indicate the output type or configuration.
20. The method according to claim 16, characterized in that The output configuration of the AI model also includes: reference signal resource configuration supported by the base station.
21. The method according to claim 14 or 15, characterized in that The method further comprises: An AI model output configuration operation type indication is sent to the UE, where the indication is used to indicate the executed AI model output configuration.
22. The method according to claim 14 or 15, characterized in that The AI model configuration includes: the input and output association relationship of the AI model.
23. The method according to claim 22, characterized in that The input-output association relationship of the AI model includes: If the input beam is a subset of the output beam, the input-output association relationship of the AI model is indicated by the reference signal resource index during configuration, that is, reference signal resources with the same index indicate a corresponding relationship; If the input beam is not a subset of the output beam, the input-output association relationship of the AI model is indicated by a configuration method in which one resource is associated with at least one other resource, or by a configuration method in which one resource is associated with at least one resource group, and the resource group contains at least one resource.
24. The method according to claim 23, characterized in that If the input beam is not a subset of the output beam, the input and output association relationship of the AI model is configured through DCI, MAC CE or RRC signaling.
25. The method according to claim 14 or 15, characterized in that The input information includes at least one of the following information: at least one pattern information; The order of pattern transmission; The number of resources required for configuration; Alternative subset of resources.
26. The method according to claim 25, characterized in that The input information further includes an input type indication, and the input type indication includes at least one of the following: Fixed pattern transmission; Transmit the patterns in the order in which they are transmitted; Random transmission pattern; Randomly transfer resources according to the required number of configured resources; Transfer all resources in the resource candidate subset; The resources in the resource candidate subset are randomly transmitted according to the required number of configured resources.
27. A beam management device, the device being located in a user equipment, characterized in that: The device comprises: A first sending unit, configured to send a trigger request and / or input information of an AI model to a base station, wherein the trigger request indicates that the UE will perform an AI operation; The first receiving unit is used to receive the AI model configuration indicated by the base station.
28. A beam management device, the device being located in a base station, characterized in that: The device comprises: A third receiving unit, configured to receive a trigger request sent by the UE and / or input information of the AI model, wherein the trigger request indicates that the UE will perform an AI operation; The second sending unit is used to send an AI model configuration indication to the UE.
29. A user equipment, characterized in that: The user equipment comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 13.
30. A base station, characterized in that: The base station comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 14 to 26.
31. A chip, characterized in that: The chip comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 13 or the method of any one of claims 14 to 26.
32. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which, when executed by a processor, implement the method according to any one of claims 1 to 13 or the method according to any one of claims 14 to 26.
