Measurement report

Abstract

A method, apparatus, and computer program are described that include obtaining a reporting framework, where the reporting framework maps one or more precision conditions to one or more reporting behaviors and / or one or more reporting behavior changes; obtaining an indication of whether an output satisfies a precision condition of the one or more precision conditions; and determining the reporting behavior and / or the one or more reporting behavior changes based at least in part on the reporting framework and the indication of whether the precision condition of the one or more precision conditions is satisfied by the output.

Description

Technical Field

[0001] This specification relates to apparatus and methods for determining reporting behavior, and apparatus and methods for verifying accuracy conditions. Background Technology

[0002] There remains interest in improving measurement reporting in wireless access networks. Summary of the Invention

[0003] The independent claims define the scope of protection for various embodiments of the invention. Embodiments and features described in this specification that do not fall within the scope of the independent claims (if any) are to be interpreted as examples that aid in understanding the various embodiments of the invention.

[0004] A first aspect provides an apparatus comprising: components for acquiring a reporting frame, wherein the reporting frame maps one or more precision conditions to one or more reporting actions and / or one or more reporting action variations; components for acquiring an indication of whether the output satisfies a precision condition among the one or more precision conditions; and components for determining a reporting action and / or one or more reporting action variations based at least in part on the reporting frame and the indication of whether the precision condition among the one or more precision conditions is satisfied by the output.

[0005] In some example embodiments, one or more precision conditions include at least one or more of the following: a threshold precision where a precision condition above it is satisfied; a threshold precision where a precision condition below it is satisfied; a threshold precision where a precision condition within it is satisfied; and a threshold precision where a precision condition outside it is satisfied.

[0006] In some example embodiments, one or more precisions of one or more precision conditions are defined in at least one or more of the following aspects: the proportion of one or more outputs that match one or more corresponding true values; the mean square error of one or more outputs relative to one or more corresponding true values; and one or more negative logarithms of one or more likelihoods, wherein the likelihood is the likelihood assigned to the output that matches the corresponding true value.

[0007] In some example embodiments, the output includes one or more of the following: the measured predicted value; the target cell; and the target beam.

[0008] In some example embodiments, each of one or more reporting actions includes the use of a corresponding measurement report configuration. Each corresponding measurement report configuration of one or more reporting actions may include at least one or more of the following parameters: reporting criteria; the event used to trigger the reporting; the reporting period / interval; and the number of reports. The corresponding measurement report configuration of one or more reporting actions may include at least one or more accuracy levels at which one or more user equipment measurements are collected. A first reporting action in one or more reporting actions may include a first measurement report configuration including a first accuracy level at which one or more user equipment measurements are collected; a second reporting action in one or more reporting actions may include a second measurement report configuration including a second accuracy level at which one or more user equipment measurements are collected, wherein the first accuracy level is higher than the second accuracy level; a first accuracy condition above a threshold can be satisfied and mapped to the second reporting action; and a second accuracy condition below a threshold can be satisfied and mapped to the first reporting action.

[0009] In some example embodiments, the reporting behavior includes an event-triggered measurement report; and the reporting framework includes one or more measurement report configurations that map one or more accuracy conditions to the triggering of the event-triggered measurement report.

[0010] In some example embodiments, the apparatus further includes: components for collecting measurement data associated with a radio access network signal; components for acquiring a reporting framework including: components for receiving the reporting framework from a network node; and components for acquiring an indication of whether the output meets one or more accuracy conditions including: components for generating the output based on the measurement data acquired by the user equipment, components for determining the accuracy of the generated output, and components for verifying whether the generated output meets the accuracy conditions. Each of the one or more reporting actions may include the use of a corresponding measurement report configuration, and the apparatus may further include: components for sending a request to a network node in response to acquiring an indication that the output meets the accuracy conditions, the request being for using the measurement report configuration included in the reporting action mapped by the reporting framework to the accuracy conditions; components for receiving a response from the network node indicating that the measurement report configuration will be used by the apparatus; and components for applying the indicated measurement report configuration. The accuracy conditions satisfied by the output can be mapped to reporting behavior including a measurement reporting configuration with event-based reporting criteria; and the apparatus may further include: components for verifying whether an event identified in the measurement configuration has occurred; and components for triggering a measurement report based on the verification that the event has occurred. The apparatus may also include: components for sending capability information associated with the user equipment to network nodes; and components for sending accuracy requirements for the output to network nodes.

[0011] In some example embodiments, the apparatus is a network node, and: the apparatus further includes: components for receiving measurement data associated with a radio access network signal from a user equipment; components for obtaining an indication of whether the output meets one or more accuracy conditions, including: components for generating the output based on the measurement data received from the user equipment, components for determining the accuracy of the generated output, and components for verifying whether the generated output meets the accuracy conditions. The apparatus may also include components for sending an instruction to the user equipment to use a measurement report configuration included by reporting behavior mapped to the met accuracy conditions. The apparatus may further include: components for receiving capability information from the user equipment; and components for obtaining a reporting framework may include: components for generating a reporting framework based on the capability information associated with the user equipment and the accuracy requirements for the output.

[0012] In some example embodiments, the apparatus is a network node, and wherein: the component for obtaining the reporting framework includes: a component for generating the reporting framework based on capability information associated with the user equipment and the accuracy requirements of the user equipment's output; the apparatus includes: a component for sending one or more accuracy conditions to the user equipment; the apparatus includes: a component for receiving from the user equipment an indication of whether the output meets one or more accuracy conditions; and the apparatus includes: a component for sending an instruction to the user equipment for using a measurement report configuration included by reporting behavior mapped to the met accuracy conditions.

[0013] The second aspect provides a method comprising: acquiring a reporting framework, wherein the reporting framework maps one or more precision conditions to one or more reporting behaviors and / or one or more reporting behavior variations; acquiring an indication of whether the output satisfies a precision condition among the one or more precision conditions; and determining a reporting behavior and / or one or more reporting behavior variations based at least in part on the reporting framework and the indication of whether the precision condition among the one or more precision conditions is satisfied by the output.

[0014] In some example embodiments, one or more precision conditions include at least one or more of the following: a threshold precision where a precision condition above it is satisfied; a threshold precision where a precision condition below it is satisfied; a threshold precision where a precision condition within it is satisfied; and a threshold precision where a precision condition outside it is satisfied.

[0015] In some example embodiments, one or more precisions of one or more precision conditions are defined in at least one or more of the following aspects: the proportion of one or more outputs that match one or more corresponding true values; the mean square error of one or more outputs relative to one or more corresponding true values; and one or more negative logarithms of one or more likelihoods, wherein the likelihood is the likelihood assigned to the output that matches the corresponding true value.

[0016] In some example embodiments, the output includes one or more of the following: the measured predicted value; the target cell; and the target beam.

[0017] In some example embodiments, each of one or more reporting actions includes the use of a corresponding measurement report configuration. Each corresponding measurement report configuration of one or more reporting actions may include at least one or more of the following parameters: reporting criteria; the event used to trigger the reporting; the reporting period / interval; and the number of reports. The corresponding measurement report configuration of one or more reporting actions may include at least one or more accuracy levels at which one or more user equipment measurements are collected. A first reporting action in one or more reporting actions may include a first measurement report configuration including a first accuracy level at which one or more user equipment measurements are collected; a second reporting action in one or more reporting actions may include a second measurement report configuration including a second accuracy level at which one or more user equipment measurements are collected, wherein the first accuracy level is higher than the second accuracy level; a first accuracy condition above a threshold can be satisfied and mapped to the second reporting action; and a second accuracy condition below a threshold can be satisfied and mapped to the first reporting action.

[0018] In some example embodiments, the reporting behavior includes an event-triggered measurement report; and the reporting framework includes one or more measurement report configurations that map one or more accuracy conditions to the triggering of the event-triggered measurement report.

[0019] In some example embodiments, the method further includes: collecting measurement data associated with a radio access network signal; obtaining a reporting framework including receiving the reporting framework from a network node; and obtaining an indication of whether the output meets one or more accuracy conditions including: generating the output based on the measurement data obtained by the user equipment, determining the accuracy of the generated output, and verifying that the generated output meets the accuracy conditions. Each of the one or more reporting actions may include the use of a corresponding measurement report configuration, and the method may further include: in response to obtaining an indication that the output meets the accuracy conditions, sending a request to the network node for using the measurement report configuration included in the reporting action mapped by the reporting framework to the accuracy conditions; receiving a response from the network node indicating that the measurement report configuration will be used by the device; and applying the indicated measurement report configuration. The accuracy conditions met by the output may be mapped to a reporting action including a measurement report configuration with an event-based reporting standard; and the method may further include: verifying whether an event identified in the measurement configuration has occurred; and triggering a measurement report based on verifying that the event has occurred. The method may further include: sending capability information associated with the user equipment to the network node; and sending an accuracy requirement for the output to the network node.

[0020] In some example embodiments, the method further includes: receiving measurement data associated with a radio access network signal from a user equipment; obtaining an indication of whether the output meets one or more accuracy conditions includes: generating the output based on the measurement data received from the user equipment, determining the accuracy of the generated output, and verifying whether the generated output meets the accuracy conditions. The method may also include: sending an instruction to the user equipment to use a measurement report configuration included by reporting behavior mapped to the met accuracy conditions. The method may further include: receiving capability information from the user equipment; and obtaining the reporting framework may include: generating a reporting framework based on the capability information associated with the user equipment and the accuracy requirements for the output.

[0021] In some example embodiments, obtaining a reporting framework includes: generating a reporting framework based on capability information associated with a user equipment and the accuracy requirements of the user equipment's output; the method includes: sending one or more accuracy conditions to the user equipment; the method includes: receiving from the user equipment an indication of whether the output meets one or more accuracy conditions; and the method includes: sending an instruction to the user equipment to use a measurement report configuration included by reporting behavior mapped to the met accuracy conditions.

[0022] A third aspect provides a computer program including a set of instructions that, when executed on a device, are configured to cause the device to perform a method comprising: acquiring a reporting frame, wherein the reporting frame maps one or more precision conditions to one or more reporting actions and / or one or more reporting action variations; acquiring an indication of whether an output satisfies a precision condition among the one or more precision conditions; and determining a reporting action and / or one or more reporting action variations based at least in part on the reporting frame and the indication of whether a precision condition among the one or more precision conditions is satisfied by the output.

[0023] In some example embodiments, the third aspect may include any other features mentioned regarding the method of the second aspect.

[0024] A fourth aspect of the present invention provides a non-transitory computer-readable medium having computer-readable code stored thereon, which, when executed by at least one processor, causes the at least one processor to perform a method comprising: acquiring a reporting frame, wherein the reporting frame maps one or more precision conditions to one or more reporting actions and / or one or more reporting action variations; acquiring an indication of whether an output satisfies a precision condition among the one or more precision conditions; and determining a reporting action and / or one or more reporting action variations based at least in part on the reporting frame and the indication of whether a precision condition among the one or more precision conditions is satisfied by the output.

[0025] The fourth aspect may include any other features mentioned regarding the methods of the second aspect.

[0026] A fifth aspect of the invention provides an apparatus having at least one processor and at least one memory storing computer-readable code thereon, which, when executed, controls the at least one processor to: acquire a reporting frame, wherein the reporting frame maps one or more precision conditions to one or more reporting actions and / or one or more reporting action variations; acquire an indication of whether the output satisfies a precision condition among the one or more precision conditions; and determine a reporting action and / or one or more reporting action variations based at least in part on the reporting frame and the indication of whether the precision condition among the one or more precision conditions is satisfied by the output.

[0027] The fifth aspect may include any other features mentioned regarding the methods of the second aspect.

[0028] A sixth aspect provides an apparatus comprising: components for acquiring measurement data of a user equipment; components for processing the measurement data of the user equipment to determine an output; components for determining an indication of the accuracy of the output; and components for comparing the indicated accuracy with one or more accuracy conditions.

[0029] In some example embodiments, the determined indication of the accuracy of the output and one or more accuracy conditions are defined in at least one or more of the following aspects: the proportion of the determined output to match the true value; the mean square error of the determined output compared to the true value; and the negative log-likelihood.

[0030] In some example embodiments, the apparatus further includes components for determining a truth value, wherein the components for determining an indication of the accuracy of the output are configured to: determine the indication of accuracy based at least in part on the output and the determined truth value, and wherein the components for determining the truth value are configured to determine the truth value based at least in part on one or more of the following: second measurement data, wherein the second measurement data is more accurate than user equipment measurement data; a processing method that is computationally more intensive than a processing method used by the components for processing user equipment measurement data to determine the output; third measurement data, wherein the third measurement data includes more data than user equipment measurement data; and fourth measurement data, wherein the fourth measurement data includes data collected after the user equipment measurement data is collected.

[0031] In some example embodiments, one or more precision conditions include at least one or more of the following: a threshold precision for which a precision condition above is satisfied, a threshold precision for which a precision condition below is satisfied, a threshold precision for which a precision condition within is satisfied, and a threshold precision for which a precision condition outside is satisfied.

[0032] In some example embodiments, the user equipment measurement data includes measurement data indicating one or more of the following: reference signal received power; reference signal received quality; received signal strength indicator; signal-to-interference-plus-noise ratio; and signal-to-noise ratio.

[0033] In some example embodiments, the output includes one or more of the following: the target cell; and the target beam. In some example embodiments, the output includes one or more measured predicted values.

[0034] In some example embodiments, the device is a user equipment, and the device further includes components for collecting user equipment measurement data. The components for collecting user equipment measurement data can be configured to collect data at different levels of accuracy.

[0035] In some example embodiments, the device is a network node, and the device includes components for receiving user equipment measurement data from the user equipment.

[0036] The seventh aspect provides a method comprising: acquiring user equipment measurement data; processing the user equipment measurement data to determine an output; determining an indication of the accuracy of the output; and comparing the indicated accuracy with one or more accuracy conditions.

[0037] In some example embodiments, the determined indication of the accuracy of the output and one or more accuracy conditions are defined in at least one or more of the following aspects: the proportion of the determined output to match the true value; the mean square error of the determined output compared to the true value; and the negative log-likelihood.

[0038] In some example embodiments, the method further includes determining a truth value, wherein determining the indication of the accuracy of the output includes: determining the indication of accuracy based at least in part on the output and the determined truth value, and wherein determining the truth value includes determining the truth value based at least in part on one or more of the following: second measurement data, wherein the second measurement data is more accurate than user equipment measurement data; a processing method that is computationally more intensive than the processing method used in the step of processing user equipment measurement data to determine the output; third measurement data, wherein the third measurement data includes more data than user equipment measurement data; and fourth measurement data, wherein the fourth measurement data includes data collected after the user equipment measurement data is collected.

[0039] In some example embodiments, one or more precision conditions include at least one or more of the following: a threshold precision for which a precision condition above is satisfied, a threshold precision for which a precision condition below is satisfied, a threshold precision for which a precision condition within is satisfied, and a threshold precision for which a precision condition outside is satisfied.

[0040] In some example embodiments, the user equipment measurement data includes measurement data indicating one or more of the following: reference signal received power; reference signal received quality; received signal strength indicator; signal-to-interference-plus-noise ratio; and signal-to-noise ratio.

[0041] In some example embodiments, the output includes one or more of the following: the target cell; and the target beam. In some example embodiments, the output includes one or more measured predicted values.

[0042] In some example embodiments, the method further includes collecting user equipment measurement data. Collecting user equipment measurement data may include collecting data at different levels of accuracy.

[0043] In some example embodiments, the method further includes receiving user equipment measurement data from the user equipment.

[0044] The eighth aspect provides a computer program including a set of instructions that, when executed on a device, are configured to cause the device to perform a method comprising: acquiring user equipment measurement data; processing the user equipment measurement data to determine an output; determining an indication of the accuracy of the output; and comparing the indicated accuracy with one or more accuracy conditions.

[0045] In some example embodiments, the eighth aspect may include any other features mentioned in relation to the method of the seventh aspect.

[0046] A ninth aspect of the present invention provides a non-transitory computer-readable medium having computer-readable code stored thereon, which, when executed by at least one processor, causes the at least one processor to perform a method comprising: acquiring user equipment measurement data; processing the user equipment measurement data to determine an output; determining an indication of the accuracy of the output; and comparing the indicated accuracy with one or more accuracy conditions.

[0047] The ninth aspect may include any other features mentioned in relation to the method of the seventh aspect.

[0048] A tenth aspect of the present invention provides an apparatus having at least one processor and at least one memory having computer-readable code stored thereon, the computer-readable code, when executed, controlling the at least one processor to: acquire user equipment measurement data; process the user equipment measurement data to determine an output; determine an indication of the accuracy of the output; and compare the indicated accuracy with one or more accuracy conditions.

[0049] The tenth aspect may include any other features mentioned regarding the method of the seventh aspect. Attached Figure Description

[0050] Exemplary embodiments will now be described by way of non-limiting example with reference to the accompanying drawings, in which:

[0051] Figure 1 This is a schematic diagram illustrating the RRC message structure of a measurement configuration according to an example embodiment;

[0052] Figure 2 This is a diagram illustrating a message flow sequence according to an example embodiment;

[0053] Figure 3 This is a high-level diagram illustrating how precision-based event entry and exit conditions are used in event triggering reporting according to an example embodiment;

[0054] Figures 4-8 This is a diagram illustrating a message flow sequence according to an example embodiment;

[0055] Figure 9 This is a block diagram illustrating an apparatus according to an example embodiment.

[0056] Figure 10 and Figure 11 This is a flowchart illustrating a method according to an example embodiment;

[0057] Figure 12 This is a diagram schematically illustrating the implementation of an ML time series prediction model according to an example embodiment.

[0058] Figure 13 This is a block diagram of the components of a system according to an example embodiment; and

[0059] Figure 14 An example of a tangible medium for storing computer-readable code is shown, which, when run by a computer, can perform methods according to the above example embodiments. Detailed Implementation

[0060] In a cellular radio access network (RAN), a user equipment (UE) can be served by a cell. A UE may wish to be served by a different cell, and therefore the network can perform a handover procedure to change the cell serving the UE.

[0061] The determination to perform a handover to another cell can be made based on measurements at the RAN network node. All or part of these measurements can be performed at the user equipment, and these measurements can be transmitted to the network node in one or more measurement reports.

[0062] Therefore, appropriate measurements performed at the UE and the appropriate information sent to the network nodes are of interest in enabling appropriate handover decisions. Other decisions by the user equipment and / or the network may also rely on appropriate UE measurements and measurement reports.

[0063] UEs and networks can take a "static" approach to perform and report measurements, and make handover decisions based on these reports. Under this static approach, high-fidelity measurements may be required to ensure accurate handover decisions. Measurement configuration

[0064] The measurements performed by the user equipment and how these measurements are transmitted to the RAN node can be specified in the measurement configuration stored by the UE.

[0065] The UE can be provided with measurement configuration via Radio Resource Control (RRC) messages provided by the network node.

[0066] Figure 1 This is a schematic diagram illustrating the RRC message structure of measurement configuration 100.

[0067] Measurement configuration 100 includes indications for one or more measurement objects 110, one or more reporting configurations 112, indications for one or more measurement identities 114, one or more quantity configurations 116, and indications for one or more measurement gaps 118.

[0068] Measurement object 110 is the entity on which the UE performs measurements. Measurement objects can vary depending on the measurement type (such as intra-frequency, inter-frequency, and RAT (Radio Access Technology) measurements). Measurement objects may include details such as the target cell frequency, frequency or time location, target reference signal, cell-specific offset, and blacklist / whitelist cell lists. Measurement objects may include object IDs that link the measurement object to one or more serving cells.

[0069] Report configuration 112 defines how the UE reports measurements. The reporting configuration may include criteria for triggering a report (e.g., an event that triggers a measurement report in the case of event-based reporting). The reporting configuration may include an indication of the reference signal (RS) type used for the measurement. For example, the reference signal may be a synchronization signal (SSB) / physical broadcast channel (PBCH) signal, or the reference signal may be a channel state information (CSI) reference signal (CSI-RS). The reporting configuration may include a reporting format. Conditional reconfiguration includes additional criteria and reference signal types to be performed.

[0070] One or more measurement identifiers 114 are a list that links measurement objects to reporting configurations. Multiple measurement identifiers can be linked to a reporting configuration, and the measurement identifiers are included in the reports sent to the network.

[0071] One or more measurement configurations 116 are configurations that determine the filtering applied to measurements during event assessment, reporting, and periodic reporting. They can have different filtering coefficients for various measurement quantities and RS types.

[0072] One or more measurement gaps 118 are time periods that the UE can use to perform specified measurements.

[0073] For each measurement object, one or more reporting configurations can be defined. Each measurement object can have one or more reporting configurations. Each reporting configuration can include one or more of the following: reporting criteria, RS type, and report format.

[0074] When the reporting criteria are determined to be met, this can trigger the UE to send a measurement report. The reporting criteria can be periodic (i.e., configured to cause the UE to send measurement reports periodically) or event-triggered (e.g., triggered by a single event).

[0075] RS type refers to the reference signal (SSB or CSI-RS) used by the UE for beam and cell measurement results.

[0076] The report format includes the quantities for each cell and each beam that the UE includes in the measurement report (e.g., RSRP), and may include other relevant information such as the maximum number of cells to be reported and the maximum number of beams per cell. L1 CSI reporting

[0077] CSI reporting can be performed at the physical layer (L1).

[0078] In physical layer CSI reporting, the UE can calculate CSI parameters (if reported), assuming the following dependencies exist between the CSI parameters (if reported): - LI can be calculated based on the reported CQI, PMI, RI, and CRI; - CQI can be calculated based on the reported PMI, RI, and CRI; - PMI can be calculated based on the reported RI and CRI; and - RI can be calculated based on the reported CRI.

[0079] CSI reporting configurations can be aperiodic and can use the Physical Uplink Shared Channel (PUSCH) to send reports. Reporting configurations can also be periodic and can use the Physical Uplink Control Channel (PUCCH) to send reports. Reporting configurations can be semi-persistent and can use PUCCH and a PUSCH activated by Downlink Control Information (DCI) to send reports. CSI-RS resources can be periodic, semi-persistent, or aperiodic. Table 1 below illustrates the combinations of CSI reporting configurations and CSI-RS resource configurations, and how to trigger (or activate) CSI reporting for each CSI-RS resource configuration. Periodic CSI-RS can be configured by higher layers. Table 1: Triggers for Potential CSI Reporting Types in CSI-RS Configuration Lifecycle Management

[0080] UE measurements can be used for radio resource management (RRM). UE RRM measurements can be input data for algorithms that can implement machine learning (ML) models.

[0081] These algorithms can be implemented using models designed to predict RRM measurements and mobility-related events.

[0082] ML models are trained models, and the “lifecycle” of such models can be managed during lifecycle management (LCM).

[0083] LCM can include the following aspects: - Data collection (which may also include relevant supporting information, if applicable); - Model training; - Function / model identifier; - Model transfer; - Model inference operations; - Function / model selection, activation, deactivation, switching, and rollback operations (which may include decisions made by the network (network-initiated or UE-initiated and requested by the network), and / or decisions made by the UE (triggered by network-configured events, UE decisions reported to the network, or UE autonomous decisions with or without UE decisions reported to the network). - Functionality / model monitoring; - Model updates; and - UE capabilities

[0084] ML models can have model IDs with associated information. ML models can also provide a given functionality.

[0085] Different models may have different requirements, such as measurement accuracy and prediction accuracy. When performing LCM procedures on the UE side and / or network side, a flexible (measurement) reporting mechanism can adapt to different ML requirements.

[0086] When deploying ML solutions for mobility (handover decisions), one or both of the network (NW) and user equipment (UE) can be equipped with AI-driven (or ML-driven) algorithms. The accuracy of predictions made by such algorithms can depend on the nature of the model inputs. For example, an algorithm for predicting inter-cell beam-level measurements for network layer (L3) mobility may require high-fidelity measurements as input to ensure high reliability of the predictions (i.e., target cell and / or target beam).

[0087] Changing the measurement accuracy of a device can be challenging for several reasons: • Achieving high-fidelity measurements may introduce complexity; • Performing high-precision measurements may require higher processing overhead, higher power consumption, activation of additional RF components, etc. • Reporting measurements with different levels of accuracy may not meet current requirements, such as the accuracy levels reported by RSRP; and • For AI / ML-based use cases, such as beam management or mobility, the UE can support different levels of prediction accuracy depending on the conditions and scenario. A static measurement reporting process may not capture the required level of AI / ML-based prediction accuracy for a given use case.

[0088] Furthermore, there are different radio conditions and scenarios where higher precision measurements may or may not be required, for example: • When the UE is within good coverage of a sufficient number of beams / cells, such as in line-of-sight (LoS) conditions, high-precision measurements may not be necessary, and predictions can be sufficiently reliable. However, in strong non-line-of-sight (NLoS) conditions, or when many beams are blocked, improved measurement accuracy can help provide reliable predictions, despite inference. • Higher accuracy measurements may be required when the UE is in a high mobility state, at the cell edge, or in a non-stationary state. Different aspects of LCM may require different levels of measurement accuracy. In one example, data collection for model training may require high-fidelity measurements. High-fidelity measurements may also be needed in ground truth data collection for model monitoring (e.g., monitoring the accuracy of predictions made by the model). • During monitoring, degradation in inference can be observed based on default / low-precision measurements. For example, it can be observed that low-precision measurements lead to sufficiently accurate model predictions / outputs. • Performing high-fidelity measurements and / or reporting high-fidelity measurement data can provide data that can be used as truth values ​​for monitoring purposes at the UE and / or NW side. • It is assumed that different AI / ML models / model inputs may be required or used when different measurement accuracies are used.

[0089] Therefore, mechanisms for identifying the ability to measure and / or report measurements at different levels of precision or accuracy, and for reporting, switching, and testing using different levels of measurement precision and / or accuracy, can be useful. These mechanisms can be useful in AI / ML use cases.

[0090] Dynamic measurement and reporting mechanisms based on various (e.g., two or more) levels of measurement accuracy (which can be useful for AI / ML use cases) can include one or more of the following aspects. UE-side implementation

[0091] In UE-side embodiments, the UE manages an algorithm that generates output using measurement data collected by the UE. In some example embodiments, this algorithm may be an AI / ML-based algorithm. The UE monitors the accuracy requirements of the algorithm's output. The algorithm's output may be part of the content of a measurement report sent by the UE to the network, or the output may be used to determine the content of that measurement report.

[0092] In this embodiment, the UE can send capability information to a network node (such as a gNB) to indicate the reporting mechanisms and parameters it supports, and the network node can provide the UE with corresponding reporting configurations to perform a dynamic reporting framework based on accuracy requirements (which may be the accuracy requirements of the output).

[0093] Reporting configurations can include those using periodic, event-triggered, or semi-persistent / non-periodic reporting mechanisms. The selected reporting configuration can be based on the indicated supported reporting mechanism. NW side implementation

[0094] In NW-side embodiments, the NW (and in some examples, specific network nodes) hosts an algorithm that generates output using measurement data collected by the UE. In some example embodiments, this algorithm may be an AI / ML-based algorithm. The NW monitors the accuracy requirements of the algorithm's output and may instruct the UE to report measurements with a specific accuracy or level of accuracy. In this embodiment, based on one or more accuracy requirements on the NW side, the NW can provide the UE with the desired configuration for executing this dynamic reporting framework.

[0095] Similarly, reporting configurations can include those using periodic, event-triggered, or semi-persistent / non-periodic reporting mechanisms. The selected reporting configuration can be based on the indicated supported reporting mechanism.

[0096] One or both of the above embodiments may use one or more configuration messages to provide dynamic precision-based reporting, the configuration message including one or more of the following aspects: • UE capability information, which may include reporting mechanisms supported by the UE and can be sent from the UE to the network. • Dynamic, precision-based reporting mechanism RRC configuration parameters, which may include reporting criteria, number of reports, intervals, etc., can be sent from the network to the UE. • Accuracy verification conditions and thresholds for UE and / or NW to evaluate accuracy requirements. In periodic and semi-persistent reporting, accuracy verification conditions and / or thresholds can be conditions used to change between measurement and / or reporting configurations with different measurement and / or reporting accuracy levels. • One or more time windows and / or durations for evaluating constraints for the UE and / or NW (e.g., accuracy verification conditions and thresholds). • A precision-based event configuration for the UE and / or NW to perform event-triggered reporting. This configuration may include event start and end conditions and associated thresholds.

[0097] Figure 2 This is a diagram illustrating message flow sequence 200. Message flow sequence 200 occurs between UE 202 and network node 204. In some embodiments, network node 204 may be a gNB.

[0098] Message flow sequence 200 corresponds to an embodiment of an algorithm that determines the output on the UE side. This algorithm can be an ML algorithm. The accuracy of the output is determined and verified against accuracy conditions. Based on this verification, the reporting configuration used by UE 202 can be changed. Message flow sequence 200 depicts changing the reporting configuration to a reporting configuration with a periodic reporting type.

[0099] At step 210, network node 204 sends a capability query to UE 202 to perform the dynamic measurement reporting process. The query can be sent via RRC signaling.

[0100] At step 212, UE 202 sends capability information to network node 204. This capability information may be sent along with the measurement accuracy requirements for the output determined at UE 202. This message may be sent via RRC signaling. The supported reporting framework (and / or the supported reporting configuration) and corresponding parameters (such as the supported reporting method type, reporting period, number of reports, etc.) may be included in this message as capability information.

[0101] At step 214, network node 204 prepares configuration parameters based on capability information to create a reporting framework. This reporting framework can be prepared for use by UE 202 to determine outputs using AI / ML, such as RRM measurement predictions (e.g., for L3 mobility).

[0102] At step 216, network node 204 sends a (re)configuration message with configuration parameters to UE 202 to support the new reporting framework (e.g., report type, period, number of reports, etc.) and other associated parameters (e.g., time window for output / prediction accuracy verification, measurement accuracy limits or thresholds for switching report configurations, etc.). In this step, other AI / ML-specific hyperparameters and time steps may also be transmitted to UE 202. This message may be an RRC message.

[0103] At step 218, UE 202 performs a measurement procedure (which may be an L1 and / or L3 measurement procedure) and collects measurement data for determining the output (in some examples, determining the output may include using AI / ML functions to determine the output).

[0104] At step 220, UE 202 determines the output. This output can be determined using the measurement data collected at step 218, and additionally or alternatively, more broadly using measurement data from UE 202. The output can be determined at least in part based on an ML solution. By generating the output using measurement data as input, UE 202 can perform AI / ML-based functions. In UE-side embodiments, AI / ML LCM processes (e.g., training, inference, monitoring, etc.) can be performed on the UE side. In one example use case, a time-series Long Short-Term Memory (LSTM) prediction model can be used to predict RRM measurements or sequences of measurement events for mobility. An example method for determining the output (in this example, RRM measurement prediction for L3 mobility) using an AI / ML model, including model layout, input, and output structures, will be described in more detail later in this specification.

[0105] At step 222, UE 202 verifies the accuracy verification conditions. This can be performed within the network configuration evaluation time window (which may be the time window configured in step 216). Verifying the accuracy verification conditions may include determining the accuracy of the output and comparing it with the accuracy verification conditions.

[0106] Accuracy verification conditions may include a threshold accuracy, and if the predicted measurement accuracy level is relatively low (compared to the threshold), UE 202 may switch to a periodic reporting configuration (which may be configured in step 216). For example, if the periodic reporting configuration is expected to provide higher accuracy than the current reporting configuration (e.g., because the periodic reporting configuration includes more frequent or more accurate measurements), and the current reporting configuration may be a conventional or default reporting configuration, the periodic reporting configuration may be configured to be used below the accuracy threshold. In another example, if the periodic reporting configuration is expected to provide lower accuracy than the current reporting configuration (e.g., because the periodic reporting configuration includes less frequent or less accurate measurements, which may consume fewer resources), and the current reporting configuration may be a conventional or default reporting configuration, the periodic reporting configuration may be configured to be used above the accuracy threshold.

[0107] There are several potential methods for formulating accuracy-based validation conditions. Three example accuracy formulas are shown below, which can be compared with a threshold in the accuracy validation step. The evaluation window size is represented as... L eval The validation dataset collected within this window is represented as D eval The validation dataset may include one or more outputs of step 220. For example, the validation dataset may include a measurement prediction set.

[0108] Option 1: Output precision f E (Compared to the true value) can be expressed as the percentage or score of the validation dataset that matches the true value. This can be done using an equation. To calculate, where It is the output data. This is the first time to verify the data. i The true value of each sample. It is an index function. When the output is a prediction of a measurement or another quantity that can be derived from one or more measurements, the true value can be that measurement or quantity (i.e., the true value can be determined based on post-hoc analysis).

[0109] Option 2: Output precision f E This can be expressed as mean squared error (where the error is the difference between the output and the corresponding true value). This can be expressed using an equation. To calculate, where It is a prediction. It is the first i The true value of the sample.

[0110] Option 3: Output Precision f E This can be represented as the negative log-likelihood. In some examples, the output may include confidence scores assigned to the prediction or determination (e.g., if the output is a beam or cell, or a predicted beam or cell, confidence scores can be assigned to the beam or cell). The negative log-likelihood can be calculated based on the negative logarithm of the confidence scores assigned to the prediction or determination that match the corresponding true value (which is not necessarily the output with the highest score). This can be expressed using an equation. To calculate, where It is the first i A prediction or a certain confidence score. For example, The precise probability value can be obtained from the output of the softmax activation function in the output layer of the ML output. Indicates the model accuracy.

[0111] In this example, where the output (which could be an ML prediction) is determined at UE 202, the truth / correct label is available on the UE side for each output. When this process is performed on the UE side, this information can be used for accuracy assessment.

[0112] Several methods exist for determining the truth value used in the accuracy validation step. The truth value can be based on information collected after the data used to determine the validation dataset has been collected (i.e., based on post-hoc analysis). One example is using a later measurement as the truth value for the prediction of that measurement. Additionally or alternatively, the truth value can be determined using a larger dataset than the dataset used to determine the validation data (e.g., data collected more frequently or over a longer period compared to the data used to determine the validation data). Additionally or alternatively, the truth value can be determined using a more computationally intensive method than the method used to determine the validation data (e.g., using a method that would typically not be feasible within the time allowed for normal output determination). Additionally or alternatively, the truth value can be determined using more accurate data than the dataset used to determine the validation data (e.g., collecting more accurate data may require additional resources).

[0113] In other examples, the truth value can be obtained from an external source. For instance, the dataset from which the validation data is determined can be determined through a test / validation setup, and the truth value can correspond to the "correct" output based on that test / validation setup, which is then used to validate accuracy.

[0114] In the example of implementing the accuracy-based verification conditions described above, the NW (or NW node 204 of the NW) configuration represents a threshold for acceptable prediction accuracy error. ε th .

[0115] if f E < ε th (Right now, f E < ε th If true, the verification phase is set to qualified, and the UE should execute the new periodic reporting framework.

[0116] if f E ≥ ε th (Right now, f E < ε th If the result is false, the verification phase is set to unqualified, and the UE should not execute the new periodic reporting framework. The traditional reporting framework with high-fidelity reporting will still be used.

[0117] In this example, the new reporting framework can include reports with lower fidelity than those provided by the traditional, default, or current reporting frameworks.

[0118] At step 224, UE 202 sends a request or indication to network node 204 to switch to the periodic reporting process. After the accuracy condition is verified at step 222, this request can be a request to switch to the periodic reporting process (i.e., the accuracy condition may not be verified at this stage, and the switch may be conditional upon verification). In some embodiments, the message can simply convey the evaluation result, which can be encoded in a bitmap.

[0119] In some embodiments, the message may include one bit of information. Based on the accuracy verification against a threshold, UE 202 may send a message 1 or 0 to indicate whether the accuracy verification condition is true.

[0120] In some embodiments, the message may include n-bit information. When multiple precision thresholds exist (which may correspond to multiple precision conditions and multiple potential reporting configurations), a general n-bit sequence indication can be sent from UE 202.

[0121] The indication message in step 224 can be sent via L1 or L2 signaling channels, for example, via MAC control element (MAC CE) or via uplink control information (UCI).

[0122] At step 226, network node 204 may evaluate the message from step 224 and prepare the corresponding preferred reporting configuration for UE 202 accordingly. In some embodiments, this step may be omitted.

[0123] At step 228, network node 204 sends a response to UE 202 instructing UE 202 to switch to the periodic reporting procedure as requested in step 224. If the request is for switching to the periodic reporting procedure after accuracy conditions have been verified, the switching instruction may be conditional upon verification of accuracy conditions. Periodic reporting of the UE 202's preferred configuration is authorized.

[0124] Network node 204 can send a 1-bit message to authorize UE 202 to perform a periodically reported request, or to reject the request.

[0125] Similarly, L1 / L2 signaling channels can be used for this indication message, such as MAC CE or DCI.

[0126] At step 230, UE 202 applies the new reporting configuration and sends a measurement report (which may have different or relatively low accuracy) to network node 204.

[0127] At step 232, network node 204 may cache the reported configuration used by the UE. This enables network node 204 to evaluate whether the current method is beneficial to continue or revert to the legacy configuration. In some embodiments, this step may be omitted.

[0128] Figure 3 Figure 300 illustrates at a higher level how precision-based event entry and exit conditions are used in event-triggered reporting. As prediction precision decreases over time, it may fall below the event entry threshold 310. A precision verification step can determine that the precision has fallen below the event entry threshold 310 and trigger event-triggered reporting 312. After the precision increases to above the event exit threshold 314, a precision verification step can determine that the precision has increased to above the exit threshold 314 and stop event-triggered reporting 312.

[0129] Figure 4 Message flow sequence 400 is shown. Message flow sequence 400 occurs between UE 202 and network node 204. In some embodiments, network node 204 may be a gNB.

[0130] Message stream sequence 400 corresponds to an embodiment of an algorithm executed on the UE side to determine the output. This algorithm can be an ML algorithm. The accuracy of the output is determined and verified against accuracy conditions. Based on this verification, event-based measurement reporting can be triggered.

[0131] At step 410, network node 204 sends a capability query to UE 202 to perform the dynamic measurement reporting process. The query can be sent via RRC signaling.

[0132] At step 412, similar to step 212, UE 202 sends capability information to network node 204. This capability information may be sent along with the measurement accuracy requirements for the output determined at UE 202. This message may be sent via RRC signaling. Supported reporting frameworks and corresponding parameters (such as supported reporting method types, reporting cycles, reporting quantities, etc.) may be included in this message as capability information.

[0133] In step 414, similar to step 214, network node 204 prepares configuration parameters based on capability information to create a reporting framework. This reporting framework can be prepared for use by UE 202 to determine outputs using AI / ML, such as RRM measurement predictions (e.g., for L3 mobility).

[0134] At step 416, similar to step 216, network node 204 sends a (re)configuration message with configuration parameters to UE 202 to support the new reporting framework, such as report type, period, number of reports, and other associated parameters (e.g., time window for output / prediction accuracy verification, measurement accuracy limits or thresholds for switching report configurations, etc.). In this step, other AI / ML-specific hyperparameters and time steps may also be transmitted to UE 202. This message may be an RRC message.

[0135] In this scenario, the reporting framework can include a reporting configuration that includes accuracy-based, event-triggered measurement reports. These events can be defined in the reporting configuration and can include event start conditions, event exit conditions, and other associated configuration parameters. For example, two corresponding thresholds can be configured to control event start and event exit.

[0136] At step 418, similar to step 218, UE 202 performs a measurement procedure (which may be an L1 and / or L3 measurement procedure) and collects measurement data to determine the output (in some examples, determining the output may include using AI / ML functions to determine the output mentioned above).

[0137] At step 420, similar to step 220, UE 202 determines the output. This output can be determined using the measurement data collected at step 218, and additionally or alternatively, more broadly using the measurement data of UE 202. The output can be determined at least partially based on an ML solution. By generating the output using measurement data as input, UE 202 can perform AI / ML-based functions. In UE-side embodiments, AI / ML LCM processes (e.g., training, inference, monitoring, etc.) can be performed on the UE side. In one example use case, a time-series Long Short-Term Memory (LSTM) prediction model can be used to predict RRM measurements of mobility or sequences of measurement events. An example method for determining the output (in this example, RRM measurement prediction for L3 mobility) using an AI / ML model, including model layout, input, and output structures, will be described in more detail later in this specification.

[0138] At step 422, UE 202 verifies the accuracy verification condition. This can be performed within the evaluation time window configured in the network (which may be the time window configured at step 416). Verifying the accuracy verification condition may include determining the accuracy of the output and comparing it with the accuracy verification condition. If the accuracy condition is verified, this may trigger measurement reporting (where the accuracy condition corresponds to an event entry trigger) or may terminate measurement reporting (where the accuracy condition corresponds to an event exit trigger).

[0139] For precision-based events, the precision verification methods and conditions can correspond to those described at step 222 of method 200. For example, two thresholds can be defined. ε Enter and ε Exit This allows for control over event entry and exit conditions based on accuracy metrics. As discussed in conjunction with step 222, in some examples, accuracy can be one or more of percentage, mean squared error, and negative log-likelihood.

[0140] Events based on UE radio quality, events based on mobility KPIs, and events based on LCM can also be defined and used as triggers for event-based reporting.

[0141] Events based on UE radio quality can be triggered using a metric based on radio coupling gain (e.g., the RSRP difference between the serving cell and its strongest neighboring cell can be compared with a threshold used for event initiation triggering). RSRPdiff_ enter and the threshold used for event exit triggering RSRPdiff_exit (For comparison). The RSRP difference can be measured in dBm.

[0142] Mobility KPI-based events can be triggered using mobility KPI-related counters. For example, one or more of the following metrics could be used: number of handovers (HO), number of handover failures (HOF), number of ping-pongs (PP), number of unnecessary HOs, and number of radio link failures (RLF), or a metric based on one or more of these metrics could be used in conjunction with a threshold. MobCounter_enter and MobCounter_exit Compare them.

[0143] LCM-based events can be constructed based on different scenarios behind the LCM process. For example, data collection events for model training can be defined and can trigger high-fidelity measurement reporting. If performance degradation is detected, model monitoring events can also be defined to verify the model's accuracy and trigger lower-precision measurement reports accordingly.

[0144] At step 424, when the event entry condition is verified, UE 202 sends an event-triggered measurement report to network node 204. Using different types of events, UE 202 can perform the measurement reporting process at different accuracy levels and can meet different accuracy requirements. For example, different types of event triggering (e.g., accuracy-based and radio quality-based) or event triggering associated with different thresholds (e.g., two different accuracy-based thresholds) can result in the execution of reporting processes with different accuracy levels.

[0145] At step 426, similar to step 232, network node 204 may cache the reported configuration used by the UE. This enables network node 204 to evaluate whether the current method is beneficial to continue or revert to the legacy configuration. In some embodiments, this step may be omitted.

[0146] Figure 5 Message flow sequence 500 is shown. Message flow sequence 500 occurs between UE 202 and network node 204. In some embodiments, network node 204 may be a gNB.

[0147] Message flow sequence 500 corresponds to an embodiment of an algorithm for determining the output executed on the UE side. This algorithm can be an ML algorithm. The accuracy of the output is determined and verified against accuracy conditions. Based on this verification, the reporting configuration used by UE 202 can be changed. Message flow sequence 500 depicts changing the reporting configuration to a reporting configuration with a semi-persistent reporting type. Accuracy condition verification can be repeated to determine additional reporting configurations / periods.

[0148] Steps 510-522 correspond to steps 210-222 of method 200, except that in this case, the reported configuration is semi-persistent, rather than periodic.

[0149] At step 524, UE 202 sends a request or indication to network node 204 to switch to the semi-persistent reporting procedure. This request can be a request to switch to the semi-persistent reporting procedure after the accuracy condition has been verified at step 522 (i.e., the accuracy condition may not be verified at this stage, and the switch may be conditional upon verification). In some embodiments, the message may simply convey the evaluation result, which can be encoded in a bitmap.

[0150] In some embodiments, the message may include one bit of information. Based on the accuracy verification against a threshold, UE 202 may send a message 1 or 0 to indicate whether the accuracy verification condition is true.

[0151] In some embodiments, the message may include n-bit information. When multiple precision thresholds exist (which may correspond to multiple precision conditions and multiple potential reporting configurations), a general n-bit sequence indication can be sent from UE 202.

[0152] The indication message in step 524 can be sent via L1 or L2 signaling channels, for example, via a MAC control element (MAC CE) or UCI.

[0153] At step 526, network node 204 may evaluate the message from step 524 and prepare the corresponding preferred reporting configuration for UE 202 accordingly. In some embodiments, this step may be omitted.

[0154] At step 528, network node 204 sends a response to UE 202, instructing UE 202 to switch to a semi-persistent reporting procedure according to the request in step 524. If the request is for switching to a semi-persistent reporting procedure after accuracy conditions have been verified, the switching instruction may be conditional upon verification of accuracy conditions. Periodic reporting configurations for UE 202 are preferably authorized. Furthermore, network node 204 may configure a method for switching reporting periods for the UE.

[0155] Network node 204 can send a 1-bit message to authorize UE 202 to perform a semi-persistent reporting request as requested, or to reject the request.

[0156] Similarly, L1 / L2 signaling channels can be used for this indication message, such as MAC CE or DCI.

[0157] At step 530, UE 202 applies the new report configuration and sends a measurement report to network node 204. The report can have a configured accuracy level and a reporting period of T1. Therefore, the report is sent once every T1.

[0158] At step 532, UE 202 monitors accuracy-based verification conditions. Once these conditions are no longer met, UE 202 can switch to a different reporting period T2.

[0159] At step 534, the UE applies a new reporting configuration to network node 204. This new reporting configuration can have different reporting periods and / or different levels of precision. The reporting period at step 534 can be T2, so in some examples T2 ≠ T1.

[0160] In the above embodiments, measurement reporting is dynamically adjusted based on whether the output of an algorithm or process maintained on the UE side (which may be an AI / ML algorithm or process) meets accuracy requirements. In another embodiment, measurement reporting can be dynamic based on whether the output of an algorithm or process maintained on the network side meets accuracy requirements. This can be a supplement to or alternative to dynamic reporting based on the accuracy of the UE-side algorithm or process.

[0161] Figure 6 Message flow sequence 600 is shown. Message flow sequence 600 occurs between UE 202 and network node 204. In some embodiments, network node 204 may be a gNB.

[0162] Message flow sequence 600 corresponds to an embodiment of an algorithm that determines the output on the network side. This algorithm can be an ML algorithm. The accuracy of the output is determined and verified against accuracy conditions. Based on this verification, the reporting configuration used by UE 202 can be changed. Message flow sequence 600 depicts changing the reporting configuration to a reporting configuration with a periodic reporting type.

[0163] Steps corresponding to 210 and 212 may be included in this process (e.g., before step 610). These steps are not necessary because the network may rely on the capabilities that UE 202 may be assumed to possess, and UE 202 may not have its own communication accuracy requirements.

[0164] Message flow sequence 600 begins at step 610, which is similar to step 214. At step 610, network node 204 prepares configuration parameters to create a reporting framework. This reporting framework can be prepared for use by network 204 to determine outputs, such as RRM measurement predictions (e.g., for L3 mobility).

[0165] At step 612, network node 204 sends a (re)configuration message with configuration parameters to UE 202 to support the new reporting framework, such as report type, period, and number of reports. This message may be an RRC message.

[0166] At step 614, UE 202 performs a measurement procedure (which may be an L1 and / or L3 measurement procedure) and collects measurement data for network node 204 to use to determine the output.

[0167] At step 616, UE 202 sends a measurement report to network node 204. This measurement report may include the data collected at step 614. Since the network may not yet have decided to use a new (i.e., non-default or non-traditional) reporting configuration, this measurement report may be a default or traditional L1 or L3 measurement report.

[0168] At step 618, network node 204 determines the output. This output can be determined using (i.e., at least partially based on) measurement data from the measurement report received at step 616. The output can be determined at least partially based on an ML solution. By generating the output using UE measurement data as input, UE 202 can perform AI / ML-based functions. In NW-side embodiments, AI / ML LCM processes (e.g., training, inference, monitoring, etc.) can be performed on the network side. In one example use case, a time-series Long Short-Term Memory (LSTM) prediction model can be used to predict RRM measurements of mobility or sequences of measurement events. An example method for determining the output (in this example, RRM measurement prediction for L3 mobility) using an AI / ML model, including model layout, input, and output structures, will be described in more detail later in this specification.

[0169] At step 620, network node 204 verifies the accuracy verification condition. This can be performed within an evaluation time window, which may be configured by the network. Verifying the accuracy verification condition may include determining the accuracy of the output and comparing it to the accuracy verification condition.

[0170] The accuracy verification process performed on the network side at step 620 can be similar to the accuracy verification process performed on the UE side at step 222 of message flow sequence 200, but it is performed on the network side based on the output calculated at the network.

[0171] At step 622, network node 204 may determine the preferred reporting period for UE 202's measurement reports. In some examples, this preferred reporting period may be determined at an earlier stage (e.g., it may be sent in the RRC configuration message at step 612).

[0172] At step 624, network node 204 sends an indication to UE 202 that it wants to switch the measurement reporting configuration (in this case, switch to a measurement reporting configuration with a periodic reporting process). This indication may include an indication of the reporting period.

[0173] The L1 / L2 signaling channel can be used for this indication message, such as MAC CE or DCI.

[0174] In steps 626a-626c, UE 202 applies the new reporting configuration and sends a measurement report (which may have different or relatively lower accuracy than the conventional or default measurement report) to network node 204.

[0175] At step 628, network node 204 may cache the reported configuration used by the UE. This enables network node 204 to evaluate whether the current method is beneficial to continue or revert to the legacy configuration. In some embodiments, this step may be omitted.

[0176] Figure 7 Message flow sequence 700 is shown. Message flow sequence 700 occurs between UE 202 and network node 204. In some embodiments, network node 204 may be a gNB.

[0177] Message flow sequence 700 corresponds to an embodiment of an algorithm that determines the output on the network side. This algorithm can be an ML algorithm. The accuracy of the output is determined and verified against accuracy conditions. Based on this verification, the reporting configuration used by UE 202 can be changed. Message flow sequence 700 depicts changing the reporting configuration to a reporting configuration with an event-triggered reporting type.

[0178] Steps corresponding to 210 and 212 may be included in this process (e.g., before step 710). These steps are not necessary because the network may rely on the capabilities that UE 202 may be assumed to possess, and UE 202 may not have its own communication accuracy requirements.

[0179] Message flow sequence 700 begins at step 710. Steps 710-720 may correspond to steps 610-620; however, the configuration parameters at step 710 may correspond to a reporting configuration with an event-triggered type.

[0180] In this case, the step corresponding to step 622 may be unnecessary because the reporting is event-triggered rather than periodic.

[0181] At step 722, network node 204 sends an instruction to UE 202 to switch the measurement reporting configuration (in this case, switch to the measurement reporting configuration with event-triggered reporting).

[0182] The L1 / L2 signaling channel can be used for this indication message, such as MAC CE or DCI.

[0183] At step 724, UE 202 performs a measurement procedure (which may be an L1 and / or L3 measurement procedure) and collects measurement data for use by network node 204.

[0184] At step 726, UE 202 sends a measurement report to network node 204 according to the configured event-based reporting configuration.

[0185] Figure 8 Message flow sequence 800 is shown. Message flow sequence 800 occurs between UE 202 and network node 204. In some embodiments, network node 204 may be a gNB.

[0186] Message flow sequence 800 corresponds to an embodiment of an algorithm that determines the output on the network side. This algorithm can be an ML algorithm. The accuracy of the output is determined and verified against accuracy conditions. Based on this verification, the reporting configuration used by UE 202 can be changed. Message flow sequence 800 depicts changing the reporting configuration to a reporting configuration with a semi-persistent reporting type.

[0187] The steps corresponding to 210 and 212 can be included in this process (e.g., before step 810). These steps are not necessary because the network can rely on the capabilities that UE 202 can be assumed to possess, and UE 202 may not have its own communication accuracy requirements.

[0188] Message flow sequence 800 begins at step 810. Steps 810-820 may correspond to steps 610-620; however, the configuration parameters at step 810 may correspond to the reporting configuration with a semi-persistent reporting type.

[0189] At step 822, network node 204 sends an indication to UE 202 that it wants to switch the measurement reporting configuration (in this case, switch to a measurement reporting configuration with a periodic reporting process). This indication may specify the reporting period T1.

[0190] The L1 / L2 signaling channel can be used for this indication message, such as MAC CE or DCI.

[0191] In steps 824a-824c, UE 202 applies the new reporting configuration and sends a measurement report (which may have different or relatively lower accuracy than the traditional or default measurement report) to network node 204. The reporting period is T1. Therefore, the report is sent once every T1.

[0192] At step 826, network node 204 monitors accuracy-based verification conditions.

[0193] At step 828, after determining at step 826 that the accuracy verification condition determined in step 820 is no longer met, network node 204 can instruct UE 202 to switch to a different reporting period T2. The L1 / L2 signaling channel can be used for this instruction message, such as MAC CE or DCI.

[0194] In steps 830a-830c, UE 202 applies a new reporting configuration to network node 204. This new reporting configuration may have a different reporting period and / or a different level of precision than the report in step 530. The reporting period in step 534 may be T2, so in some examples T2 ≠ T1.

[0195] Figure 9 This is a block diagram illustrating an example device 900 according to an example embodiment of the present invention.

[0196] The device 900 includes a processor 910 and a memory 912.

[0197] Processor 910 can execute instructions stored in memory 912. Memory 912 can store the instructions executed by processor 910.

[0198] In some embodiments, device 900 may be a user equipment, such as UE 202. In some embodiments, device may be a network node, such as network node 204. In some embodiments, network node and device 900 may be gNB.

[0199] The processor 910 can be coupled to a receiver and / or transmitter, which allows the device 900 to receive and / or send data to other devices respectively.

[0200] The processor 910 may optionally be coupled to the measuring device 914. For example, a receiving antenna can be considered a measuring device capable of acquiring radio signals. Further measurements can be derived from the radio signals using various known measuring devices and processing techniques to determine quantities such as RSRP.

[0201] Figure 10 This is a flowchart illustrating an example method 1000 according to an exemplary embodiment of the present invention. In some exemplary embodiments, method 1000 may be performed by device 900.

[0202] Method 1000 begins with step 1010. At step 1010, a reporting framework is obtained. This reporting framework maps one or more precision conditions to one or more reporting actions and / or one or more changes in reporting actions.

[0203] One or more precision conditions may include at least one or more of the following: a threshold precision where a precision condition above it is satisfied; a threshold precision where a precision condition below it is satisfied; a threshold precision where a precision condition within it is satisfied; and a threshold precision where a precision condition outside it is satisfied.

[0204] One or more precision conditions can be defined in at least one or more of the following ways: the proportion of one or more outputs that match one or more corresponding true values; the mean square error of one or more outputs relative to one or more corresponding true values; and one or more negative logarithms of one or more likelihoods, where the likelihood is the likelihood assigned to the output that matches the corresponding true value. Based on the chosen precision metric, the precision condition can be expressed accordingly. Different precision expressions can be applied to different outputs and methods of determining the output. For example, mean square error may not be appropriate when the output is an ID, but it may be more appropriate when the output is a measurement or predictive measurement of a value defined on a continuous scale such as RSRP.

[0205] At step 1012, an indication is obtained as to whether the output satisfies one or more precision conditions.

[0206] The output may include one or more of the following: the measured predicted value; the target cell; and the target beam.

[0207] At step 1014, at least in part based on the acquired reporting framework and the acquired indication of whether the precision conditions in one or more precision conditions are satisfied by the output, the reporting behavior and / or one or more reporting behavior changes are determined.

[0208] One or more reporting actions. Each reporting action may include the use of a corresponding measurement report configuration. This allows for dynamic changes to the measurement report configuration based on accuracy conditions, and thus provides dynamic reporting actions based on accuracy conditions. Each corresponding measurement report configuration for one or more reporting actions may include at least one or more of the following parameters: reporting criteria; the event used to trigger the report; the reporting period / interval; and the number of reports. The corresponding measurement report configuration for one or more reporting actions may include at least one or more accuracy levels at which one or more user equipment measurements will be collected.

[0209] The reporting framework may include a first reporting action and a second reporting action. The first reporting action, one or more of the reported actions, may include a first measurement report configuration that includes a first accuracy level, at which one or more user equipment measurements are collected. The second reporting action, one or more of the reported actions, may include a second measurement report configuration that includes a second accuracy level, at which one or more user equipment measurements are collected, wherein the first accuracy level is higher than the second accuracy level. A first accuracy condition can be satisfied above a threshold and mapped to the second reporting action, and a second accuracy condition can be satisfied below a threshold and mapped to the first reporting action. This allows for dynamic switching between measurement report configurations with different accuracies.

[0210] In some example embodiments, reporting behavior may include event-triggered measurement reports. The reporting framework may include one or more measurement report configurations that map one or more accuracy conditions to triggering the event-triggered measurement report. Using the accuracy conditions used to trigger the event-triggered measurement report, the measurement report may be dynamically sent in response to the accuracy conditions.

[0211] In some embodiments, method 1000 is performed by a user equipment (such as UE 202) and further includes collecting measurement data associated with radio access network signals (e.g., using measurement device 914). Obtaining a reporting framework may include receiving a reporting framework from a network node (such as network node 204), and obtaining an indication of whether the output meets accuracy conditions may include generating an output based on the acquired measurement data. The indication of the accuracy of the generated output may be determined at least in part based on the generated output, and whether the generated output meets accuracy conditions may be verified based on the accuracy indication and the accuracy conditions.

[0212] If method 1000 is performed by the user equipment, method 1000 may further include: in response to an indication that the output meets the accuracy condition, sending a request to the network node for using the measurement report configuration included in the reporting behavior mapped to the accuracy condition by the reporting framework. The user equipment can then receive from the network node an indication that the measurement report configuration will be used by the user equipment, and apply the measurement report configuration.

[0213] If method 1000 is executed by a user equipment and the accuracy conditions satisfied by the output are mapped to a reporting behavior that includes a measurement report configuration with event-based reporting criteria, the method may further include verifying whether an event identified in the measurement configuration has occurred, and triggering a measurement report based on the verification that the event has occurred.

[0214] If method 1000 is executed by the user equipment, the method may further include sending capability information associated with the user equipment and the accuracy requirements of the user equipment's output to the network node. Based on the user equipment's capabilities and requirements, the network node can provide a more suitable reporting framework.

[0215] In some embodiments, method 1000 may be performed by a network node (e.g., device 900 may be a network node, such as a gNB). When method 1000 is performed by a network node, it may further include: receiving measurement data associated with a radio access network signal from a user equipment, and obtaining an indication of whether the output satisfies one or more accuracy conditions may include: generating an output based on the measurement data received from the user equipment, determining the accuracy of the generated output, and verifying whether the generated output satisfies the accuracy conditions. Therefore, the network node can verify the accuracy of the output of a network-side process that uses user equipment measurement data as input.

[0216] When method 1000 is executed by a network node, the method may further include: sending an instruction to a user equipment to use a measurement report configuration comprised of reporting behaviors mapped to satisfied accuracy conditions. Therefore, the network node can dynamically change the measurement reporting based on the accuracy of the output of the network-side process.

[0217] When method 1000 is performed by a network node, the method may further include: receiving capability information from the user equipment, and obtaining the reporting framework may include: generating the reporting framework based on the capability information associated with the user equipment and the accuracy requirements for the output. In some examples, the reporting framework may depend on the user equipment capabilities that the user equipment is expected to have, so it may not be necessary to obtain the capability information from the user equipment in all examples.

[0218] In some embodiments, method 1000 can be performed by a network node, and obtaining the reporting framework can include generating a reporting framework based on capability information associated with the user equipment and the accuracy requirements of the user equipment's output. The method may further include sending one or more accuracy conditions to the user equipment and receiving from the user equipment an indication of whether the output meets one or more accuracy conditions. The apparatus can then send an instruction to the user equipment to use a measurement report configuration included by reporting behavior mapped to the met accuracy conditions. Therefore, the output and whether the output meets the accuracy verification conditions can be determined at the user equipment and indicated to the network, and the network can determine whether to instruct the user equipment to use the measurement report configuration based on whether the accuracy verification conditions are met, thereby providing dynamic measurement reports.

[0219] Figure 11This is a flowchart illustrating an example method 1100 according to an exemplary embodiment of the present invention. In some exemplary embodiments, method 1100 may be performed by device 900.

[0220] Method 1100 begins at step 1110. At step 1010, user equipment measurement data is acquired. In some embodiments, the user equipment measurement data is acquired from a measurement device 914 coupled to device 900, and device 900 is the user equipment. In some embodiments, the user equipment measurement data is acquired from another device. For example, the user equipment measurement data may be received from the user equipment at device 900. The user equipment measurement data may include measurements associated with a radio access network signal. The user equipment measurement data may include measurement data indicating one or more of the following: reference signal received power; reference signal received quality; received signal strength indicator; signal-to-interference-plus-noise ratio; and signal-to-noise ratio. The user equipment measurement data may be collected at different levels of accuracy (e.g., using different hardware or hardware combinations that produce measurements of different accuracies).

[0221] At step 1112, user equipment measurement data is processed to determine the output. Processing the user equipment measurement data may include using an AI / ML model to determine the output. The output may predict the measurement data values. The output may predict the target cell or target beam.

[0222] At step 1114, an indication of the accuracy of the output is determined. The accuracy may be determined at least in part based on the output and the true value. The true value may be determined based on at least one of: second measurement data that is more accurate than the user equipment measurement data; a processing method that is computationally more intensive than the processing method used in the step of processing the user equipment measurement data to determine the output; third measurement data that includes more data than the user equipment measurement data; and fourth measurement data that includes data collected after the user equipment measurement data was collected.

[0223] At step 1116, the indicated precision is compared with one or more precision conditions. The indication of the output precision and one or more precision conditions may be defined in at least one or more of the following aspects: the proportion of the determined output matching the true value; the mean square error of the determined output compared to the true value; and the negative log-likelihood. Precision conditions may include one or more of the following: a threshold precision where a precision condition above it is satisfied, a threshold precision where a precision condition below it is satisfied, a threshold precision where a precision condition within its range is satisfied, and a threshold precision where a precision condition outside its range is satisfied. ML model implementation example

[0224] Figure 12This is a diagram schematically illustrating an implementation example of the ML time series forecasting model 1200. Model 1200 can be used for inter-frequency mobility management and HO decision-making.

[0225] In the example structure of model 1200, input frame 1210 can be a sequence of L2-RSRP values ​​1212 and / or a sequence of beam IDs 1214. These two inputs can be concatenated in operation 1220 before being fed into the ML model. This example uses L2-RSRP input values ​​from 294 beams (14 beams from each of 21 gNBs) within a 3-second window. In this example, the gNB measures a synchronization signal block (SSB) every 20ms, so if the total input length is 3 seconds, the measurement duration covers 150 samples. The input frames (i.e., L2-RSRP + beam ID) are fed into the ML model to obtain the time-series output 1260, in this example, which is a Long Short-Term Memory (LSTM) recurrent neural network (RNN). In this example, the concatenated inputs are fed into one or more dense layers 1230 before being fed into one or more LSTM / RNN layers 1240 and before being fed into another or more dense layers 1250. The output of model 1260 can be any of the following: a predicted L2-RSRP value, or a key marker (such as a HO indicator metric / predicted HO event within the prediction window). The main advantage of model 1200 is that it allows for the delivery of additional information about beam / cell selection on top of the traditional L3 mobility management framework. enter

[0226] As mentioned above, time t The input sequence at the position is represented as Each sample Q It is the following tuple: The input RSRP vector is measured from 294 beams (14 beams from 21 gNBs), i.e. The input frame has changed from the past. N The measurement begins in milliseconds, and it contains information specific to each measurement or beam ID. q This is a past sample. In this example, gNB measures SSB every 20ms. q =150 measurement samples in total input length N =Collected within 3000ms. Therefore, the input data is a 150×588 matrix. Output (labels)

[0227] The labeled data used to train the model is a given prediction window. MA sequence of RSRP samples in ms, containing each measurement or beam ID. m 1. Past samples. In this example, the predicted output is... M =RSRP measured within 100ms from 294 beams (14 beams from 21 gNBs), which includes for each index m =5 samples. Therefore, the label data is a 294×5 matrix. . loss function

[0228] Implement a time series forecasting model to learn historical RSRP plus each beam index. For future RSRP values Dependencies. LSTM has been proven to be an effective method for solving long-term dependency problems; this method has been used... Figure 10 The model design in this paper. This model includes the relationship between each input. Q i Associated parameter set Earlier Multiply by weight and with later Add them together to capture time-series dependencies. Output layer With each tag P i Related. A loss function is defined to evaluate the model's predictions. With model detection P Errors between them, for example: train

[0229] During training, optimization algorithms such as stochastic gradient descent (SGD) can be applied to tune each individual element. This minimizes the average prediction error of the samples collected from the UE. Model reliability assessment and evaluation are performed after the training step is completed. ML model implementation example

[0230] Figure 12 An example implementation of ML time series forecasting model 1200 is illustrated schematically. Model 1200 can be used for inter-frequency mobility management and HO decision-making.

[0231] In the example structure of model 1200, input frame 1210 can be a sequence of L2-RSRP values ​​1212 and / or a sequence of beam IDs 1214. These two inputs can be concatenated in operation 1220 before being fed into the ML model. This example uses L2-RSRP input values ​​from 294 beams (14 beams from each of 21 gNBs) within a 3-second window. In this example, the gNB measures a synchronization signal block (SSB) every 20ms, so if the total input length is 3 seconds, the measurement duration covers 150 samples. The input frames (i.e., L2-RSRP + beam ID) are fed into the ML model to obtain the time-series output 1260, in this example, which is a Long Short-Term Memory (LSTM) recurrent neural network (RNN). In this example, the concatenated inputs are fed into one or more dense layers 1230 before being fed into one or more LSTM / RNN layers 1240 and before being fed into another or more dense layers 1250. The output of model 1260 can be any of the following: a predicted L2-RSRP value, or a key marker (such as a HO indicator metric / predicted HO event within the prediction window). The main advantage of model 1200 is that it allows for the delivery of additional information about beam / cell selection on top of the traditional L3 mobility management framework. enter

[0232] As mentioned above, time t The input sequence at the position is represented as Each sample Q It is the following tuple: The input RSRP vector is measured from 294 beams (14 beams from 21 gNBs), i.e. The input frame has changed from the past. N The measurement begins in milliseconds, and it contains information specific to each measurement or beam ID. q This is a past sample. In this example, gNB measures SSB every 20ms. q =150 measurement samples in total input length N =Collected within 3000ms. Therefore, the input data is a 150×588 matrix. Output (labels)

[0233] The labeled data used to train the model is a given prediction window. M A sequence of RSRP samples in ms, containing each measurement or beam ID. m 1. Past samples. In this example, the predicted output is... M=RSRP measured within 100ms from 294 beams (14 beams from 21 gNBs), which includes for each index m =5 samples. Therefore, the label data is a 294×5 matrix. . loss function

[0234] Implement a time series forecasting model to learn historical RSRP plus each beam index. For future RSRP values Dependencies. LSTM has been proven to be an effective method for solving long-term dependency problems; this method has been used... Figure 10 The model design in this paper. This model includes the relationship between each input. Q i Associated parameter set Earlier Multiply by weight and with later Add them together to capture time-series dependencies. Output layer With each tag P i Related. A loss function is defined to evaluate the model's predictions. With model detection P Errors between them, for example: train

[0235] During training, optimization algorithms such as stochastic gradient descent (SGD) can be applied to tune each individual element. This minimizes the average prediction error of the samples collected from the UE. Model reliability assessment and evaluation are performed after the training step is completed.

[0236] For the sake of completeness, Figure 13 This is a schematic diagram of components of one or more example embodiments previously described, collectively referred to below as processing system 1900. For example, processing system 1900 may consist of devices mentioned in the following claims.

[0237] The processing system 1900 may include a processor 1902, a memory 1904 tightly coupled to the processor and consisting of RAM 1914 and ROM 1912, and optional user inputs 1910 and a display 1918. The processing system 1900 may include one or more network / device interfaces 1908 for connecting to a network / device, such as a wired or wireless modem. The network / device interface 1908 may also operate as a connection to other devices, such as devices that are not network-side devices. Therefore, direct connection between devices is possible without network involvement.

[0238] The processor 1902 is connected to each of the other components to control their operation.

[0239] Memory 1904 may include non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD). The ROM 1912 of memory 1904 stores the operating system 1915, and may also store software applications 1916. The RAM 1914 of memory 1904 is used by processor 1902 for temporary data storage. The operating system 1915 may contain code that, when executed by the processor, implements aspects of methods 1000 and 1100 described above, as well as aspects of message flow sequences 200, 400, 500, 600, 700, and 800. Note that in the case of small devices, memory is best suited for small size applications, i.e., hard disk drives (HDDs) or solid-state drives (SSDs) are not always used.

[0240] The processor 1902 can take any suitable form. For example, it can be a microcontroller, multiple microcontrollers, a processor, or multiple processors.

[0241] The processing system 1900 can be a standalone computer, server, console, or its network. The processing system 1900 and the necessary structural components can all be internal to a device (such as an IoT device), i.e., embedded in a very small size.

[0242] In some example embodiments, the processing system 1900 may also be associated with external software applications. These may be applications stored on a remote server device / app, and may run partially or exclusively on the remote server device / app. These applications may be referred to as cloud-hosted applications. The processing system 1900 may communicate with the remote server device / app to utilize the software applications stored there.

[0243] Figure 14A tangible medium in the form of a removable memory unit 2010 storing computer-readable code, which, when run by a computer, can perform the methods according to the example embodiments described above. The removable memory unit 2010 may be a memory stick, such as a USB memory stick, having internal memory 2030 for storing computer-readable code. The internal memory 2030 can be accessed by a computer system via a connector 2020. Of course, other forms of tangible storage media can be used, as will be apparent to those skilled in the art. The tangible medium can be any device / apparatus capable of storing data / information that can be exchanged between devices / apparatus / networks.

[0244] Embodiments of the present invention can be implemented using software, hardware, application logic, or a combination of software, hardware, and application logic. The software, application logic, and / or hardware can reside on memory or any computer medium. In example embodiments, the application logic, software, or instruction set is maintained on any of a variety of conventional computer-readable media. In the context of this document, "memory" or "computer-readable medium" can be any non-transitory medium or component that can contain, store, transmit, propagate, or transfer instructions for use by or in conjunction with an instruction execution system, apparatus, or device, such as a computer.

[0245] In the relevant context, references to “computer-readable medium,” “computer program product,” “computer program tangibly embodied,” or “processor” or “processing circuit system” should be understood to include not only computers with different architectures, such as single / multiprocessor architectures and sequencer / parallel architectures, but also special-purpose circuits, such as field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), signal processing devices / apparatus, and other devices / apparatus. References to computer programs, instructions, code, etc., should be understood to represent the software (such as the programmable content of hardware devices / apparatus) of a programmable processor firmware as processor instructions, or the configuration or configuration settings of fixed-function devices / apparatus, gate arrays, programmable logic devices / apparatus, etc.

[0246] If necessary, the different functions discussed herein can be executed in different orders and / or concurrently with each other. Furthermore, one or more of the functions described above can be optional or can be combined, if needed. Similarly, it should be understood that... Figure 2 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 , Figure 10 and Figure 11 The flowcharts and signal diagrams are merely examples, and the various operations depicted therein may be omitted, reordered, and / or combined.

[0247] It should be understood that the above exemplary embodiments are purely illustrative and do not limit the scope of the invention. Other variations and modifications will become apparent to those skilled in the art after reading this specification.

[0248] Furthermore, the disclosure of this application should be understood to include any novel feature or combination of novel features or any generalization thereof explicitly or implicitly disclosed herein, and new claims may be made in the course of this application or any application derived therefrom to cover any such feature and / or combination of such features.

[0249] Although various aspects of the invention are set forth in the independent claims, other aspects of the invention include other combinations of features from the exemplary embodiments and / or dependent claims with features of the independent claims, and not just combinations expressly set forth in the claims.

[0250] It should also be noted that while various examples have been described above, these descriptions should not be considered limiting. Rather, several changes and modifications may be made without departing from the scope of the invention as defined in the appended claims.

Claims

1. An apparatus comprising: Components for acquiring a reporting framework, wherein the reporting framework maps one or more precision conditions to one or more reporting behaviors and / or one or more reporting behavior variations; A component for obtaining an indication of whether the output meets one or more accuracy conditions; as well as A component for determining reporting behavior and / or one or more changes in reporting behavior, based at least in part on the reporting framework and the indication of whether the precision condition in one or more precision conditions is satisfied by the output.

2. The apparatus according to claim 1, wherein: The one or more accuracy conditions include at least one or more of the following: a threshold accuracy where the above accuracy conditions are satisfied; in The threshold precision for which the following precision conditions are met; the threshold precision for which the precision conditions are met within its range; and the threshold precision for which the precision conditions are met outside its range.

3. The apparatus according to claim 1 or 2, wherein the precision of one or more of the one or more precision conditions is defined in at least one or more of the following aspects: The proportion of one or more outputs that match one or more corresponding truth values; The mean square error of one or more outputs relative to one or more corresponding true values; as well as One or more negative logarithms of one or more likelihoods, wherein the likelihood is the likelihood assigned to the output that matches the corresponding truth value.

4. The apparatus according to any one of claims 1 to 3, wherein the output comprises one or more of the following: The predicted value of the measurement; Target community; and Target beam.

5. The apparatus according to any one of claims 1 to 4, wherein: Each of the one or more reporting actions includes the use of the corresponding measurement report configuration.

6. The apparatus according to claim 5, wherein: Each corresponding measurement report configuration of the one or more reporting behaviors includes at least one or more of the following parameters: reporting criteria; and the event used to trigger the reporting. Reporting period / interval; and the number of reports.

7. The apparatus according to any one of claims 5 or 6, wherein the corresponding measurement report configuration for the one or more reporting behaviors includes at least one or more accuracy levels, at which one or more user equipment measurements are collected.

8. The apparatus according to claim 7, wherein: The first reporting action in the one or more reporting actions includes a first measurement report configuration, which includes a first accuracy level, at which one or more user equipment measurements will be collected; The second reporting action in the one or more reporting actions includes a second measurement report configuration, which includes a second accuracy level, at which one or more user equipment measurements will be collected, wherein the first accuracy level is higher than the second accuracy level; The first precision condition is met above the threshold and is mapped to the second reporting behavior; and The second precision condition is satisfied below the threshold and is mapped to the first reporting behavior.

9. The apparatus according to any one of claims 1 to 4, wherein: The reporting behavior includes event-triggered measurement reports; as well as The reporting framework includes one or more measurement report configurations that map one or more accuracy conditions to the triggering of event-triggered measurement reports.

10. The apparatus according to any one of claims 1 to 9, wherein the apparatus is a user equipment, and wherein: The apparatus further includes: components for collecting measurement data associated with wireless access network signals; The component for obtaining the reporting framework includes: a component for receiving the reporting framework from a network node; and The component for obtaining an indication of whether the output meets one or more accuracy conditions includes: a component for generating the output based on measurement data acquired by the user equipment, a component for determining the accuracy of the generated output, and a component for verifying whether the generated output meets the accuracy conditions.

11. The apparatus of claim 10, wherein each of the one or more reporting actions includes the use of a corresponding measurement report configuration, and the apparatus comprises: A component for sending a request to a network node in response to an indication that the output meets the accuracy condition, the request being used to utilize the measurement report configuration included in the reporting behavior mapped by the reporting framework to the accuracy condition; A component for receiving a response from the network node, the response indicating that the measurement report configuration will be used by the device; as well as Components for applying the indicated measurement report configuration.

12. The apparatus of claim 11, wherein the accuracy condition satisfied by the output is mapped to a reporting behavior including a measurement report configuration having event-based reporting criteria; the apparatus further comprises: A component used to verify whether the event identified in the measurement configuration has occurred; as well as A component used to trigger a measurement report based on verification that the event has occurred.

13. The apparatus according to any one of claims 10 to 12, further comprising: Components for sending capability information associated with the user equipment to the network node; as well as A component used to send the accuracy requirements for the output to the network node.

14. The apparatus according to any one of claims 1 to 9, wherein the apparatus is a network node, and wherein: The apparatus further includes: a component for receiving measurement data associated with a wireless access network signal from a user equipment; The component for obtaining an indication of whether the output meets one or more accuracy conditions includes: a component for generating the output based on the measurement data received from the user equipment, a component for determining the accuracy of the generated output, and a component for verifying whether the generated output meets the accuracy conditions.

15. The apparatus of claim 14, wherein the apparatus includes a component for sending instructions to the user equipment, the instructions being configured to use a measurement report configuration included by the reporting behavior mapped to the satisfied accuracy condition.

16. The apparatus according to claim 14 or 15, wherein: The apparatus further includes: components for receiving capability information from the user equipment; and The components for obtaining the reporting framework include: components for generating the reporting framework based on capability information associated with the user equipment and accuracy requirements for the output.

17. The apparatus according to any one of claims 1 to 9, wherein the apparatus is a network node, and wherein: The components used to obtain the reporting framework include: components for generating the reporting framework based on capability information associated with the user equipment and the accuracy requirements of the user equipment's output; The device includes: a component for transmitting the one or more accuracy conditions to the user equipment; The apparatus includes: a component for receiving from a user equipment an indication of whether the output satisfies one or more accuracy conditions; and The apparatus includes a component for sending instructions to the user equipment, the instructions being configured to use a measurement report configuration included by the reporting behavior mapped to the satisfied accuracy conditions.

18. A method comprising: Obtain a reporting framework, wherein the reporting framework maps one or more precision conditions to one or more reporting behaviors and / or one or more reporting behavior variations; Obtain an indication of whether the output meets one or more of the precision conditions; as well as The reporting behavior and / or one or more reporting behavior changes are determined at least in part based on the reporting framework and the indication of whether the precision condition in one or more precision conditions is satisfied by the output.

19. The method of claim 18, wherein: The method further includes: collecting measurement data associated with wireless access network signals; Obtaining the reporting framework includes: receiving the reporting framework from a network node; and Obtaining an indication of whether the output meets one or more accuracy conditions includes: generating the output based on the collected measurement data, determining the accuracy of the generated output, and verifying whether the generated output meets the accuracy conditions.

20. The method of claim 18, wherein: The method further includes: receiving measurement data associated with a wireless access network signal from a user equipment; Obtaining an indication of whether the output meets one or more accuracy conditions includes: generating the output based on the measurement data received from the user equipment, determining the accuracy of the generated output, and verifying whether the generated output meets the accuracy conditions.

Images