A method and apparatus used in a node for wireless communication and artificial intelligence

Abstract

The application discloses a method and device used in a node for wireless communication and artificial intelligence. A first node receives a first configuration information block, the first configuration information block indicating a first inference configuration; receives a first RS resource set, the first RS resource set including one or more RS resources; performs at least one performance monitoring for the first inference configuration every first time length; the performance monitoring depends on measurement on the first RS resource set and the first inference configuration; and the first time length depends on the output of the performance monitoring and the first inference configuration. The application proposes criteria and conditions for relaxing performance monitoring in combination with AI characteristics, which is conducive to identifying model changes while saving energy at the terminal and ensuring the reliability and robustness of inference.

Description

Technical Field

[0001] This application relates to signal transmission methods and apparatus in wireless communication systems, and more particularly to methods and apparatus for monitoring inference performance in AI / ML (Artificial Intelligence / Machine Learning). Background Technology

[0002] Leveraging AI / ML (Artificial Intelligence / Machine Learning) technologies to enhance 5G network performance is a crucial component of achieving deep integration of 5G and AI / ML and building intelligent dimensions for 5G-Advanced (5.5G) networks. The 3GPP (3rd Generation Partnership Project) standards organization initiated research on standards for RAN (Radio Access Networks) intelligence starting with Rel-16 (Release-16), primarily focusing on intelligent use cases, enhanced data collection, and the potential impact on RAN nodes and interfaces. Rel-18 formally established a project for AI / ML-based 5G air interface enhancement, initiating international standardization work on the integration of 5G air interface and AI / ML, mainly focusing on researching use cases, lifecycle management (LCM), simulation verification, and data collection.

[0003] Currently, the development of AI / ML has entered the large-scale model stage. Large-scale communication models can realize autonomous networks and intelligent services, support network operation optimization, and improve network efficiency. The deep integration of communication and AI is an important direction for the future evolution of communication. AI will empower the development and upgrading of 5G, 5.5G to 6G, bringing new management models such as automated management of frequency bands and traffic, real-time analysis of user data and network load, and prediction of network status. Summary of the Invention

[0004] LCM (Limited Management) of AI / ML models refers to the management of the entire process from creation to termination of an AI / ML model. 3GPP, in Rel-18, proposed that LCM for AI / ML models includes one or more of the following: data collection, model training, model registration, model deployment, model configuration, model inference operation, model selection / switching / activation / deactivation / fallback operation, model monitoring, model update, and model transfer. Among these, how the terminal can conserve energy while identifying changes in model performance during AI / ML model monitoring is a pressing technical problem that needs to be solved.

[0005] To address the aforementioned issues, this application discloses a solution. It should be noted that while this application is initially intended for AI / ML scenarios, it can also be applied to other non-AI / ML scenarios. Furthermore, adopting a unified design scheme for different scenarios (such as other non-AI / ML scenarios, including but not limited to Vehicle to Everything (V2X), capacity enhancement systems, short-range communication systems, NTN (Non-Terrestrial Network), IoT (Internet of Things), and URLLC (Ultra-Reliable Low-Latency Communication) networks) helps reduce hardware complexity and cost. Where there is no conflict, embodiments and features in any node of this application can be applied to any other node. Where there is no conflict, embodiments and features in any embodiment of this application can be arbitrarily combined with each other.

[0006] In particular, the interpretation of terms, nouns, functions, and variables in this application (unless otherwise specified) can be found in the definitions of the TS38 and TS37 series of 3GPP (3rd Generation Partnership Project) Technical Specifications (TS). Where necessary, reference can be made to TS38.211, TS38.212, TS38.213, TS38.214, TS38.215, TS38.300, TS38.304, TS38.305, TS38.321, TS38.331, TS37.355, and TS38.423 in the 3GPP technical specifications to aid in understanding this application.

[0007] As an example, the interpretation of terms in this application is based on the definitions in the 3GPP specification protocol TS38 series.

[0008] As an example, the interpretation of terms in this application is based on the definitions in the 3GPP specification protocol TS37 series.

[0009] As an example, the interpretation of the terms used in this application is based on the definitions in 3GPP specification protocol Rel-17.

[0010] As an example, the interpretation of the terms used in this application is based on the definitions in 3GPP specification protocol Rel-18.

[0011] This application discloses a method for a first node in wireless communication and artificial intelligence, comprising:

[0012] Receive a first configuration information block, the first configuration information block indicating a first inference configuration;

[0013] Receive a first RS resource set, wherein the first RS resource set includes one or more RS resources;

[0014] Perform at least one performance check for the first inference configuration every first time period;

[0015] The performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the results of the performance monitoring and the first inference configuration.

[0016] As an example, the problem this application aims to solve includes: how to save energy while recognizing changes in model performance.

[0017] As an example, the problem this application aims to solve includes: how the first node determines the monitoring cycle of the model.

[0018] As an example, the features of the above method include: in this application, the first node performs performance monitoring at least once for a given inference configuration every first time period, so that the first node monitors the model inference performance and identifies changes in model performance.

[0019] As an example, the features of the above method include: in this application, the first time length depends on the corresponding inference configuration and the results of performance monitoring, so that the first node can relax the monitoring of model performance when the model performance is good, thereby achieving the purpose of saving energy; furthermore, performance monitoring for each inference configuration can ensure that the first node monitors the inference performance of the inference process associated with each inference configuration.

[0020] As an example, the features of the above method include: in this application, the first node determines the inference performance based on the inference configuration and the performance monitoring results corresponding to the inference configuration, and achieves terminal energy saving by extending the performance monitoring cycle when the inference performance of the model or functionality is good.

[0021] As an example, the advantages of the above method include: this application supports the deep integration of AI and communication, improves the adaptability and intelligence level of the communication system, and thus enhances the performance, efficiency and user experience of the communication system.

[0022] As an example, the advantages of the above method include: achieving terminal energy saving in a manner that has a minimal impact on inference performance monitoring.

[0023] As an example, the advantages of the above method include good compatibility and feasibility.

[0024] According to one aspect of this application, the above method is characterized by comprising:

[0025] Receive the first information block, which configures the inference performance good criterion;

[0026] The first time length depends on whether the reasoning performance criterion is met.

[0027] As an example, the characteristics of the above method include: the good inference performance is specific to the inference configuration.

[0028] As an example, the features of the above method include: the first time length depends on whether the model or functionality associated with the first inference configuration meets the good inference performance criterion.

[0029] As an example, the features of the above method include: if the model or functionality associated with the first inference configuration meets the good inference performance criterion, the first node performs relaxed performance monitoring for the first inference configuration; otherwise, the first node does not perform relaxed performance monitoring for the first inference configuration. The relaxed performance monitoring includes using a longer time interval for performance monitoring, i.e., relaxing performance monitoring by extending the first time length.

[0030] As an example, the benefits of the above method include: adjusting the monitoring cycle according to inference performance helps the UE save energy.

[0031] As an example, the advantages of the above method include: defining criteria for good inference performance, which facilitates implementation and deployment.

[0032] As an example, the benefits of the above method include: improving the robustness of inference and increasing the credibility of the model.

[0033] According to one aspect of this application, the above method is characterized in that the result of the performance monitoring includes a first metric, and the good inference performance criterion includes one of the following:

[0034] - The first metric satisfies a first threshold, where the first threshold is predefined or configured by a higher-layer signaling layer;

[0035] - Within a second time length, the first metric satisfies a first threshold, the second time length is not less than the first time length, and the first threshold is predefined or configured by a higher-layer signaling;

[0036] - At least K consecutive times the first metric satisfies the first threshold, where K is a positive integer greater than 1, and the first threshold is predefined or configured by a higher-layer signaling.

[0037] As an example, the characteristics of the above method include: the good inference performance criterion depends on the performance monitoring results performed for the first inference configuration.

[0038] As an example, the characteristics of the above method include: the performance monitoring is performed on the UE side.

[0039] As an example, the features of the above method include: the first threshold is configured at least per inference.

[0040] As an example, the advantages of the above method include: this application provides multiple implementation methods, which are helpful for commercial use and deployment.

[0041] As an example, the benefits of the above method include: enhanced robustness, support for trend judgment and detection of model degradation.

[0042] As an example, the advantages of the above method include: good quantification inference performance criteria and simple implementation.

[0043] According to one aspect of this application, the above method is characterized by comprising:

[0044] Receive a second information block, which configures low mobility criteria and at least the former of criteria not at the cell edge;

[0045] Wherein, the first inference is configured for low mobility, and the first time length depends on whether at least the former of the low mobility criterion and the non-cell edge criterion is met.

[0046] As an example, the features of the above method include: the low mobility criterion and the not-at-cell-edge criterion are configured per UE.

[0047] As an example, the characteristics of the above method include: the inference process does not output irregular extreme values; when the UE is in a low mobility state, the channel changes relatively slowly when the terminal is stationary or moving at low speed; low variance inference input helps to improve inference performance, thereby reducing the model monitoring frequency or extending the model monitoring cycle.

[0048] As an example, the features of the above method include: when the first node is not at the cell edge, it usually indicates higher signal strength and quality; low-noise inference input helps to improve inference performance, which in turn can reduce the model monitoring frequency or extend the model monitoring cycle.

[0049] As an example, the advantages of the above method include: good compatibility.

[0050] As an example, the benefits of the above method include: reducing inference performance monitoring requirements, thereby reducing UE power consumption.

[0051] As an example, the advantages of the above method include: avoiding redundant calculations while maintaining performance stability.

[0052] According to one aspect of this application, the above method is characterized by comprising:

[0053] Send a first message, the first message including the results of the performance monitoring;

[0054] Monitor target signaling, wherein the target signaling is MAC layer signaling or physical layer signaling;

[0055] The target signaling indicates the first time length from multiple time lengths.

[0056] As an example, the features of the above method include: the performance detection includes UE-assisted performance monitoring.

[0057] As an example, the characteristics of the above method include: the results of the performance monitoring include intermediate-level KPIs for model monitoring.

[0058] As an example, the characteristics of the above method include: the results of the performance monitoring include whether the model or functionality associated with the first inference configuration meets the good inference performance criteria.

[0059] As an example, the advantages of the above method include: this application is applicable to situations where the base station and the terminal jointly perform model performance monitoring or the base station and the terminal adopt a bilateral model, and provides an implementation scheme for the base station and the terminal to jointly perform model performance monitoring or the base station and the terminal to perform relaxed performance monitoring using a bilateral model, which helps to further reduce UE power consumption.

[0060] As an example, the advantages of the above method include: UE-assisted performance detection can reduce the amount of UE computation and reduce the computational pressure on the UE.

[0061] As an example, the advantages of the above method include: dynamically changing the period used for inference performance monitoring based on the signaling of the protocol layer below the RRC layer.

[0062] According to one aspect of this application, the above method is characterized in that the model or functionality associated with the first inference configuration is for a higher level.

[0063] As an example, the features of the above method include: the model or functionality associated with the first inference configuration is for the MAC layer or the RRC layer.

[0064] As an example, the characteristics of the above method include: the results of the performance monitoring require filtering by a higher-level filter.

[0065] As an example, the characteristics of the above method include: the result of the performance monitoring requires at least one timer or counter, which is maintained at a higher level.

[0066] As an example, the advantages of the above method include: reducing the impact on the physical layer while ensuring energy saving in the terminal.

[0067] As an example, the advantages of the above method include: higher-level models have lower requirements for real-time performance and accuracy, but higher deployment costs. Therefore, it is more reasonable to reduce the monitoring frequency for higher-level models or functions, resulting in greater system benefits.

[0068] As an example, the benefits of the above method include enhanced fault tolerance and robustness.

[0069] According to one aspect of this application, the above method is characterized in that the first time length depends on a first parameter, the first parameter indicating the downlink reception period.

[0070] As an example, the features of the above method include: the downlink reception period includes STMC.

[0071] As an example, the features of the above method include: the downlink reception period includes DRX.

[0072] As an example, the advantages of the above method include better compatibility.

[0073] As an example, the advantages of the above method include: significantly reducing terminal power consumption and avoiding scheduling conflicts.

[0074] As an example, the benefits of the above method include: improved monitoring effectiveness.

[0075] According to one aspect of this application, the above method is characterized in that the first node is a user equipment.

[0076] According to one aspect of this application, the above method is characterized in that the first node is a terminal.

[0077] This application discloses a method for a second node in wireless communication and artificial intelligence, comprising:

[0078] Send a first configuration information block, which indicates a first inference configuration;

[0079] Send a first RS resource set, which includes one or more RS resources;

[0080] Wherein, the receiver of the first configuration information block performs at least one performance monitoring for the first inference configuration every first time length, the performance monitoring depending on the measurement on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0081] As an example, the features of the above method include: the second node includes a base station and a core network.

[0082] As an example, the features of the above method include: the second node includes a core network.

[0083] As an example, the features of the above method include: the second node includes an entity for deploying AI / ML models.

[0084] As an example, the features of the above method include: the second node includes a node for deploying AI / ML models.

[0085] As an example, the features of the above method include: the second node includes a base station.

[0086] As an example, the features of the above method include: the second node is a base station.

[0087] As an example, the features of the above method include: the second node is an LMF.

[0088] As an example, the features of the above method include: the second node is OAM.

[0089] As an example, the features of the above method include: the second node is a gNB.

[0090] As an example, the features of the above method include: the second node is a network device, which includes at least one of a core network device and an access network device.

[0091] As an example, the features of the above method include: the second node is a device that provides wireless communication function services, can communicate with terminal devices, and is usually located on the network side.

[0092] As an example, the features of the above method include: the base station in this application includes a core network.

[0093] As an example, the features of the above method include: the base station in this application includes core network equipment.

[0094] As an example, the features of the above method include: the base station in this application includes an entity for deploying AI / ML models.

[0095] As an example, the features of the above method include: the base station in this application includes nodes for deploying AI / ML models.

[0096] According to one aspect of this application, the above method is characterized by comprising:

[0097] Send the first information block, which configures the inference performance good criterion;

[0098] The first time length depends on whether the reasoning performance criterion is met.

[0099] According to one aspect of this application, the above method is characterized in that the result of the performance monitoring includes a first metric, and the good inference performance criterion includes one of the following:

[0100] - The first metric satisfies a first threshold, where the first threshold is predefined or configured by a higher-layer signaling layer;

[0101] - Within a second time length, the first metric satisfies a first threshold, the second time length is not less than the first time length, and the first threshold is predefined or configured by a higher-layer signaling;

[0102] - At least K consecutive times the first metric satisfies the first threshold, where K is a positive integer greater than 1, and the first threshold is predefined or configured by a higher-layer signaling.

[0103] According to one aspect of this application, the above method is characterized by comprising:

[0104] Send a second information block, which configures the low mobility criterion and at least the former of the criteria not at the cell edge;

[0105] Wherein, the first inference is configured for low mobility, and the first time length depends on whether at least the former of the low mobility criterion and the non-cell edge criterion is met.

[0106] According to one aspect of this application, the above method is characterized by comprising:

[0107] Receive a first message, the first message including the results of the performance monitoring;

[0108] Send target signaling, which is MAC layer signaling or physical layer signaling;

[0109] The target signaling indicates the first time length from multiple time lengths.

[0110] According to one aspect of this application, the above method is characterized in that the model or functionality associated with the first inference configuration is for a higher level.

[0111] According to one aspect of this application, the above method is characterized in that the first time length depends on a first parameter, the first parameter indicating the downlink reception period.

[0112] According to one aspect of this application, the method described above is characterized in that the second node is a base station.

[0113] This application discloses a first node for use in wireless communication and artificial intelligence, comprising:

[0114] A first processor receives a first configuration information block, the first configuration information block indicating a first inference configuration; receives a first RS resource set, the first RS resource set including one or more RS resources; and performs at least one performance monitoring for the first inference configuration every first time period.

[0115] The performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0116] This application discloses a second node for use in wireless communication and artificial intelligence, comprising:

[0117] The second transmitter sends a first configuration information block, which indicates a first inference configuration; and sends a first RS resource set, which includes one or more RS resources.

[0118] Wherein, the receiver of the first configuration information block performs at least one performance monitoring for the first inference configuration every first time length, the performance monitoring depending on the measurement on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0119] As an example, compared with conventional solutions, this application has the following advantages, but is not limited to:

[0120] This application supports the deep integration of AI and communication to improve the adaptability and intelligence of communication systems, thereby enhancing the performance, efficiency, and user experience of communication systems.

[0121] This application provides various implementation methods for relaxing performance monitoring, which is helpful for the commercialization and deployment of AI / ML;

[0122] By combining AI characteristics, we propose criteria and conditions for relaxing performance monitoring, which is beneficial for identifying model changes while saving energy at the terminal, thus ensuring the reliability and robustness of inference. Attached Figure Description

[0123] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0124] Figure 1 A flowchart of the first node transmission according to an embodiment of this application is shown;

[0125] Figure 2 A schematic diagram of a network architecture according to an embodiment of this application is shown;

[0126] Figure 3 A schematic diagram of an embodiment of a wireless protocol architecture for the user plane and control plane according to an embodiment of this application is shown;

[0127] Figure 4 A schematic diagram of a first communication device and a second communication device according to an embodiment of this application is shown;

[0128] Figure 5 A first flowchart illustrating the transmission between a first node and a second node according to an embodiment of this application is shown;

[0129] Figure 6 A second flowchart illustrating the transmission between a first node and a second node according to an embodiment of this application is shown;

[0130] Figure 7 A third flowchart illustrating the transmission between a first node and a second node according to an embodiment of this application is shown;

[0131] Figure 8 A fourth flowchart illustrating the transmission between a first node and a second node according to an embodiment of this application is shown;

[0132] Figure 9 A schematic diagram of a first time length according to an embodiment of this application is shown;

[0133] Figure 10 A schematic diagram of a first inference configuration according to an embodiment of this application is shown;

[0134] Figure 11 A schematic diagram of a first parameter according to an embodiment of this application is shown;

[0135] Figure 12 A schematic diagram illustrating the deployment of RAN domain AI / ML functionality according to an embodiment of this application is shown;

[0136] Figure 13 A schematic diagram illustrating the deployment of AI / ML functions in a UE according to an embodiment of this application is shown;

[0137] Figure 14 A schematic diagram of an artificial intelligence or machine learning-based processing system according to an embodiment of this application is shown;

[0138] Figure 15 A schematic diagram illustrating artificial intelligence or machine learning according to an embodiment of this application is shown;

[0139] Figure 16 A structural block diagram of a processing apparatus for a first node according to an embodiment of this application is shown;

[0140] Figure 17 A structural block diagram of a processing apparatus for a second node according to an embodiment of this application is shown. Detailed Implementation

[0141] The technical solutions of this application will be further described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Considering performance, flexibility, complexity, overhead, and compatibility, those skilled in the art are motivated to flexibly combine the embodiments in different drawings without conflict, including but not limited to the accompanying drawings. Figure 1 Examples and appendices Figure 5 -Appendix Figure 17 The embodiments in the appendix Figure 5 Examples and appendices Figure 6 -Appendix Figure 17 Examples, etc.

[0142] Example 1

[0143] Example 1 illustrates a flowchart of a first node transmission according to an embodiment of this application, as shown in the attached diagram. Figure 1 As shown. In the appendix Figure 1 In this diagram, each box represents a step. Specifically, the order of the steps within the boxes does not indicate a specific temporal sequence between them.

[0144] The first node receives a first configuration information block in step 101, the first configuration information block indicating a first inference configuration; receives a first RS resource set in step 102, the first RS resource set including one or more RS resources; and performs at least one performance monitoring for the first inference configuration every first time length in step 103.

[0145] In Example 1, the performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0146] As one example, the first node is a user equipment (UE).

[0147] As one example, the first node is a terminal.

[0148] As an example, the first node in this application includes a core network device that provides services for terminal AI.

[0149] As an example, the first node in this application includes an application layer device that provides services for terminal AI.

[0150] As an example, the first node in this application includes an Agent (intelligent agent) that provides services for the terminal AI.

[0151] As an example, the first node in this application includes a Handset (phone).

[0152] As an example, the first node is the first node in this application.

[0153] As an example, RS stands for Reference Signal.

[0154] As an example, the first node receives the first configuration information block.

[0155] As one embodiment, the first configuration information block is carried by higher layer signaling.

[0156] As an example, the first configuration information block is transmitted via RRC (Radio Resource Control) signaling.

[0157] As an example, the first configuration information block includes one or more RRC IEs (Information Elements).

[0158] As an example, the first configuration information block includes one or more fields in an RRC IE.

[0159] As one embodiment, the first configuration information block includes information from one or more domains of each of the plurality of RRC IEs.

[0160] As an example, the first configuration information block is an RRC IE.

[0161] As one example, the first configuration information block includes ServingCellConfigCommonSIB IE.

[0162] As one example, the first configuration information block includes one or more domains in the ServingCellConfigCommonSIB IE.

[0163] As an example, the first configuration information block includes ServingCellConfigCommon IE.

[0164] As an example, the first configuration information block includes one or more domains in the ServingCellConfigCommon IE.

[0165] As an example, the first configuration information block includes ServingCellConfig IE.

[0166] As one example, the first configuration information block includes one or more domains in ServingCellConfig IE.

[0167] As an example, the first configuration information block includes CSI-MeasConfig IE.

[0168] As an example, the first configuration information block includes one or more domains in the CSI-MeasConfig IE.

[0169] As an example, the first configuration information block includes MeasConfig IE.

[0170] As an example, the first configuration information block includes one or more domains in the MeasConfig IE.

[0171] As an example, the first configuration information block includes MeasObjectToAddModListIE.

[0172] As one example, the first configuration information block includes one or more fields in MeasObjectToAddModListIE.

[0173] As one example, the first configuration information block includes one or more domains in MeasObjectIE.

[0174] As an example, the first configuration information block includes MeasObjectNR IE.

[0175] As an example, the first configuration information block includes MeasObjec6G IE.

[0176] As an example, the first configuration information block includes MeasObjec6G IE.

[0177] As one embodiment, the first configuration information block includes CSI-RS-ResourceConfigMobilityIE.

[0178] As one example, the first configuration information block includes one or more domains in the CSI-RS-ResourceConfigMobility IE.

[0179] As an example, the name of the signaling used to transmit the first configuration information block includes CSI.

[0180] As an example, the name of the signaling used to transmit the first configuration information block includes CSI-RS.

[0181] As an example, the name of the signaling used to transmit the first configuration information block includes Config.

[0182] As an example, the name of the signaling used to transmit the first configuration information block includes Meas.

[0183] As an example, the name of the signaling used to transmit the first configuration information block includes Measurement.

[0184] As one example, the name of the signaling used to transmit the first configuration information block includes Object.

[0185] As an example, the name of the signaling used to transmit the first configuration information block includes Inference.

[0186] As an example, the name of the signaling used to transmit the first configuration information block includes AI.

[0187] As an example, the name of the signaling used to transmit the first configuration information block includes ML.

[0188] As an example, the first configuration information block indicates the first inference configuration.

[0189] As an example, the first configuration information block configures the first inference configuration.

[0190] As an example, the first configuration information block carries the first inference configuration.

[0191] As one embodiment, the first configuration information block includes the first inference configuration.

[0192] As one embodiment, the first inference configuration indicates some or all of the inference parameters.

[0193] As a sub-implementation of this embodiment, some or all of the parameters of the inference indicate at least one of the input, output, and model of the inference process associated with the first inference configuration.

[0194] As a sub-example of this embodiment, some or all of the parameters of the inference indicate a set of RS resources for generating the dataset of the inference process associated with the first inference configuration.

[0195] As a sub-example of this embodiment, some or all of the parameters of the inference include the beamwidth of the RS resource used to generate the dataset of the inference process associated with the first inference configuration.

[0196] As a sub-example of this embodiment, some or all of the parameters of the inference include the number of RS resources used to generate the dataset of the inference process associated with the first inference configuration.

[0197] As a sub-implementation of this embodiment, some or all of the parameters of the inference include the model parameters associated with the first inference configuration.

[0198] As a sub-example of this embodiment, some or all of the parameters of the inference include parameters used to construct the model associated with the first inference.

[0199] As a sub-implementation of this embodiment, some or all of the parameters of the inference include some or all of the parameters used to construct the model associated with the first inference.

[0200] As an example, the first inference configuration indicates an inference process.

[0201] As an example, the first inference configuration includes configuration information for a first operation, which includes at least the former of inference and monitoring.

[0202] As an example, the first inference configuration instructs the first node to perform higher-level inference parameters.

[0203] As an example, the first inference configuration instructs the first node to perform inference and monitoring of higher-level parameters.

[0204] As an example, the first inference configuration corresponds to a model.

[0205] As an example, the first inference configuration corresponds to multiple models.

[0206] As an example, the first inference configuration corresponds to a functionality.

[0207] As an example, the first inference configuration corresponds to an applicable functionality.

[0208] As an example, the first inference configuration is associated with one or more models.

[0209] As an example, the first inference configuration is associated with a functionality.

[0210] As an example, the first inference configuration is associated with an applicable functionality.

[0211] As an example, the first inference configuration deploys one or more models.

[0212] As an example, the first inference configuration deploys a functionality.

[0213] As an example, the first inference configuration deploys an applicable functionality.

[0214] As an example, the first inference configuration corresponds to an inference-Id.

[0215] As an example, the inference refers to infer or inference.

[0216] As an example, the reasoning refers to prediction or prediction.

[0217] As an example, the model described in this application refers to an AI model or an ML model.

[0218] As an example, the model described in this application is AIAgent.

[0219] As an example, the first node receives the first RS resource set.

[0220] As one embodiment, the first RS resource set includes one or more RS resources.

[0221] As an example, the first RS resource set includes an RS resource.

[0222] As one embodiment, the first RS resource set includes multiple RS resources.

[0223] As an example, the first RS resource set is periodic.

[0224] As one embodiment, the first RS resource set includes periodic RS resources.

[0225] As an example, the first RS resource set is semi-persistent (SP).

[0226] As an example, the first RS resource set includes SPRS resources.

[0227] As an example, the RS resources in the first RS resource set belong to the same cell.

[0228] As an example, in the first RS resource set, there are two RS resources belonging to different cells.

[0229] As an example, the RS resources in the first RS resource set belong to the same BWP (BandWidthPart).

[0230] As an example, in the first RS resource set, there are two RS resources belonging to different BWPs.

[0231] As an example, each RS resource in the first RS resource set includes at least one port(s).

[0232] As an example, each RS resource in the first RS resource set includes at least one antenna port(s).

[0233] As an example, each RS resource in the first RS resource set includes at least one RS port.

[0234] As an example, each RS resource in the first RS resource set includes a CSI-RS (Channel State Information Reference Signal) port.

[0235] As an example, the first RS resource set includes at least one RS.

[0236] As an example, the first RS resource set includes at least one RS resource.

[0237] As one embodiment, the first RS resource set includes downlink RS resources.

[0238] As one embodiment, the first RS resource set includes RS resources for time-frequency resource tracking.

[0239] As an example, the first RS resource set includes PTRS (Phase-Tracking Reference Signal) resources.

[0240] As one embodiment, the first RS resource set includes RS resources for positioning.

[0241] As an example, the first RS resource set includes PRS (Positioning Reference Signal) resources.

[0242] As one embodiment, the first RS resource set includes RS resources for channel estimation.

[0243] As one embodiment, the first RS resource set includes RS resources for demodulation.

[0244] As an example, the first RS resource set includes DMRS (DeModulation Reference Signal) resources.

[0245] As an example, the first RS resource set includes RS resources for sensing.

[0246] As an example, the first RS resource set includes RS resources for ISAC (Integrated Sensing and Communication).

[0247] As an example, the first RS resource set includes ISAC-RS resources.

[0248] As an example, the first RS resource set includes IRS (ISAC Reference Signal) resources.

[0249] As one embodiment, the first RS resource set includes RS resources for mobility management.

[0250] As one embodiment, the first RS resource set includes RS resources for cell-level mobility management.

[0251] As one embodiment, the first RS resource set includes RS resources for beam-level mobility management.

[0252] As one embodiment, the first RS resource set includes RS resources for synchronization.

[0253] As one embodiment, the first RS resource set includes synchronization signals in at least 5G systems and systems after 5G systems.

[0254] As an example, the first RS resource set includes at least a synchronization signal in a 6G system.

[0255] As one embodiment, the first RS resource set includes at least a synchronization signal (SS).

[0256] As one embodiment, the first RS resource set includes at least a Primary Synchronization Signal (PSS).

[0257] As one embodiment, the first RS resource set includes at least a Secondary Synchronization Signal (SSS).

[0258] As an example, the first RS resource set includes at least PBCH (Physical Broadcast Channel).

[0259] Typically, the PBCH, PSS, and SSS are received in consecutive symbols and form an SS / PBCH block.

[0260] As an example, any RS resource in the first RS resource set is a CSI-RS resource or an SSB resource.

[0261] As an example, any RS resource in the first RS resource set is identified by an NZP-CSI-RS-ResourceId or SSB-Index.

[0262] As an example, the first RS resource set is identified by a CSI-ResourceConfigId.

[0263] As an example, the first RS resource set is a CSI-RS resource set.

[0264] As an example, the first RS resource set is an NZP (Non-Zero-Power) CSI-RS resource set.

[0265] As an example, the first RS resource set is a CSI-SSB resource set.

[0266] As an example, each RS resource in the first RS resource set is a CSI-RS resource.

[0267] As an example, each RS resource in the first RS resource set is an NZP CSI-RS resource.

[0268] As an example, each RS resource in the first RS resource set is identified by an NZP-CSI-RS-ResourceId.

[0269] As an example, the first RS resource set includes SSB.

[0270] As an example, the first RS resource set includes SSB resources.

[0271] As an example, the first RS resource set includes at least one SSB.

[0272] As an example, the first RS resource set includes an SSB.

[0273] As one embodiment, the first RS resource set includes multiple SSBs.

[0274] As an example, the first RS resource set includes at least one SSB in an SSB burst set.

[0275] As an example, each RS resource in the first RS resource set is identified by an SSB-Index.

[0276] As an example, each RS resource in the first RS resource set is an SSB resource.

[0277] As an example, SSB in this application refers to Synchronization Signal Block.

[0278] As an example, the SSB mentioned in this application refers to the SS / PBCH block.

[0279] As an example, the first node performs the performance monitoring at least once for the first inference configuration every first time period.

[0280] As an example, the first time length is one of a plurality of time lengths.

[0281] As an example, the first time length is one of Q1 candidate time lengths, where Q1 is a positive integer greater than 2.

[0282] As an example, the unit of the first time length is seconds (s).

[0283] As an example, the unit of the first time length is milliseconds (ms).

[0284] As an example, the unit of the first time length is a subframe.

[0285] As an example, the unit of the first time length is a slot.

[0286] As an example, the unit of the first time length is a symbol.

[0287] As an example, the first node performs the performance monitoring at least once for the first inference configuration every first time period.

[0288] As an example, performance monitoring for an inference configuration in this application includes: performance monitoring for the inference process corresponding to the inference configuration.

[0289] As an example, performance monitoring for an inference configuration in this application includes performance monitoring for a model associated with the inference configuration.

[0290] As an example, performance monitoring for an inference configuration in this application includes performance monitoring for multiple models associated with the inference configuration.

[0291] As an example, performance monitoring for an inference configuration in this application includes performance monitoring for some or all of the multiple models associated with the inference configuration.

[0292] As an example, performance monitoring for an inference configuration in this application includes performance monitoring for a function associated with the inference configuration.

[0293] As an example, performance monitoring for an inference configuration in this application includes performance monitoring for an applicable functionality associated with the inference configuration.

[0294] As one example, the performance monitoring includes: inference performance monitoring.

[0295] As one example, the performance monitoring includes: predictive performance monitoring.

[0296] As an example, the performance monitoring includes inference latency monitoring.

[0297] As one example, the performance monitoring includes: resource consumption monitoring.

[0298] As one example, the performance monitoring includes: data distribution detection.

[0299] As an example, the performance monitoring is only performed on the first inference configuration.

[0300] As an example, the performance monitoring targets the first inference configuration and at least one inference configuration other than the first inference configuration.

[0301] As an example, the performance monitoring targets multiple inference configurations, with the first inference configuration being one of the multiple inference configurations.

[0302] As a sub-implementation of this embodiment, the first configuration information block indicates the plurality of inference configurations.

[0303] As a sub-implementation of this embodiment, the multiple inference configurations are associated with the same functionality.

[0304] As a sub-implementation of this embodiment, the RS resources corresponding to the plurality of inference configurations are QCLs with the RS resources in the first RS resource set.

[0305] As an example, QCL in this application refers to Quasi Co-Location.

[0306] As an example, QCL in this application refers to Quasi Co-Located.

[0307] As an example, the QCL types described in this application include typeA, typeB, typeC, and typeD.

[0308] As an example, the QCL parameters of type A in this application include Doppler shift, Doppler spread, average delay, and delay spread; the QCL parameters of type B include Doppler shift and Doppler spread; the QCL parameters of type C include Doppler shift and average delay; and the QCL parameters of type D include spatial Rx parameters.

[0309] As an example, the specific definitions of type A, type B, type C and type D in this application can be found in section 5.1.5 of 3GPP TS (Technical Specification) 38.214.

[0310] As one example, the performance monitoring relies on measurements on the first RS resource set and the first inference configuration.

[0311] As one embodiment, the measurement on the first RS resource set includes: measuring RS transmitted on the RS resources included in the first RS resource set.

[0312] As one embodiment, the measurement on the first RS resource set includes: measuring the RS transmitted on the RS resources included in the first RS resource set.

[0313] As one embodiment, the measurement on the first RS resource set includes: measuring the RS transmitted on a portion of the RS resources included in the first RS resource set.

[0314] As one example, the measurement includes: channel measurement.

[0315] As one example, the measurement includes: received power measurement.

[0316] As an example, the measurement includes: path loss (PL) measurement.

[0317] As an example, the measurement includes a measurement of RSRP (Reference Signal Received Power).

[0318] As an example, the measurement includes a measurement of RSRQ (Reference Signal Received Quality).

[0319] As an example, the measurement includes a measurement of RSSI (Received Signal Strength Indicator).

[0320] As an example, the measurement includes a measurement of SNR (Signal To Noise Ratio).

[0321] As one example, the measurement includes: intra-frequency measurement.

[0322] As one example, the measurement includes: inter-frequency measurement.

[0323] As one example, the measurement includes: inter-RAT (RadioAccess Technology) measurement.

[0324] As an example, the measurement includes: Intra-RAT measurement.

[0325] As one example, the measurement includes: channel measurement.

[0326] As one example, the measurement includes: received power measurement.

[0327] As one example, the measurement includes: channel matrix measurement.

[0328] As an example, the measurement includes: raw channel matrix measurement.

[0329] As one example, the measurement includes: eigenvector and eigenvalue measurements.

[0330] As an example, the measurement includes: Block Error Rate (BLER) measurement.

[0331] As an example, the measurement includes: Bit Error Rate (BER) measurement.

[0332] As one example, the measurement includes: delay spread measurement.

[0333] As one example, the measurement includes: Doppler displacement measurement.

[0334] As one example, the measurement includes: Doppler extension measurement.

[0335] As one example, the measurement includes: average delay measurement.

[0336] As one example, the measurement includes: average gain measurement.

[0337] As one example, the measurement includes: interference measurement.

[0338] As one example, the measurement includes: interference channel matrix measurement.

[0339] As one example, the measurement includes: a measurement of the interference covariance matrix.

[0340] As one example, the measurement includes: interference feature vector measurement.

[0341] As one example, the measurement includes: interference characteristic value measurement.

[0342] As one example, the measurement includes: interference beam measurement.

[0343] As one example, the measurement includes: interference power measurement.

[0344] As one example, the measurement includes: interference variance measurement.

[0345] As one example, the measurement includes: interference power spectral density measurement.

[0346] As an example, the first inference configuration indicates a second RS resource set, and the performance monitoring depends on measurements on the first RS resource set and the second RS resource set.

[0347] As an example, the inputs for the performance monitoring include measurements on the first RS resource set.

[0348] As one example, the input for performance monitoring includes some or all of the resources in the first RS resource set.

[0349] As an example, the output of the performance monitoring depends on measurements on the first RS resource set.

[0350] As an example, the input of the performance monitoring depends on a portion of the RS resources on the first RS resource set, and the output of the performance monitoring depends on measurements on other RS ​​resources in the first RS resource set.

[0351] As an example, the output of the performance monitoring includes the first metric in this application, which depends on the output of the first inference configuration and the measurement on the first RS resource set.

[0352] As an example, the output of the performance monitoring includes the first metric in this application, which the first node obtains by comparing the output of the model associated with the first inference configuration with the measurement on the first RS resource set.

[0353] As a sub-example of this embodiment, the input of the model associated with the first inference configuration is QCL with the RS resources in the first RS resource set.

[0354] As a sub-example of this embodiment, the input to the model associated with the first inference configuration is some or all of the RS resources in the first RS resource set.

[0355] As a sub-example of this embodiment, the input of the model associated with the first inference configuration depends on the transmission occasion of the RS resources in the first RS resource set.

[0356] As an example, the first node performs the performance detection accordingly.

[0357] As an example, the first time length depends on the output of the performance monitoring and the first inference configuration.

[0358] As an example, the first time length is one of a plurality of time lengths, the first inference configuration configures the plurality of time lengths, and the output of the performance monitoring indicates the first time length from the plurality of time lengths.

[0359] As a sub-example of this embodiment, the plurality of time lengths are respectively a first candidate time length and a second candidate time length. When the output of the performance monitoring is better than or no worse than a first threshold, the first time length is the first candidate time length. When the output of the performance monitoring is worse than or no worse than the first threshold, the first time length is the second candidate time length.

[0360] As an example, the first inference configuration is one of a plurality of inference configurations, the plurality of inference configurations corresponding to a plurality of time length groups, each of the plurality of time length groups including one or more time lengths, the first inference configuration indicating a first time length group from the plurality of time length groups, and the performance monitoring result indicating the first time length from the first time length group.

[0361] As a sub-implementation of this embodiment, the plurality of time length groups are predefined.

[0362] As a sub-implementation of this embodiment, the plurality of time length groups are configured.

[0363] As a sub-implementation of this embodiment, the plurality of inference configurations are higher-layer signaling configurations.

[0364] As an example, the first time length depends on a first time parameter and a second time parameter, the first inference configuration indicates the first time parameter, and the performance monitoring output indicates the second time parameter.

[0365] As a sub-implementation of this embodiment, the first time length is linearly related to the first time parameter.

[0366] As a sub-implementation of this embodiment, the first time length is linearly related to the second time parameter.

[0367] As a sub-implementation of this embodiment, the first time length also depends on time parameters other than the first time parameter and the second time parameter.

[0368] Example 2

[0369] Example 2 illustrates a schematic diagram of a network architecture according to an embodiment of this application, as shown in the attached diagram. Figure 2 As shown.

[0370] Appendix Figure 2Network architecture 200 is described. Network architecture 200 refers to the network architectures of LTE (Long-Term Evolution), LTE-A (Long-Term Evolution Advanced), 5G systems, 5G-Advanced, and future 6G systems. The network architectures of LTE, LTE-A, 5G systems, 5G-Advanced, and future 6G systems are referred to as EPS (Evolved Packet System). The 5G NR or LTE network architecture may be referred to as 5GS (5G System) / EPS or some other suitable terminology; the 6G network architecture may be referred to as 6GS (6G System) / EPS or some other suitable terminology.

[0371] The network architecture 200 may include one or more UEs 201, a RAN (Radio Access Network) 202, a core network 210, an HSS (Home Subscriber Server) / UDM (Unified Data Management) 220, and an Internet service 230. The network architecture 200 may interconnect with other access networks, but these entities / interfaces are not shown for simplicity.

[0372] As attached Figure 2As shown, the network architecture 200 provides packet switching services; however, those skilled in the art will readily understand that the various concepts presented throughout this application can be extended to networks providing circuit-switched services or other cellular networks. The RAN 202 includes Node B 203 and other nodes 204. Node B 203 provides user and control plane protocol termination toward the UE 201. Node B 203 may be connected to other nodes 204 via an Xn interface (e.g., backhaul). Node B 203 may also be referred to as eNB (evolved Node B), gNB, base station, base transceiver station, wireless base station, wireless transceiver, transceiver function, Basic Service Set (BSS), Extended Service Set (ESS), TRP (Transmitter Receiver Point), or some other suitable term. Node B 203 provides UE 201 with an access point to the core network 210; the core network 210 is a 5GC (5G Core network) / EPC (Evolved Packet Core), or the core network 210 is a 6GC (6G Core network). Examples of the UE 201 include cellular phones, smartphones, Session Initiation Protocol (SIP) phones, laptops, personal digital assistants (PDAs), satellite radios, GPS devices, multimedia devices, video devices, digital audio players (e.g., MP3 players), cameras, game consoles, drones, aircraft, narrowband physical network devices, machine-type communication devices, land vehicles, automobiles, wearable devices, or any other similar functional devices. Those skilled in the art may also refer to the UE 201 as a mobile station, subscriber station, mobile unit, subscriber unit, radio unit, remote unit, mobile device, radio device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, radio terminal, remote terminal, handheld device, user agent, mobile client, client, or any other suitable term. The Node B 203 is connected to the core network 210 via an S1 / NG interface.The core network 210 includes an MME (Mobility Management Entity) / AMF (Authentication Management Field) / SMF (Session Management Function) 211, other MMEs / AMFs / SMFs 214, an S-GW (Service Gateway) / UPF (User Plane Function) 212, and a P-GW (Packet Data Network Gateway) / UPF 213. The MME / AMF / SMF 211 is the control node that handles signaling between the UE 201 and the core network 210. Generally, the MME / AMF / SMF 211 provides bearer and connection management. All user IP (Internet Protocol) packets are transmitted through the S-GW / UPF 212, which is itself connected to the P-GW / UPF 213. The P-GW provides UE IP address allocation and other functions. The P-GW / UPF 213 is connected to the Internet service 230. The Internet service 230 includes carrier-compliant Internet protocol services, specifically including the Internet, intranet, IMS (IP Multimedia Subsystem), and packet-switched streaming services.

[0373] As an example, the first node in this application includes the UE 201.

[0374] As an example, the second node in this application includes node B 203.

[0375] As an example, node B 203 is a macrocell base station.

[0376] As an example, node B 203 is a microcell base station.

[0377] As an example, node B 203 is a pico cell base station.

[0378] As an example, node B 203 is a femtocell.

[0379] As an example, node B 203 is a base station device that supports large latency differences.

[0380] As an example, node B 203 is a flight platform device.

[0381] As an example, node B 203 is a satellite device.

[0382] As one embodiment, the node B 203 is a test device (e.g., a transceiver device simulating part of the base station's functions, a signaling tester).

[0383] As an example, the UE 201 includes a mobile phone.

[0384] As an example, the UE 201 is a vehicle including a car.

[0385] As an example, the wireless link from the UE 201 to the node B 203 is an uplink, which is used to perform uplink transmissions.

[0386] As an example, the radio link from the node B 203 to the UE 201 is a downlink, which is used to perform downlink transmissions.

[0387] As an example, the wireless link between the node B 203 and the UE 201 includes a cellular link.

[0388] As an example, the node B 203 and the UE 201 are connected via the Uu air interface.

[0389] As an example, the sender of the first configuration information block in this application includes the node B 203.

[0390] As an example, the recipient of the first configuration information block in this application includes the UE 201.

[0391] As an example, the sender of the first RS resource set in this application includes the node B 203.

[0392] As an example, the recipient of the first RS resource set in this application includes the UE 201.

[0393] As an example, the sender of the first information block in this application includes the node B 203.

[0394] As an example, the recipient of the first information block in this application includes the UE 201.

[0395] As an example, the sender of the second information block in this application includes the node B 203.

[0396] As an example, the recipient of the second information block in this application includes the UE 201.

[0397] As an example, the sender of the first message in this application includes the UE 201.

[0398] As an example, the recipient of the first message in this application includes the node B 203.

[0399] As an example, the sender of the target signaling described in this application includes the node B 203.

[0400] As an example, the recipient of the target signaling described in this application includes the UE 201.

[0401] As an example, node B 203 supports the deployment of network-side (NW-side) AI / ML models.

[0402] As an example, the UE 201 supports the deployment of UE-side AI / ML models.

[0403] As an example, the UE 201 supports a 5G system.

[0404] As an example, the node B 203 supports a 5G system.

[0405] As an example, the UE 201 supports at least a 6G system.

[0406] As an example, the node B 203 supports at least a 6G system.

[0407] Example 3

[0408] Example 3 illustrates a schematic diagram of an embodiment of a wireless protocol architecture for the user plane and control plane according to an embodiment of this application, as shown in the attached diagram. Figure 3 As shown.

[0409] Figure 3 This is a schematic diagram illustrating an embodiment of a wireless protocol architecture for the user plane 350 and the control plane 300. Figure 3The wireless protocol architecture for the control plane 300 between the first communication node device (UE or RSU in V2X, onboard equipment or onboard communication module) and the second node device (gNB, UE or RSU in V2X, onboard equipment or onboard communication module), or between two UEs, is illustrated using three layers: Layer 1 (L1), Layer 2 (L2), and Layer 3 (L3). L1 is the lowest layer and implements various PHY (Physical layer) signal processing functions. L1 will be referred to as PHY 301 in this document. L2305 sits above PHY 301 and is responsible for the link between the first and second node devices, or between two UEs, via PHY 301. L2305 includes a MAC (Medium Access Control) sublayer 302, an RLC (Radio Link Control) sublayer 303, and a PDCP (Packet Data Convergence Protocol) sublayer 304, which terminate at the second node device. The PDCP sublayer 304 provides multiplexing between different radio bearers and logical channels. It also provides security through encrypted data packets and supports cross-cell mobility between the second communication node devices and the first communication node device. The RLC sublayer 303 provides upper-layer packet segmentation and reassembly, retransmission of lost packets, and packet reordering to compensate for out-of-order reception due to HARQ (Hybrid Automatic Repeat Quest). The MAC sublayer 302 provides multiplexing between logical and transport channels. It is also responsible for allocating various radio resources (e.g., resource blocks) within a cell among the first communication node devices. The MAC sublayer 302 is also responsible for HARQ operations. The RRC (Radio Resource Control) sublayer 306 in L3 of the control plane 300 is responsible for obtaining radio resources (i.e., radio bearers) and using RRC signaling between the second communication node device and the first communication node device to configure the lower layer.The wireless protocol architecture of user plane 350 includes Layer 1 (L1) and Layer 2 (L2). The wireless protocol architecture for the first and second communication node devices in user plane 350 is largely the same as the corresponding layers and sublayers in control plane 300 for Physical Layer 351, PDCP sublayer 354 in L2355, RLC sublayer 353 in L2355, and MAC sublayer 352 in L2355. However, PDCP sublayer 354 also provides header compression for upper-layer packets to reduce wireless transmission overhead. L2355 in user plane 350 also includes SDAP (Service Data Adaptation Protocol) sublayer 356. SDAP sublayer 356 is responsible for mapping between QoS (Quality of Service) streams and Data Radio Bearer (DRB) to support service diversity. Although not illustrated, the first communication node device may have several upper layers above L2355, including a network layer (e.g., IP (Internet Protocol) layer) terminating at the P-GW on the network side and an application layer terminating at the other end of the connection (e.g., remote UE, server, etc.).

[0410] As an example, Appendix Figure 3 The wireless protocol architecture described herein is applicable to the first node in this application.

[0411] As an example, Appendix Figure 3 The wireless protocol architecture described herein is applicable to the second node in this application.

[0412] As an example, the first configuration information block in this application is generated in the RRC 306.

[0413] As an example, in this application, the first information block is generated in the RRC 306.

[0414] As an example, the second information block in this application is generated in the RRC 306.

[0415] As an example, the target signaling described in this application is generated in MAC 302 or MAC 352.

[0416] As an example, the target signaling described in this application is generated in the PHY301 or the PHY351.

[0417] As an example, in this application, the first message is generated in the RRC 306.

[0418] As an example, in this application, the first message is generated by MAC 302 or MAC 352.

[0419] As an example, in this application, the first message is generated in the PHY301 or the PHY351.

[0420] As an example, the higher layer mentioned in this application refers to the layer above the physical layer.

[0421] As an example, the higher layer described in this application includes the RRC layer.

[0422] As an example, the higher-layer signaling described in this application includes RRC IE.

[0423] As an example, the higher-level signaling described in this application includes RRC messages.

[0424] As an example, the higher layer described in this application includes the MAC layer.

[0425] As an example, the higher-layer signaling described in this application includes MAC CE.

[0426] Example 4

[0427] Example 4 illustrates a schematic diagram of a first communication device and a second communication device according to an embodiment of this application, as shown in the attached diagram. Figure 4 As shown. (Attached) Figure 4 This is a block diagram of a first communication device 410 and a second communication device 450 communicating with each other in an access network.

[0428] The first communication device 410 includes a controller / processor 475, a memory 476, a receiver processor 470, a transmitter processor 416, a multi-antenna receiver processor 472, a multi-antenna transmitter processor 471, a transmitter / receiver 418, and an antenna 420.

[0429] The second communication device 450 includes a controller / processor 459, a memory 460, a data source 467, a transmitting processor 468, a receiving processor 456, a multi-antenna transmitting processor 457, a multi-antenna receiving processor 458, a transmitter / receiver 454, and an antenna 452.

[0430] In the transmission from the first communication device 410 to the second communication device 450, at the first communication device 410, upper-layer data packets from the core network are provided to the controller / processor 475. The controller / processor 475 implements L2 functionality. In the DL, the controller / processor 475 provides header compression, encryption, packet segmentation and reordering, multiplexing between logical and transport channels, and radio resource allocation to the second communication device 450 based on various priority metrics. The controller / processor 475 is also responsible for HARQ operation, retransmission of lost packets, and signaling to the second communication device 450. The transmit processor 416 and the multi-antenna transmit processor 471 implement various signal processing functions for L1 (i.e., the physical layer). Transmit processor 416 performs encoding and interleaving to facilitate forward error correction (FEC) at the second communication device 450, and mapping of signal clusters based on various modulation schemes (e.g., Binary Phase Shift Keying (BPSK), Quadrature Phase Shift Keying (QPSK), M-PSK, and M-Quadrature Amplitude Modulation (M-QAM)). Multi-antenna transmit processor 471 performs digital spatial precoding on the encoded and modulated symbols, including codebook-based precoding and non-codebook-based precoding, and beamforming processing, generating one or more parallel streams. The transmit processor 416 then maps each parallel stream to a subcarrier, multiplexes the modulated symbols with a reference signal (e.g., a pilot) in the time and / or frequency domains, and then uses an inverse fast fourier transform (IFFT) to generate a physical channel carrying the time-domain multicarrier symbol stream. The multi-antenna transmit processor 471 then performs transmit analog precoding / beamforming operations on the time-domain multicarrier symbol stream. Each transmitter 418 converts the baseband multicarrier symbol stream provided by the multi-antenna transmit processor 471 into an RF stream, which is then provided to a different antenna 420.

[0431] In the transmission from the first communication device 410 to the second communication device 450, at the second communication device 450, each receiver 454 receives a signal through its corresponding antenna 452. Each receiver 454 recovers the information modulated onto the radio frequency carrier and converts the radio frequency stream into a baseband multicarrier symbol stream, which is then provided to the receiver processor 456. The receiver processor 456 and the multi-antenna receiver processor 458 implement various L1 signal processing functions. The multi-antenna receiver processor 458 performs receive analog precoding / beamforming operations on the baseband multicarrier symbol stream from the receiver 454. The receiver processor 456 uses a Fast Fourier Transform (FFT) to convert the baseband multicarrier symbol stream after the receive analog precoding / beamforming operations from the time domain to the frequency domain. In the frequency domain, the physical layer data signal and the reference signal are demultiplexed by the receiver processor 456, where the reference signal is used for channel estimation, and the data signal is recovered in the multi-antenna receiver processor 458 after multi-antenna detection to recover any parallel stream destined for the second communication device 450. Symbols on each parallel stream are demodulated and recovered in the receive processor 456, generating soft decisions. The receive processor 456 then decodes and deinterleaves the soft decisions to recover the upper-layer data and control signals transmitted by the first communication device 410 over the physical channel. The upper-layer data and control signals are then provided to the controller / processor 459. The controller / processor 459 implements L2 functionality. The controller / processor 459 may be associated with a memory 460 storing program code and data. The memory 460 may be referred to as computer-readable media. In the DL, the controller / processor 459 provides multiplexing, packet reassembly, decryption, header decompression, and control signal processing between the transmission and logical channels to recover upper-layer packets from the core network. The upper-layer packets are then provided to all protocol layers above L2. Various control signals may also be provided to L3 for L3 processing. The controller / processor 459 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operation.

[0432] In the transmission from the second communication device 450 to the first communication device 410, at the second communication device 450, a data source 467 is used to provide upper-layer data packets to the controller / processor 459. The data source 467 represents all protocol layers above L2. Similar to the transmission functions at the first communication device 410 described in the DL, the controller / processor 459 implements header compression, encryption, packet segmentation and reordering, and multiplexing between logical and transport channels based on the radio resource allocation of the first communication device 410, implementing L2 functions for the user plane and control plane. The controller / processor 459 is also responsible for HARQ operations, retransmission of lost packets, and signaling to the first communication device 410. Transmit processor 468 performs modulation mapping and channel coding processing, while multi-antenna transmit processor 457 performs digital multi-antenna spatial precoding, including codebook-based and non-codebook-based precoding, and beamforming processing. Subsequently, transmit processor 468 modulates the generated parallel stream into a multi-carrier / single-carrier symbol stream. After analog precoding / beamforming operations in multi-antenna transmit processor 457, the stream is provided to different antennas 452 via transmitter 454. Each transmitter 454 first converts the baseband symbol stream provided by multi-antenna transmit processor 457 into a radio frequency symbol stream before providing it to antenna 452.

[0433] In the transmission from the second communication device 450 to the first communication device 410, the function at the first communication device 410 is similar to the receiving function at the second communication device 450 described in the transmission from the first communication device 410 to the second communication device 450. Each receiver 418 receives radio frequency signals through its corresponding antenna 420, converts the received radio frequency signals into baseband signals, and provides the baseband signals to the multi-antenna receiving processor 472 and the receiving processor 470. The receiving processor 470 and the multi-antenna receiving processor 472 jointly implement the L1 function. The controller / processor 475 implements the L2 function. The controller / processor 475 may be associated with a memory 476 storing program code and data. The memory 476 may be referred to as computer-readable media. The controller / processor 475 provides multiplexing, packet reassembly, decryption, header decompression, and control signal processing between the transmission and logical channels to recover upper-layer data packets from the second communication device 450. The upper-layer data packets from the controller / processor 475 may be provided to the core network. The controller / processor 475 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operation.

[0434] As one embodiment, the second communication device 450 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The second communication device 450 apparatus at least receives the first configuration information block in this application, the first configuration information block indicating a first inference configuration; receives the first RS resource set in this application, the first RS resource set including one or more RS resources; performs at least one performance monitoring for the first inference configuration for each first time length in this application; the performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0435] As one embodiment, the second communication device 450 includes: a memory storing a computer-readable instruction program that produces actions when executed by at least one processor, the actions including: receiving the first configuration information block in this application; receiving the first RS resource set in this application; and performing at least one performance monitoring for the first inference configuration in this application for each first time length in this application.

[0436] As one embodiment, the first communication device 410 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The first communication device 410 transmits at least the first configuration information block in this application, the first configuration information block indicating a first inference configuration; transmits the first RS resource set in this application, the first RS resource set including one or more RS resources; the recipient of the first configuration information block is the second communication device 450, the second communication device 450 performing at least one performance monitoring for the first inference configuration every first time length in this application; the performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0437] As one embodiment, the first communication device 410 includes: a memory storing a computer-readable instruction program, which generates actions when executed by at least one processor, the actions including: sending the first configuration information block in this application; and sending the first RS resource set in this application.

[0438] As an example, the first node in this application includes the second communication device 450.

[0439] As an example, the second node in this application includes the first communication device 410.

[0440] As an example, at least one of {the antenna 420, the transmitter 418, the transmitter processor 416, the multi-antenna transmitter processor 471, the controller / processor 475, and the memory 476} is used to transmit the first configuration information block in this application; at least one of {the antenna 452, the receiver 454, the receiver processor 456, the multi-antenna receiver processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first configuration information block in this application.

[0441] As an example, at least one of {the antenna 420, the transmitter 418, the transmitter processor 416, the multi-antenna transmitter processor 471, the controller / processor 475, and the memory 476} is used to transmit the first RS resource set in this application; at least one of {the antenna 452, the receiver 454, the receiver processor 456, the multi-antenna receiver processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first RS resource set in this application.

[0442] As an example, at least one of {the antenna 452, the receiver 454, the receiver processor 456, the multi-antenna receiver processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to perform at least one performance monitoring for the first inference configuration in this application for the first time length in each application.

[0443] As an example, at least one of {the antenna 420, the transmitter 418, the transmitter processor 416, the multi-antenna transmitter processor 471, the controller / processor 475, and the memory 476} is used to transmit the first information block in this application; at least one of {the antenna 452, the receiver 454, the receiver processor 456, the multi-antenna receiver processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first information block in this application.

[0444] As an example, at least one of {the antenna 420, the transmitter 418, the transmitter processor 416, the multi-antenna transmitter processor 471, the controller / processor 475, and the memory 476} is used to transmit the second information block in this application; at least one of {the antenna 452, the receiver 454, the receiver processor 456, the multi-antenna receiver processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the second information block in this application.

[0445] As an example, at least one of {the antenna 452, the transmitter / receiver 454, the transmission processor 468, the multi-antenna transmission processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first message in this application; at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} is used to receive the first message in this application.

[0446] As an example, at least one of {the antenna 420, the transmitter 418, the transmission processor 416, the multi-antenna transmission processor 471, the controller / processor 475, and the memory 476} is used to transmit the target signaling described in this application; at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the target signaling described in this application.

[0447] Example 5

[0448] Example 5 illustrates a first flowchart of transmission between a first node and a second node according to an embodiment of this application, as shown in the attached diagram. Figure 5 As shown. In the appendix Figure 5 In this embodiment, the first node U1 and the second node N2 communicate via a wireless link. It should be noted that the order in this embodiment does not limit the signal transmission order or the order of implementation in this application.

[0449] For the first node U1, in step S510, a first configuration information block is received; in step S511, a first RS resource set is received; and in step S512, at least one performance monitoring is performed for the first inference configuration every first time length.

[0450] For the second node N2, the first configuration information block is sent in step S520; the first RS resource set is sent in step S521.

[0451] In Example 5, the first configuration information block indicates the first inference configuration; the first RS resource set includes one or more RS resources; the performance monitoring depends on the measurement on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0452] As an example, the first node U1 is the first node in this application.

[0453] As an example, the second node N2 is the second node in this application.

[0454] As one embodiment, the air interface between the second node N2 and the first node U1 includes a wireless interface between the base station equipment and the user equipment.

[0455] As one embodiment, the air interface between the second node N2 and the first node U1 includes a wireless interface between the relay node device and the user equipment.

[0456] As one embodiment, the air interface between the second node N2 and the first node U1 includes a wireless interface between user equipment and user equipment.

[0457] As one example, the second node N2 and the first node U1 communicate via the Uu interface.

[0458] As one example, the second node N2 is the maintenance base station of the serving cell of the first node U1.

[0459] As an example, the transmission channel occupied by the first configuration information block includes DL-SCH (DownLink-Shared Channel).

[0460] As an example, the physical layer channel occupied by the first configuration information block includes PDSCH (Physical Downlink Shared Channel).

[0461] As an example, step S521 is after step S510; step S511 is after step S520.

[0462] As an example, step S512 is performed after step S511.

[0463] As an example, steps S512 and S511 occur simultaneously.

[0464] As an example, this application does not limit the order of steps S512 and S511. One possible implementation is that the first node U1 periodically measures the first RS resource set, and performs performance monitoring at least once every first time length based on the measurement results of the first RS resource set.

[0465] Example 6

[0466] Example 6 illustrates a second flowchart of transmission between a first node and a second node according to an embodiment of this application, as shown in the attached diagram. Figure 6 As shown. In the appendix Figure 6 In this process, the first node U3 and the second node N4 communicate via a wireless link.

[0467] For the first node U3, the first information block is received in step S630.

[0468] For the second node N4, the first information block is sent in step S640.

[0469] In Example 6, the first information block is configured with a good inference performance criterion; the first time length depends on whether the good inference performance criterion is met.

[0470] As an example, the first node U3 is the first node in this application.

[0471] As an example, the second node N4 is the second node in this application.

[0472] As one embodiment, the air interface between the second node N4 and the first node U3 includes a wireless interface between the base station equipment and the user equipment.

[0473] As one embodiment, the air interface between the second node N4 and the first node U3 includes a wireless interface between the relay node device and the user equipment.

[0474] As one embodiment, the air interface between the second node N4 and the first node U3 includes a wireless interface between user equipment and user equipment.

[0475] As one example, the second node N4 and the first node U3 communicate via the Uu interface.

[0476] As one example, the second node N4 is the sustaining base station for the serving cell of the first node U3.

[0477] As an example, the first node U3 receives the first information block.

[0478] As one embodiment, the first information block is carried by higher-level signaling.

[0479] As an example, the first information block is transmitted via RRC signaling.

[0480] As one example, the first information block includes one or more RRC IEs.

[0481] As an example, the first information block includes one or more fields in an RRC IE.

[0482] As one embodiment, the first information block includes information from one or more fields of each of the plurality of RRC IEs.

[0483] As an example, the first information block is an RRC IE.

[0484] As an example, the first information block is UE-specific.

[0485] As one example, the name of the signaling used to transmit the first information block includes Object.

[0486] As one example, the name of the signaling used to transmit the first information block includes Inference.

[0487] As an example, the name of the signaling used to transmit the first information block includes AI.

[0488] As an example, the name of the signaling used to transmit the first information block includes ML.

[0489] As one example, the name of the signaling used to transmit the first information block includes "good".

[0490] As one example, the name of the signaling used to transmit the first information block includes monitoring.

[0491] As an example, the first information block configures the good inference performance criterion.

[0492] As an example, the good inference performance criterion is specific to the inference configuration.

[0493] As an example, the good inference performance criterion is specific to a particular inference configuration.

[0494] As an example, the good inference performance criterion is specific to a particular model.

[0495] As an example, the good inference performance criterion is specific to a particular functionality.

[0496] As an example, the results of the performance monitoring include a first metric.

[0497] As an example, the first metric corresponds to a performance metric.

[0498] As an example, the first metric corresponds to a KPI (Key Performance Indicator).

[0499] As an example, the first metric is the intermediate KPI.

[0500] As an example, the first metric is the eventual KPI.

[0501] As an example, the first metric includes the classification model KPI.

[0502] As an example, the first metric includes accuracy.

[0503] As an example, the first metric includes precision or accuracy.

[0504] As an example, the first metric includes recall.

[0505] As an example, the first metric includes sensitivity.

[0506] As an example, the first metric includes the F-score.

[0507] As an example, the first metric includes the F1-score.

[0508] As a sub-implementation of this embodiment, the F1-score depends on precision and recall.

[0509] As an example, the first metric includes AUC (Area Under Curve).

[0510] As a sub-example of this embodiment, the curve is a characteristic curve.

[0511] As a sub-example of this embodiment, the curve is the ROC (Receiver Operating Characteristic Curve).

[0512] As an example, the first metric includes the True Positive Rate (TPR).

[0513] As an example, the first metric includes the false positive rate (FPR).

[0514] As an example, the first metric includes RSRP prediction accuracy (RSRP accuracy).

[0515] As an example, the first metric includes regression model KPIs.

[0516] As an example, candidates for the first metric include one or more of GCS (Generalized Cosine Similarity), SGSC (Squared Generalized Cosine Similarity), NMSE (Normalized Mean Squared Error), truth groundCSI, equivalent MSE (equivalent Mean Squared Error), and numerical spectral efficiency gap.

[0517] As an example, candidates for the first parameter include one or more of throughput, BLER (Block Error Rate), and hypothetical BLER.

[0518] As an example, the first metric is NMSE.

[0519] As an example, the first metric is SGCS.

[0520] As an example, the first metric is truth ground CSI.

[0521] As an example, the first metric includes Euclidean distance.

[0522] As an example, the first metric includes Manhattan distance.

[0523] As an example, the first metric includes L1 distance.

[0524] As an example, the first metric includes the Pearson correlation coefficient.

[0525] As an example, the first metric includes Spearman's Rank Correlation Coefficient.

[0526] As an example, the type of the first metric depends on the type of the model or functionality associated with the first inference configuration.

[0527] As an example, the type of the first metric is predefined.

[0528] As an example, the types of models include classification models and regression models.

[0529] As an example, the type of the first metric includes a metric for a classification model and a metric for a regression model.

[0530] As an example, the first configuration information block indicates the type of the first metric.

[0531] As an example, the first node implements the calculation of the first metric accordingly.

[0532] As one example, the first threshold is predefined or configured by a higher-level signaling layer.

[0533] As an example, the first threshold is predefined.

[0534] As an example, the first threshold is fixed.

[0535] As one example, the first threshold is configured by higher-layer signaling.

[0536] As an example, the first threshold is configured via RRC signaling.

[0537] As an example, the first threshold is configured by the first configuration information block.

[0538] As an example, the first threshold is configured in the first information block.

[0539] As an example, the first threshold is indicated by the first information block.

[0540] As an example, the first metric is confidence level, and the first threshold corresponds to a number between 0 and 100.

[0541] As an example, the good inference performance criterion includes the first metric satisfying the first threshold.

[0542] As an example, the good inference performance criterion includes that, within a second time length, the first metric satisfies a first threshold, and the second time length is not less than the first time length.

[0543] As an example, the good inference performance criterion includes that, within a second time length, the first metric obtained by the first node satisfies a first threshold.

[0544] As an example, the good inference performance criterion includes that the time for the first metric to satisfy the first threshold is not less than the second time length.

[0545] As an example, the good inference performance criterion includes the first metric satisfying a first threshold and the duration being not less than the second time length.

[0546] As an example, the good inference performance criterion includes that the first metric satisfies a first threshold for the second time length.

[0547] As one example, the second time length is greater than the first time length.

[0548] As one example, the second time length is equal to the first time length.

[0549] As one example, the second time length is configured by higher-layer signaling.

[0550] As one example, the second time length is configured via RRC signaling.

[0551] As one embodiment, the second time length is configured through the first information block.

[0552] As an example, the second time length is predefined.

[0553] As one embodiment, the second time length is M times the first time length, where M is a positive integer.

[0554] As a sub-implementation of this embodiment, M is a higher-layer signaling configuration.

[0555] As a sub-implementation of this embodiment, M is an RRC signaling configuration.

[0556] As a sub-implementation of this embodiment, M depends on the first inference configuration.

[0557] As a sub-implementation of this embodiment, M is indicated by the first inference configuration.

[0558] As a sub-implementation of this embodiment, M is indicated by the first information block.

[0559] As a sub-implementation of this embodiment, M is fixed.

[0560] As an example, the good inference performance criterion includes at least K consecutive times the first metric satisfies a first threshold, where K is a positive integer greater than 1.

[0561] As an example, the good inference performance criterion includes K consecutive times the first metric satisfies a first threshold, where K is a positive integer greater than 1.

[0562] As an example, the good inference performance criterion includes K consecutive times the first metric satisfies a first threshold within the first time length, where K is a positive integer greater than 1.

[0563] As an example, the good inference performance criterion includes K consecutive times the first metric satisfies a first threshold within the second time length, where K is a positive integer greater than 1.

[0564] As an example, the good inference performance criterion includes satisfying a first threshold after K consecutive filtering of the first metric, where K is a positive integer greater than 1.

[0565] As an example, K equals 3.

[0566] As an example, K is a higher-layer signaling configuration.

[0567] As an example, K is an RRC signaling configuration.

[0568] As an example, K depends on the first inference configuration.

[0569] As an example, K is indicated by the first inference configuration.

[0570] As an example, K is indicated by the first information block.

[0571] As an example, K is fixed.

[0572] As an example, the first metric satisfying the first threshold means that the first metric is better than the first threshold.

[0573] As an example, the first metric satisfying the first threshold means that the first metric is not worse than the first threshold.

[0574] As an example, the first metric satisfying the first threshold means that the first metric is greater than the first threshold.

[0575] As an example, the first metric satisfying the first threshold means that the first metric is not less than the first threshold.

[0576] As an example, the first metric satisfying the first threshold means that the first metric is less than the first threshold.

[0577] As an example, the first metric satisfying the first threshold means that the first metric is not greater than the first threshold.

[0578] As an example, the first time length depends on whether the good inference performance criterion is met.

[0579] As an example, whether the good inference performance criterion is met is specific to the model associated with the first inference configuration, and the good inference performance depends on the output of the performance monitoring.

[0580] As an example, whether the good inference performance criterion is met is specific to the model associated with the first inference configuration. The good inference performance refers to whether the output of the performance monitoring meets the good inference performance criterion.

[0581] As an example, the first time length depends on whether the model or functionality associated with the first inference configuration meets the good inference performance criterion.

[0582] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length; if the model or functionality associated with the first inference configuration satisfies the good inference performance criterion, the first time length is the first candidate time length; if the model or functionality associated with the first inference configuration does not satisfy the good inference performance criterion, the first time length is the second candidate time length.

[0583] As a sub-implementation of this embodiment, the length of the first candidate time is greater than the length of the second candidate time.

[0584] As a sub-implementation of this embodiment, both the first candidate time length and the second candidate time length are indicated by higher-layer signaling.

[0585] As a sub-implementation of this embodiment, both the first candidate time length and the second candidate time length are predefined.

[0586] As a sub-implementation of this embodiment, the second candidate time length is indicated by higher-layer signaling, while the first candidate time length is UE-implemented.

[0587] As an example, the first time length belongs to a first time length group, and the candidates of the first time length group include a first candidate time length group and a second candidate time length group; if the model or functionality associated with the first inference configuration satisfies the good inference performance criterion, the first time length group is the first candidate time length group; if the model or functionality associated with the first inference configuration does not satisfy the good inference performance criterion, the first time length group is the second candidate time length group.

[0588] As a sub-implementation of this embodiment, the first candidate time length group and the second candidate time length group each include one or more time lengths, each of the one or more time lengths corresponding to an inference configuration, and the first inference configuration indicating the first time length from the first time length group.

[0589] As an example, the first time length depends on a first time parameter, which in turn depends on whether the model or functionality associated with the first inference configuration meets the good inference performance criterion.

[0590] As a sub-implementation of this embodiment, the first time length is linearly related to the first time parameter. When the model or functionality associated with the first inference configuration satisfies the good inference performance criterion, the value of the first time parameter is a first value; when the model or functionality associated with the first inference configuration satisfies the good inference performance criterion, the value of the first time parameter is a second value, and the first value is greater than the second value.

[0591] As an example, the transmission channel occupied by the first information block includes DL-SCH.

[0592] As an example, the physical layer channel occupied by the first information block includes PDSCH.

[0593] As an example, step S630 occurs before step S510; step S640 occurs before step S520.

[0594] As an example, step S630 is after step S510; step S640 is after step S520.

[0595] As an example, steps S630 and S510 occur simultaneously; steps S640 and S520 occur simultaneously.

[0596] As a sub-implementation of this embodiment, the first configuration information block and the first information block are carried by the same PDSCH.

[0597] As a sub-implementation of this embodiment, the first configuration information block and the first information block belong to different domains of the same RRC IE.

[0598] Example 7

[0599] Example 7 illustrates a third flowchart of transmission between a first node and a second node according to an embodiment of this application, as shown in the attached diagram. Figure 7 As shown. In the appendix Figure 7 In this process, the first node U5 and the second node N6 communicate via a wireless link.

[0600] For the first node U5, the second information block is received in step S750.

[0601] For the second node N6, the second information block is sent in step S760.

[0602] In embodiment 7, the second information block is configured with at least the former of the low mobility criterion and the non-cell edge criterion; the first inference is configured for low mobility, and the first time length depends on whether at least the former of the low mobility criterion and the non-cell edge criterion is satisfied.

[0603] As an example, the first node U5 is the first node in this application.

[0604] As an example, the second node N6 is the second node in this application.

[0605] As one embodiment, the air interface between the second node N6 and the first node U5 includes a wireless interface between the base station equipment and the user equipment.

[0606] As one embodiment, the air interface between the second node N6 and the first node U5 includes a wireless interface between the relay node device and the user equipment.

[0607] As one embodiment, the air interface between the second node N6 and the first node U5 includes a wireless interface between user equipment and user equipment.

[0608] As one example, the second node N6 and the first node U5 communicate via the Uu interface.

[0609] As one example, the second node N6 is the sustaining base station for the serving cell of the first node U5.

[0610] As one example, the first node U5 receives the second information block.

[0611] As one embodiment, the second information block is carried by higher-level signaling.

[0612] As one embodiment, the second information block is transmitted via RRC signaling.

[0613] As one embodiment, the second information block includes one or more RRC IEs.

[0614] As one example, the second information block includes one or more fields in an RRC IE.

[0615] As one embodiment, the second information block includes information from one or more fields of each of the plurality of RRC IEs.

[0616] As an example, the second information block is an RRC IE.

[0617] As one embodiment, the second information block is transmitted via SIB (System Information Block).

[0618] As one example, the second information block is transmitted via SIB2.

[0619] As an example, the name of the signaling used to transmit the second information block includes lowMobilityEvaluation.

[0620] As one example, the name of the signaling used to transmit the second information block includes cellEdgeEvaluation.

[0621] As an example, the name of the signaling used to transmit the second information block includes relaxedMeasurement.

[0622] As an example, the second information block configures at least the former of the low mobility criterion and the non-cell edge criterion.

[0623] As one embodiment, the second information block configures the low mobility criteria.

[0624] As an example, the second information block configures the low mobility criterion and the not-at-cell-edge criterion.

[0625] As an example, the low mobility criterion satisfies Srxlev Ref The difference between subtracting Srxlev and S is less than S. SearchDeltaP Wherein Srxlev is the Srxlev value of the current serving cell, and Srxlev... Ref It is the reference Srxlev value of the currently serving cell, where S... SearchDeltaP It is the threshold for the change in Srxlev measured during relaxation.

[0626] As an example, the "not at the cell edge" criterion satisfies Srxlev less than S SearchThresholdP Wherein, Srxlev is the Srxlev value of the current serving cell, and S... SearchThresholdP It is the Srxlev threshold for relaxation measurement.

[0627] As an example, the "not at the cell edge" criterion satisfies Squal less than S SearchThresholdQ Wherein, Squal is the Squal value of the current serving cell, and S... SearchThresholdQ It is the Squal threshold for relaxation measurement.

[0628] As an example, the "not at the cell edge" criterion satisfies Srxlev less than S SearchThresholdP And Squal is less than S SearchThresholdQ Wherein, Srxlev is the Srxlev value of the current serving cell, and S... SearchThresholdP It is the Srxlev threshold for relaxation measurement, where Squal is the Squal value of the current serving cell, and S... SearchThresholdQ It is the Squal threshold for relaxation measurement.

[0629] As an example, Srxlev refers to the received strength value selected by the cell.

[0630] As an example, the unit of Srxlev is dB.

[0631] As an example, the Srxlev Ref The unit is dB.

[0632] As an example, Squal refers to the quality value selected by the cell.

[0633] As an example, the unit of Squal is dB.

[0634] As an example, the S SearchDeltaP The unit is dB.

[0635] As an example, the S SearchDeltaP It is configured by higher-level signaling.

[0636] As an example, the S SearchDeltaP It's configured with SIB2.

[0637] As an example, the S SearchDeltaP It is configured in the second information block.

[0638] As an example, the S SearchThresholdP It is configured by higher-level signaling.

[0639] As an example, the S SearchThresholdP It's configured with SIB2.

[0640] As an example, the S SearchThresholdP It is configured in the second information block.

[0641] As an example, the S SearchThresholdQ It is configured by higher-level signaling.

[0642] As an example, the S SearchThresholdQ It's configured with SIB2.

[0643] As an example, the S SearchThresholdQ It is configured in the second information block.

[0644] As an example, the specific definition of Srxlev can be found in section 5.2.3.2 of 3GPP TS38.304v i.0.0.

[0645] As an example, the Srxlev Ref For the specific definition, please refer to section 5.2.4.9.1 of 3GPP TS38.304v i.0.0.

[0646] As an example, the first inference is configured for low mobility.

[0647] As an example, the first inference is configured as a low-mobility UE.

[0648] As an example, the model or functionality associated with the first inference configuration is for a UE in a low-mobility scenario.

[0649] As an example, the training of the model or functionality associated with the first inference configuration is based on data collected by low-mobility UEs.

[0650] As an example, the first inference configuration is applicable when the first node U5 is in a low mobility state.

[0651] As an example, the reasoning of the first reasoning configuration is applicable when the first node U5 is in a low mobility state.

[0652] As an example, the first inference configuration indicates that it applies when the first node U5 is in a low mobility state.

[0653] As an example, the first inference configuration is applied when the first node U5 meets the low mobility criterion.

[0654] As an example, the model or functionality associated with the first inference configuration meets the requirements for model applicability when the first node U5 processes low mobility states.

[0655] As an example, the first time length depends on whether at least the low mobility criterion and the non-cell edge criterion are met.

[0656] As one embodiment, the second information block configures the low mobility criterion, and the first time length depends on whether the low mobility criterion is met.

[0657] As a sub-implementation of this embodiment, the candidates for the first time length include a first candidate time length and a second candidate time length; if the first node U5 satisfies the low mobility criterion, the first time length is the first candidate time length; if the first node U5 does not satisfy the low mobility criterion, the first time length is the second candidate time length.

[0658] As a supplementary embodiment of this sub-example, the length of the first candidate time is greater than the length of the second candidate time.

[0659] As a supplementary embodiment of this sub-example, both the first candidate time length and the second candidate time length are indicated by higher-layer signaling.

[0660] As an additional embodiment of this sub-example, both the first candidate time length and the second candidate time length are predefined.

[0661] As a supplementary embodiment of this sub-example, the second candidate time length is indicated by higher-layer signaling, while the first candidate time length is UE-implemented.

[0662] As a sub-implementation of this embodiment, the first time length belongs to the first time length group, and the candidates of the first time length group include the first candidate time length group and the second candidate time length group; if the first node U5 satisfies the low mobility criterion, the first time length group is the first candidate time length group; if the first node U5 does not satisfy the low mobility criterion, the first time length group is the second candidate time length group.

[0663] As an additional embodiment of this sub-example, the first candidate time length group and the second candidate time length group each include one or more time lengths, each of the one or more time lengths corresponding to an inference configuration, and the first inference configuration indicating the first time length from the first time length group.

[0664] As a sub-implementation of this embodiment, the first time length depends on a second time parameter, which in turn depends on whether the first node U5 meets the low mobility criterion.

[0665] As an example, the second information block configures the low mobility criterion and the not-at-cell-edge criterion, and the first time length depends on whether the low mobility criterion is met.

[0666] As a sub-example of this embodiment, the first node U5 is not configured with a first higher-level parameter, which indicates that the first node U5 needs to simultaneously meet the low mobility criterion and the not-at-cell-edge criterion.

[0667] As a sub-implementation of this embodiment, the candidates for the first time length include a first candidate time length and a second candidate time length; if the first node U5 satisfies the low mobility criterion, the first time length is the first candidate time length; if the first node U5 does not satisfy the low mobility criterion, the first time length is the second candidate time length.

[0668] As a supplementary embodiment of this sub-example, the length of the first candidate time is greater than the length of the second candidate time.

[0669] As a supplementary embodiment of this sub-example, both the first candidate time length and the second candidate time length are indicated by higher-layer signaling.

[0670] As an additional embodiment of this sub-example, both the first candidate time length and the second candidate time length are predefined.

[0671] As a supplementary embodiment of this sub-example, the second candidate time length is indicated by higher-layer signaling, while the first candidate time length is UE-implemented.

[0672] As a sub-implementation of this embodiment, the first time length belongs to the first time length group, and the candidates of the first time length group include the first candidate time length group and the second candidate time length group; if the first node U5 satisfies the low mobility criterion, the first time length group is the first candidate time length group; if the first node U5 does not satisfy the low mobility criterion, the first time length group is the second candidate time length group.

[0673] As an additional embodiment of this sub-example, the first candidate time length group and the second candidate time length group each include one or more time lengths, each of the one or more time lengths corresponding to an inference configuration, and the first inference configuration indicating the first time length from the first time length group.

[0674] As a sub-implementation of this embodiment, the first time length depends on a second time parameter, which in turn depends on whether the first node U5 meets the low mobility criterion.

[0675] As one embodiment, the second information block configures the low mobility criterion and the not-at-cell-edge criterion, and the first time length depends on whether the low mobility criterion and the not-at-cell-edge criterion are met.

[0676] As a sub-example of this embodiment, the first node U5 is configured with a first higher-level parameter, which indicates that the first node U5 needs to simultaneously meet the low mobility criterion and the not-at-cell-edge criterion.

[0677] As a sub-implementation of this embodiment, the candidates for the first time length include a first candidate time length and a second candidate time length; if the first node U5 simultaneously satisfies the low mobility criterion and the not-at-cell-edge criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0678] As a supplementary embodiment of this sub-example, the length of the first candidate time is greater than the length of the second candidate time.

[0679] As a supplementary embodiment of this sub-example, both the first candidate time length and the second candidate time length are indicated by higher-layer signaling.

[0680] As an additional embodiment of this sub-example, both the first candidate time length and the second candidate time length are predefined.

[0681] As a supplementary embodiment of this sub-example, the second candidate time length is indicated by higher-layer signaling, while the first candidate time length is UE-implemented.

[0682] As a sub-implementation of this embodiment, the first time length belongs to the first time length group, and the candidates of the first time length group include the first candidate time length group and the second candidate time length group; if the first node U5 simultaneously satisfies the low mobility criterion and the not at the cell edge criterion, the first time length group is the first candidate time length group; otherwise, the first time length group is the second candidate time length group.

[0683] As an additional embodiment of this sub-example, the first candidate time length group and the second candidate time length group each include one or more time lengths, each of the one or more time lengths corresponding to an inference configuration, and the first inference configuration indicating the first time length from the first time length group.

[0684] As a sub-implementation of this embodiment, the first time length depends on a second time parameter, which depends on whether the first node U5 simultaneously satisfies the low mobility criterion and the not-at-cell-edge criterion.

[0685] As a sub-implementation of this embodiment, the first time length is linearly related to the second time parameter. When the first node simultaneously satisfies the low mobility criterion and the non-cell edge criterion, the value of the second time parameter is a first value; otherwise, the value of the second time parameter is a second value, and the first value is greater than the second value.

[0686] As one embodiment, the transmission channel occupied by the second information block includes DL-SCH.

[0687] As one embodiment, the physical layer channel occupied by the second information block includes PDSCH.

[0688] As an example, step S750 precedes step S510; step S760 precedes step S520.

[0689] As an example, step S750 is after step S510; step S760 is after step S520.

[0690] As an example, steps S750 and S510 occur simultaneously; steps S760 and S520 occur simultaneously.

[0691] As a sub-implementation of this embodiment, the first configuration information block and the second information block are carried by the same PDSCH.

[0692] As a sub-implementation of this embodiment, the first configuration information block and the second information block belong to different domains of the same RRC IE.

[0693] As an example, step S750 occurs before step S630; step S760 occurs before step S640.

[0694] As an example, step S750 is after step S630; step S760 is after step S640.

[0695] As an example, steps S750 and S630 occur simultaneously; steps S760 and S640 occur simultaneously.

[0696] As a sub-implementation of this embodiment, the first information block and the second information block are carried by the same PDSCH.

[0697] As a sub-implementation of this embodiment, the first information block and the second information block belong to different domains of the same RRC IE.

[0698] Example 8

[0699] Example 8 illustrates a fourth flowchart of transmission between a first node and a second node according to an embodiment of this application, as shown in the attached diagram. Figure 8 As shown. In the appendix Figure 8 In this embodiment, the first node U7 and the second node N8 communicate via a wireless link. It should be noted that the order in this embodiment does not limit the signal transmission order or the order of implementation in this application.

[0700] For the first node U7, a first message is sent in step S870; target signaling is monitored in step S871.

[0701] For the second node N8, the first message is received in step S880.

[0702] In Example 8, the first message includes the result of the performance monitoring; monitoring target signaling, which is MAC layer signaling or physical layer signaling; the target signaling indicates the first time length from multiple time lengths.

[0703] As an example, the first node U7 is the first node in this application.

[0704] As an example, the second node N8 is the second node in this application.

[0705] As one embodiment, the air interface between the second node N8 and the first node U7 includes a wireless interface between the base station equipment and the user equipment.

[0706] As one embodiment, the air interface between the second node N8 and the first node U7 includes a wireless interface between the relay node device and the user equipment.

[0707] As one embodiment, the air interface between the second node N8 and the first node U7 includes a wireless interface between user equipment and user equipment.

[0708] As one example, the second node N8 and the first node U7 communicate via the Uu interface.

[0709] As one example, the second node N8 is the sustaining base station for the serving cell of the first node U7.

[0710] As an example, MAC refers to Media Access Control.

[0711] As an example, MAC refers to Media Access Control layer.

[0712] As an example, the first node U7 sends a first message.

[0713] As an example, the first node U7 periodically sends the first message.

[0714] As an example, the first node U7 sends the first message every first time interval.

[0715] As an example, the first message includes the inference result corresponding to the first inference configuration.

[0716] As an example, the first message does not include the reasoning result corresponding to the first reasoning.

[0717] As an example, the first message is carried by UCI (Uplink Control Information).

[0718] As an example, the first message is carried by L1 (Layer 1) signaling.

[0719] As an example, the first message includes the results of the performance monitoring.

[0720] As an example, the first message reports the results of the performance monitoring.

[0721] As an example, the results of the performance monitoring include confidence levels.

[0722] As an example, the first message indicates that the first node meets the good inference performance criterion.

[0723] As an example, the first message indicates that the first node meets the conditions for relaxing performance monitoring.

[0724] As an example, the first message requests to enter the relaxation performance monitoring.

[0725] As one example, the first message includes a request to enter relaxed performance monitoring.

[0726] As an example, the first message includes the inference output of the first inference configuration.

[0727] As an example, the first inference configuration indicates the temporal resources occupied by the first message.

[0728] As an example, the first inference configuration indicates the frequency domain resources occupied by the first message.

[0729] As one embodiment, the first inference configuration indicates the period for the first message reporting.

[0730] As an example, the first message includes the first metric in this application.

[0731] As an example, the results of the performance monitoring are intermediate-level KPIs.

[0732] As an example, the performance testing of the model or functionality associated with the first inference configuration is jointly performed by the first node U7 and the second node U8.

[0733] As an example, the model associated with the first inference configuration is a two-side model.

[0734] As an example, the functionality associated with the first inference configuration is two-sided.

[0735] As an example, the first node U7 monitors the target signaling.

[0736] As an example, the second node N8 sends the target signaling.

[0737] As an example, the second node N8 did not send the target signaling.

[0738] As an example, the monitoring means detection.

[0739] As an example, the monitoring means receiving.

[0740] As an example, the monitoring means searching.

[0741] As an example, the term "monitoring" includes listening.

[0742] As an example, the monitoring means including CRC (Cyclic redundancy check) verification.

[0743] As an example, the monitoring refers to blind decoding.

[0744] As an example, the monitoring refers to coherent detection.

[0745] As an example, the monitoring refers to energy detection.

[0746] As an example, the target signaling is MAC layer signaling or physical layer signaling.

[0747] As an example, the target signaling is MAC layer signaling.

[0748] As an example, the target signaling includes a MAC CE (Control Element).

[0749] As an example, the target signaling is MAC CE.

[0750] As an example, the target signaling includes a MAC PDU (Protocol Data Unit).

[0751] As an example, the target signaling is a MAC PDU.

[0752] As one example, the target signaling includes a MAC subheader.

[0753] As an example, the target signaling is the MAC subheader.

[0754] As an example, the target signaling is physical layer signaling.

[0755] As an example, the target signaling is L1 signaling.

[0756] As an example, the target signaling includes information in one or more fields of a DCI (Downlink Control Information).

[0757] As an example, the target signaling is a DCI.

[0758] As an example, the target signaling indicates the first time length from a plurality of time lengths.

[0759] As an example, the multiple time lengths form a time length table, and the target information indicates a row in the table.

[0760] As an example, the plurality of time lengths are predefined.

[0761] As an example, the plurality of time lengths are pre-configured.

[0762] As one embodiment, the target signaling includes a first field, which indicates the first time length via code points.

[0763] As one embodiment, the plurality of time lengths are two time lengths, and the target signaling includes a first field indicating the first time length from the two time lengths.

[0764] As an example, the target signaling indicates the first time length by indicating the time domain resources occupied by the first message.

[0765] As one embodiment, the target information indicates the first time length by indicating the period of the first message.

[0766] As an example, the target signaling indicates the first time length by instructing the model or functionality associated with the first inference configuration to enter a relaxed monitoring state.

[0767] As an example, the first node determines the first time length by whether or not the target signaling is detected.

[0768] As an example, the first message is transmitted on PUSCH (Physical Uplink Shared Channel).

[0769] As an example, the first message is transmitted on PUCCH (Physical Uplink Control Channel).

[0770] As an example, the transmission channel occupied by the target signaling includes DL-SCH.

[0771] As an example, the physical layer channel occupied by the target signaling includes PDSCH.

[0772] As an example, the physical layer channel occupied by the target signaling includes the PDCCH (Physical Downlink Control Channel).

[0773] As an example, step S871 is performed after step S870.

[0774] As an example, step S870 is performed after step S511.

[0775] As an example, this application does not limit the order of steps S870, S511 and S512. One possible implementation is that the first node in this application periodically measures the first RS resource set, and performs at least one performance monitoring based on the measurement results of the first RS resource set every first time length, and reports the results of the at least one performance monitoring to the second node in this application through the first message.

[0776] Example 9

[0777] Example 9 illustrates a schematic diagram of a first time length according to an embodiment of this application, as shown in the attached diagram. Figure 9 As shown.

[0778] As an example, the first node performs the performance monitoring at least once for the first inference configuration every first time period.

[0779] As an example, the first node performing at least one performance monitoring for the first inference configuration every first time length means that the first node should be able to evaluate whether the model or functionality associated with the first inference configuration meets the inference performance criteria every first time length.

[0780] As a sub-implementation of this embodiment, the inference performance criteria include criteria for whether the model or functionality is applicable.

[0781] As a sub-implementation of this embodiment, the inference performance criterion includes the good inference performance criterion described in this application.

[0782] As an example, the meaning of the first node performing at least one performance monitoring for the first inference configuration every first time length includes: the first node should calculate the performance metric of the model or functionality associated with the first inference configuration at least every first time length.

[0783] As an example, the first node performing at least one performance monitoring for the first inference configuration every first time length means that the first node should filter at least two performance metrics of the model or functionality associated with the first inference configuration, and the interval between the at least two performance metrics is at least equal to half of the first time length.

[0784] As an example, the meaning of the first node performing at least one performance monitoring for the first inference configuration every first time length includes: the first node performing at least one performance monitoring for the first inference configuration within the most recent first time length.

[0785] As an example, the meaning of the first node performing at least one performance monitoring for the first inference configuration every first time length includes: the first node should determine whether the model or functionality associated with the first inference configuration meets the good inference performance criteria within the most recent first time length.

[0786] As an example, the first node performing at least one performance monitoring for the first inference configuration every first time length means that the first node should filter at least two performance metrics of the model or functionality associated with the first inference configuration within the most recent first time length, and the interval between the at least two performance metrics is at least equal to half of the first time length.

[0787] Example 10

[0788] Example 10 illustrates a schematic diagram of a first inference configuration according to an embodiment of this application, as shown in the attached diagram. Figure 10 As shown. In the appendix Figure 10 In this context, the model or functionality associated with the first inference configuration is for a higher level.

[0789] As an example, the model or functionality associated with the first inference configuration is for a higher level.

[0790] As an example, the model or functionality associated with the first inference configuration is deployed at a higher level.

[0791] As an example, the model or functionality associated with the first inference configuration is for the MAC layer.

[0792] As an example, the model or functionality associated with the first inference configuration is deployed at the MAC layer.

[0793] As an example, the model or functionality associated with the first inference configuration is for L3.

[0794] As an example, the model or functionality associated with the first inference configuration is for RLM (Radio Link Monitoring).

[0795] As an example, the model or functionality associated with the first inference configuration is for BFD (BeamFailure Detection).

[0796] As an example, the model or functionality associated with the first inference configuration is for mobility management.

[0797] As an example, the model or functionality associated with the first inference configuration is for RRM (Radio Resource Management) measurements.

[0798] As an example, the model or functionality associated with the first inference configuration is for positioning.

[0799] As an example, the model or functionality associated with the first inference configuration is for cell selection.

[0800] As an example, the model or functionality associated with the first inference configuration is for cell-reselection.

[0801] As an example, the model or functionality associated with the first inference configuration is for L3-RSRP prediction.

[0802] As an example, the results of the performance monitoring require L3 filtering.

[0803] As an example, the results of the performance monitoring require at least L2 execution.

[0804] As an example, the result of the performance monitoring is a filtered value.

[0805] As an example, the results of the performance monitoring do not need to be dynamically reported.

[0806] As an example, the results of the performance monitoring depend on at least one counter.

[0807] As a sub-implementation of this embodiment, each of the at least one counter is maintained by the MAC layer.

[0808] As a sub-implementation of this embodiment, each of the at least one counter is maintained by the RRC layer.

[0809] As a sub-implementation of this embodiment, the at least one counter includes multiple counters, at least one of which is maintained by the MAC layer, and at least one of which is maintained by the RRC layer.

[0810] As an example, the performance monitoring structure relies on at least one timer.

[0811] As a sub-implementation of this embodiment, each of the at least one timer is maintained by the MAC layer.

[0812] As a sub-implementation of this embodiment, each of the at least one timer is maintained by the RRC layer.

[0813] As a sub-implementation of this embodiment, at least one of the multiple timers is maintained by the MAC layer and at least one of the multiple timers is maintained by the RRC layer.

[0814] Example 11

[0815] Example 11 illustrates a schematic diagram of a first parameter according to an embodiment of this application, as shown in the attached diagram. Figure 11 As shown. In the appendix Figure 11 In this context, the first time length depends on a first parameter, which indicates the downlink reception period.

[0816] As an example, the first time length depends on the first parameter.

[0817] As an example, the first parameter indicates the first time length.

[0818] As an example, the first parameter indicates a first time value, which is used to determine the first time length.

[0819] As an example, the first parameter indicates the downlink reception period.

[0820] As an example, the first parameter indicates the downlink reception period, the downlink reception period indicates a first time value, and the first time value is used to determine the first time length.

[0821] As an example, the first time length is equal to a positive integer multiple of the first time value that is greater than 1.

[0822] As an example, the first time value is equal to the DRX (Discontinuous Reception) period of the first node.

[0823] As an example, the first time value is the period of the SSB.

[0824] As a sub-implementation of this embodiment, the first RS resource set includes the SSB.

[0825] As an example, the first time value is related to the configuration period of the first RS resource set.

[0826] As an example, the first time value is equal to the configuration period of the first RS resource set.

[0827] As an example, the first time value is equal to a positive integer multiple of the configuration period of the first RS resource set.

[0828] As an example, the first time value is related to the SMTC (SSB-based Measurement Timing Configuration) period.

[0829] As a sub-example of the above embodiments, the SMTC period is configured by SSB-MeasurementTimingConfiguration.

[0830] As an example, the first time length is related to the SMTC period.

[0831] As an example, the first node enters an energy-saving mode during the downlink reception cycle and outside the included onDuration time period.

[0832] As an example, the first node abandons monitoring the downlink PDCCH during the downlink reception period and outside the included onDuration time period.

[0833] Example 12

[0834] Example 12 illustrates a schematic diagram of RAN domain AI / ML function deployment according to an embodiment of this application, as shown in the attached diagram. Figure 12 As shown. In the appendix Figure 12 In this context, gNB can be replaced with network equipment such as eNB or 6G base stations.

[0835] In Example 12, the management of the ML inference functions of multiple base stations is completed by the RAN domain management function 1202, that is, data interaction with the RAN domain MnS (Management Service) consumer / cross-domain management 1201 (as shown in the attached figure). Figure 12 (As shown by the dashed arrow in the diagram). The RAN domain ML training function 1203 is located in the RAN domain management function 1202; while the ML inference function is located in the base station, that is, the AI / ML inference function 1204 is located in gNB 1205, the AI / ML inference function 1206 is located in gNB 1207, and so on.

[0836] AI / ML related functions include ML training (also known as AI training or AI / ML training), ML testing, and ML inference (also known as AI inference or AI / ML inference), etc. ML training, ML testing, and ML inference functions can be deployed independently or co-located. Deployment of AI / ML related functions can be implemented through software, such as downloading and / or running executable files; or it can be implemented through a combination of software and hardware, such as accelerating specific computing units through hardware to improve computing speed or save power.

[0837] ML training functionality can be deployed in a cross-domain management system or a domain-specific management system; the domain-specific management system is used to manage the RAN domain or the CN (Core Network) domain. For example, ML training functionality for MDA (Management Data Analytics) can be deployed in MDAF (Management Data Analytics Function); ML training for network data analytics can be deployed in NWDAF (Network Data Analytics Function), meaning the ML training functionality is an MTLF (Model Training Logical Function).

[0838] The ML inference function can also be deployed in a cross-domain management system or a domain-specific management system; for example, the ML inference function is MDAF, or the ML inference function is AnLF (Analytics Logical Function) located in NWDAF.

[0839] Similarly, ML testing capabilities can also be deployed in cross-domain management systems or domain-specific management systems.

[0840] Optionally, the management of ML inference function can also be completed by the base station itself, that is, each base station can independently interact with the RAN domain MnS consumer / cross-domain management 1201.

[0841] It should be noted that Example 12 is merely a non-limiting implementation; optionally, the ML training function of the RAN domain may also be deployed at the base station; or optionally, some base stations may deploy both the ML inference function and the ML training function of the RAN domain, while some base stations may only deploy the ML inference function.

[0842] As an example, one of the gNBs (or base stations) in Example 12 is the second node of this application.

[0843] Example 13

[0844] Example 13 illustrates a schematic diagram of the deployment of AI / ML functions in a UE according to an embodiment of this application, as shown in the attached diagram. Figure 13 As shown. In the appendix Figure 13 In this context, the RAN domain ML training function 1304 is optional.

[0845] UE function 1303 is deployed in the first node of this application, and the UE function 1303 includes AI / ML inference function 1305; the AI / ML inference function 1305 uses an ML model (also called an AI model) for inference; an ML model is typically trained before being used for AI / ML inference.

[0846] As an example, the UE function 1303 includes a RAN domain ML training function 1304, which runs training data through an ML model to obtain a relevant loss and adjusts the parameters of the ML model based on the calculated loss; the ML training includes at least one of ML initial training, ML re-training, and reinforcement learning.

[0847] The above embodiments can reduce the complexity of the base station, or save air interface resources caused by reporting training data; however, the above embodiments place high demands on the processing capabilities of the UE side.

[0848] Optionally, the UE function 1303 also includes a CN domain ML training function ( Figure 13 (Not included in the text).

[0849] Optionally, the UE function 1303 also includes an AI / ML deployment function. Figure 13 It is not included in the list, which is used to load ML models and data.

[0850] As an example, the first node indicates whether it supports ML training function (RAN domain or CN domain) through capability reporting. The capability reporting is RRC signaling or NAS (Non-Access Stratum) signaling.

[0851] As an example, the ML model and the associated metadata are loaded by the first node from a network device or a remote server.

[0852] Optionally, the UE function 1303 is an MnS producer that provides data to the CN domain MnF (Management Function) and / or the RAN domain MnF and / or the cross-domain management system 1301 for management or analysis (as shown by the double arrow 1302).

[0853] Optionally, the UE function 1303 is an MnS consumer that loads data from the CN domain MnF and / or RAN domain MnF and / or cross-domain management system 1301 for AI / ML-related management, such as managing data requests, ML model activation, and / or ML training (as shown by double arrow 1302).

[0854] As an example, the ML model is based on NN (Neural Networks).

[0855] As an example, the ML model is based on ANN (Artificial Neural Networks).

[0856] As an example, the ML model is based on CNN (Convolutional Neural Networks).

[0857] As an example, the ML model is based on the LLM (Large Language Model) architecture.

[0858] As an example, the ML model is based on the Transformer architecture.

[0859] As an example, the ML model is based on LSTM (Long Short-Term Memory network).

[0860] As an example, the ML model is based on MLP (MultiLayer Perceptron).

[0861] As an example, the ML model is based on GAN (Generative Adversarial Networks).

[0862] As an example, the ML model is based on a lightweight neural network.

[0863] As a sub-example of this embodiment, the lightweight neural network includes one or more of MobileNet, ShuffleNet, and SqueezeNet.

[0864] Example 14

[0865] Example 14 illustrates a schematic diagram of a processing system based on artificial intelligence or machine learning according to an embodiment of this application, as shown in the attached diagram. Figure 14 As shown. In the appendix Figure 14 In this context, the processing system based on artificial intelligence or machine learning includes a first processor, a second processor, a third processor, and a fourth processor.

[0866] In Example 14, the first processor sends a first dataset to the second processor and a second dataset to the third processor; the second processor generates a target first-class parameter set based on the first dataset, and sends the generated target first-class parameter set to the third processor; the third processor processes the second dataset using the target first-class parameter set to obtain a first-class output, optionally, the third processor sends the first-class output to the fourth processor. (See Appendix...) Figure 14 In this configuration, the first type of feedback and the second type of feedback are optional; the second processor includes ML training functionality; and the third processor includes ML inference functionality.

[0867] As one embodiment, the fourth processor includes ML testing functionality.

[0868] As one embodiment, the fourth processor includes performance monitoring / evaluation of the ML model.

[0869] As an example, the third processor sends a first type of feedback to the second processor; the first type of feedback is used to trigger the recalculation or update of the target first type of parameter set, that is, to trigger ML initial training or ML retraining.

[0870] As one embodiment, the fourth processor sends a second type of feedback to the first processor; the second type of feedback is used to generate the first dataset or the second dataset, or the second type of feedback is used to trigger the sending of the first dataset or the sending of the second dataset.

[0871] As one embodiment, the first processor generates the first dataset and the second dataset based on the measurement of the reference signal.

[0872] As one embodiment, the third processor belongs to the first node, and the fourth processor belongs to the second node.

[0873] As an example, the third processor belongs to the first node.

[0874] As an example, the first dataset includes training data.

[0875] As one embodiment, the second processor is used to train an ML model, and the trained model is described by the target first class of parameter sets.

[0876] As an example, the second processor belongs to the first node; the above method avoids passing the first dataset to the second node.

[0877] As an example, the second processor belongs to the second node in this application; the above method supports joint training and optimizes system performance.

[0878] As an example, the second processor belongs to the core network; the above method supports network-wide joint training, further optimizing system performance.

[0879] As an example, the second dataset includes inference data.

[0880] As an example, the third processor constructs a model based on the target first type of parameter group, and then inputs the second dataset into the constructed model to obtain the first type of output.

[0881] As an example, the third processor generates a recovery dataset based on the first type of output, and the error between the recovery dataset and the second dataset is used to generate the first type of feedback.

[0882] As an example, the first type of feedback is used to reflect the performance of the trained model; when the performance of the trained model fails to meet the requirements, the second processing opportunity will recalculate the target first type of parameter set.

[0883] As an example, when the error is too large or the update has not been performed for too long, the performance of the trained model is considered to be unsatisfactory.

[0884] As an example, the target first type of parameter group includes one or more of the following: convolution kernel, pooling kernel, pooling function, activation function, parameters of the pooling function, or parameters of the activation function.

[0885] As an example, the target first type of parameter group includes one or more of the following: convolution kernel size, number of convolution layers, convolution stride, pooling kernel size, pooling kernel stride, pooling function, activation function, or number of feature maps.

[0886] Example 15

[0887] Example 15 illustrates a schematic diagram based on artificial intelligence or machine learning according to an embodiment of this application, as shown in the attached diagram. Figure 15 As shown. In the appendix Figure 15 In this process, the first and second operations belong to the first stage, the third operation belongs to the second stage, the fourth operation belongs to the third stage, and the fifth operation belongs to the fourth stage; the arrowed lines indicate the sequence of the process.

[0888] As an example, the first operation includes AI / ML training, the second operation includes AI / ML testing, the third operation includes AI / ML emulation, the fourth operation includes AI / ML entity loading, and the fifth operation includes AI / ML inference.

[0889] As one embodiment, the first stage includes a training phase, the second stage includes an emulation phase, the third stage includes a deployment phase, and the fourth stage includes an inference phase.

[0890] As an example, the first stage includes AI / ML model training.

[0891] As an example, the first stage includes AI / ML model training and AI / ML testing.

[0892] As an example, the AI / ML model training includes initial training and re-training of one or a group of AI / ML entities.

[0893] As an example, the training of the AI / ML model depends on training data.

[0894] As an example, the training of the AI / ML model depends on the first node performing the data collection.

[0895] As an example, the AI / ML model training includes AI / ML entity validation.

[0896] As an example, the AI / ML entity verification is used to evaluate the performance of the AI / ML entity.

[0897] As an example, the AI / ML entity verification relies on verification data.

[0898] As an example, if the AI / ML entity verification results do not meet expectations, the AI / ML model will be retrained.

[0899] As an example, the AI / ML testing includes testing the validated AI / ML entities to estimate the performance of the trained AI / ML model.

[0900] As an example, if the AI / ML test results meet expectations, the AI / ML entity proceeds to the next stage; otherwise, the AI / ML model will be retrained.

[0901] As an example, the AI / ML test relies on test data.

[0902] As one embodiment, the second stage includes AI / ML simulation, which performs AI / ML entity reasoning in a simulation environment.

[0903] As an example, the AI / ML simulation estimates the performance of AI / ML entity reasoning in a simulation environment before using AI / ML entities.

[0904] As one embodiment, the second stage is optional.

[0905] As an example, the third stage includes AI / ML entity loading, which is to obtain trained AI / ML entities to obtain the desired AI / ML inference function.

[0906] As an example, the third stage is optional.

[0907] As an example, the third stage is no longer needed when the training and inference functions are co-located.

[0908] As an example, the fourth stage includes AI / ML inference.

[0909] Example 16

[0910] Example 16 illustrates a structural block diagram of a processing apparatus for a first node according to an embodiment of this application, as shown in the attached diagram. Figure 16 As shown. In the appendix Figure 16 In the first node, the processing device 1600 includes a first processor 1601 and a first transmitter 1602, wherein the first transmitter 1602 is optional.

[0911] In embodiment 16, the first processor 1601 receives a first configuration information block, the first configuration information block indicating a first inference configuration; the first processor 1601 receives a first RS resource set, the first RS resource set including one or more RS resources; and performs at least one performance monitoring for the first inference configuration every first time period.

[0912] In Example 16, the performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0913] As one embodiment, the first processor 1601 receives a first information block, the first information block configuring a good inference performance criterion; the first time length depends on whether the good inference performance criterion is met.

[0914] As an example, the performance monitoring results include a first metric, and the good inference performance criterion includes one of the following:

[0915] - The first metric satisfies a first threshold, where the first threshold is predefined or configured by a higher-layer signaling layer;

[0916] - Within a second time length, the first metric satisfies a first threshold, the second time length is not less than the first time length, and the first threshold is predefined or configured by a higher-layer signaling;

[0917] - At least K consecutive times the first metric satisfies the first threshold, where K is a positive integer greater than 1, and the first threshold is predefined or configured by a higher-layer signaling.

[0918] As one embodiment, the first processor 1601 receives a second information block, the second information block being configured with at least the former of the low mobility criterion and the non-cell edge criterion; the first inference is configured for low mobility, and the first time length depends on whether at least the former of the low mobility criterion and the non-cell edge criterion is satisfied.

[0919] As one embodiment, the first transmitter 1602 sends a first message, the first message including the result of the performance monitoring; the first processor 1601 monitors target signaling, the target signaling being MAC layer signaling or physical layer signaling; the target signaling indicates the first time length from multiple time lengths.

[0920] As an example, the model or functionality associated with the first inference configuration is for a higher level.

[0921] As an example, the first time length depends on a first parameter, which indicates the downlink reception period.

[0922] Typically, the first processor 1601 receives the first information block and the second information block, wherein the first time length depends on whether at least two of the following criteria are met: good inference performance criterion, low mobility criterion, and not at cell edge criterion.

[0923] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length. If the first node satisfies the good inference performance criterion and the low mobility criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0924] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length. If the first node satisfies the good inference performance criterion, the low mobility criterion, and the non-cell edge criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0925] As a sub-implementation of the above three embodiments, the first candidate time length is indicated by the sender of the first configuration information block.

[0926] As a sub-implementation of the above three embodiments, the first candidate time length is predefined.

[0927] As a sub-implementation of the above three embodiments, the second candidate time length is indicated by the sender of the first configuration information block.

[0928] As a sub-implementation of the above three embodiments, the second candidate time length is predefined.

[0929] As a sub-implementation of the above three embodiments, the first candidate time length is greater than the second candidate time length.

[0930] As an example, the first node 1600 is a user equipment.

[0931] As an example, the first node 1600 is a terminal.

[0932] As an example, the first node 1600 is a relay node device.

[0933] As an example, the first processor 1601 includes at least one of the following in embodiment 4: the antenna 452, the receiver 454, the receiver processor 456, the multi-antenna receiver processor 458, the controller / processor 459, the memory 460, and the data source 467.

[0934] As an example, the first transmitter 1602 includes at least one of the following in embodiment 4: the antenna 452, the transmitter 454, the transmission processor 468, the multi-antenna transmission processor 457, the controller / processor 459, the memory 460, and the data source 467.

[0935] Example 17

[0936] Example 17 illustrates a structural block diagram of a processing apparatus for a second node according to an embodiment of this application, as shown in the attached diagram. Figure 17 As shown. In the appendix Figure 17 In the second node, the processing device 1700 includes a second transmitter 1701 and a first receiver 1702, wherein the first receiver 1702 is optional.

[0937] In embodiment 17, the second transmitter 1701 sends a first configuration information block, the first configuration information block indicating a first inference configuration; the second transmitter 1701 sends a first RS resource set, the first RS resource set including one or more RS resources.

[0938] In Example 17, the recipient of the first configuration information block performs at least one performance monitoring for the first inference configuration every first time length, the performance monitoring depending on the measurement on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

[0939] As one embodiment, the second transmitter 1701 transmits a first information block, the first information block being configured with a good inference performance criterion; the first time length depends on whether the good inference performance criterion is met.

[0940] As an example, the performance monitoring results include a first metric, and the good inference performance criterion includes one of the following:

[0941] - The first metric satisfies a first threshold, where the first threshold is predefined or configured by a higher-layer signaling layer;

[0942] - Within a second time length, the first metric satisfies a first threshold, the second time length is not less than the first time length, and the first threshold is predefined or configured by a higher-layer signaling;

[0943] - At least K consecutive times the first metric satisfies the first threshold, where K is a positive integer greater than 1, and the first threshold is predefined or configured by a higher-layer signaling.

[0944] As one embodiment, the second transmitter 1701 transmits a second information block configured with at least the former of the low mobility criterion and the non-cell edge criterion; the first inference is configured for low mobility, and the first time length depends on whether at least the former of the low mobility criterion and the non-cell edge criterion is met.

[0945] As one embodiment, the first receiver 1702 receives a first message, the first message including the result of the performance monitoring; the second transmitter 1701 sends target signaling, the target signaling being MAC layer signaling or physical layer signaling; the target signaling indicates the first time length from multiple time lengths.

[0946] As an example, the first receiver 1702 receives a first message, which includes the result of the performance monitoring; the second transmitter 1701 does not send target signaling, which is MAC layer signaling or physical layer signaling; the first time length is a default value.

[0947] As an example, the model or functionality associated with the first inference configuration is for a higher level.

[0948] As an example, the first time length depends on a first parameter, which indicates the downlink reception period.

[0949] Typically, the second transmitter 1701 transmits the first information block and the second information block, the first time length depending on whether at least two of the following criteria are met: good inference performance criterion, low mobility criterion, and not at cell edge criterion.

[0950] Typically, the second transmitter 1701 transmits the first information block and the second information block, the first time length depending on whether at least two of the following criteria are met: good inference performance criterion, low mobility criterion, and not at cell edge criterion.

[0951] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length. If the sender of the first configuration information block satisfies the good inference performance criterion and the low mobility criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0952] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length. If the sender of the first configuration information block satisfies the good inference performance criterion, the low mobility criterion, and the non-cell edge criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0953] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length. If the sender of the first configuration information block satisfies the good inference performance criterion and the low mobility criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0954] As an example, the candidates for the first time length include a first candidate time length and a second candidate time length. If the sender of the first configuration information block satisfies the good inference performance criterion, the low mobility criterion, and the non-cell edge criterion, the first time length is the first candidate time length; otherwise, the first time length is the second candidate time length.

[0955] As a sub-implementation of the above three embodiments, the first candidate time length is indicated by the second node.

[0956] As a sub-implementation of the above three embodiments, the first candidate time length is predefined.

[0957] As a sub-implementation of the above three embodiments, the second candidate time length is indicated by the second node.

[0958] As a sub-implementation of the above three embodiments, the second candidate time length is predefined.

[0959] As a sub-implementation of the above three embodiments, the first candidate time length is greater than the second candidate time length.

[0960] As one example, the second node 1700 is a base station device.

[0961] As one embodiment, the second node 1700 is a user equipment.

[0962] As an example, the second node 1700 is a TRP.

[0963] As one embodiment, the second transmitter 1701 includes at least one of the following in embodiment 4: the antenna 420, the transmitter 416, the transmission processor 416, the multi-antenna transmission processor 471, the controller / processor 475, and the memory 476.

[0964] As an example, the first receiver 1702 includes at least one of the following in embodiment 4: the antenna 420, the receiver 416, the receiver processor 470, the multi-antenna receiver processor 472, the controller / processor 475, and the memory 476.

[0965] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, hard disk, or optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Correspondingly, each module unit in the above embodiments can be implemented in hardware or in the form of software functional modules. This application is not limited to any specific combination of software and hardware. The user equipment, terminal, and UE in this application include, but are not limited to, drones, communication modules on drones, remote-controlled aircraft, aircraft, small aircraft, mobile phones, tablets, laptops, vehicle-mounted communication equipment, vehicles, RSUs, wireless sensors, internet cards, IoT terminals, RFID (Radio Frequency Identification) terminals, NB-IoT (Narrow Band Internet of Things) terminals, MTC (Machine Type Communication) terminals, eMTC (enhanced MTC) terminals, data cards, internet cards, vehicle-mounted communication equipment, low-cost mobile phones, low-cost tablets, and other wireless communication devices. The base stations or system equipment in this application include, but are not limited to, macrocell base stations, microcell base stations, small cell base stations, home base stations, relay base stations, eNB (evolved Node B), gNB, TRP, GNSS (Global Navigation Satellite System), relay satellites, satellite base stations, airborne base stations, RSUs, unmanned aerial vehicles, and test equipment, such as transceivers or signaling testers that simulate some functions of a base station, and other wireless communication equipment.

[0966] Those skilled in the art will understand that the present invention can be practiced in other specified forms without departing from its core or essential characteristics. Therefore, the embodiments disclosed herein should in any way be considered descriptive rather than restrictive. The scope of the invention is defined by the appended claims rather than the foregoing description, and all modifications within their equivalent meaning and scope are considered to be included therein.

Claims

1. A first node used in wireless communication and artificial intelligence, characterized in that, include: A first processor receives a first configuration information block, wherein the first configuration information block indicates a first inference configuration; Receive a first RS resource set, which includes one or more RS resources; perform at least one performance monitoring for the first inference configuration every first time period; The performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

2. The first node according to claim 1, characterized in that, include: The first processor receives a first information block, which is configured with a good inference performance criterion. The first time length depends on whether the reasoning performance criterion is met.

3. The first node according to claim 2, characterized in that, The results of the performance monitoring include a first metric, and the criteria for good inference performance include one of the following: - The first metric satisfies a first threshold, where the first threshold is predefined or configured by a higher-layer signaling layer; - Within a second time length, the first metric satisfies a first threshold, the second time length is not less than the first time length, and the first threshold is predefined or configured by a higher-layer signaling; - At least K consecutive times the first metric satisfies the first threshold, where K is a positive integer greater than 1, and the first threshold is predefined or configured by a higher-layer signaling.

4. The first node according to any one of claims 1 to 3, characterized in that, include: The first processor receives a second information block, the second information block being configured with low mobility criteria and at least the former of not being in the cell edge criteria; Wherein, the first inference is configured for low mobility, and the first time length depends on whether at least the former of the low mobility criterion and the non-cell edge criterion is met.

5. The first node according to any one of claims 1 to 4, characterized in that, include: The first transmitter sends a first message, the first message including the results of the performance monitoring; The first receiver monitors target signaling, which is MAC layer signaling or physical layer signaling. The target signaling indicates the first time length from multiple time lengths.

6. The first node according to any one of claims 1 to 5, characterized in that, The model or functionality associated with the first inference configuration is for a higher level.

7. The first node according to any one of claims 1 to 6, characterized in that, The first time length depends on the first parameter, which indicates the downlink reception period.

8. A second node used in wireless communication and artificial intelligence, characterized in that, include: The second transmitter sends a first configuration information block, which indicates a first inference configuration. Send a first RS resource set, which includes one or more RS resources; Wherein, the receiver of the first configuration information block performs at least one performance monitoring for the first inference configuration every first time length, the performance monitoring depending on the measurement on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

9. A method for use as a first node in wireless communication and artificial intelligence, characterized in that, include: Receive a first configuration information block, the first configuration information block indicating a first inference configuration; Receive a first RS resource set, wherein the first RS resource set includes one or more RS resources; Perform at least one performance check for the first inference configuration every first time period; The performance monitoring depends on measurements on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

10. A method for a second node used in wireless communication and artificial intelligence, characterized in that, include: Send a first configuration information block, which indicates a first inference configuration; Send a first RS resource set, which includes one or more RS resources; Wherein, the receiver of the first configuration information block performs at least one performance monitoring for the first inference configuration every first time length, the performance monitoring depending on the measurement on the first RS resource set and the first inference configuration; the first time length depends on the output of the performance monitoring and the first inference configuration.

Images