Beam prediction

Abstract

There is provided a method, apparatus and system for beam prediction, and means for measuring a first set of reference signal resources of a set of configured reference signal resources for a respective beam of a network entity for determining at least one predicted beam, wherein the configured reference signal resource set is associated with prediction-based beam reporting or prediction-based beam switching; means for receiving, from a network entity, an indication that a target transmission configuration indicator (TCI) state corresponding to the at least one predicted beam is to be activated or indicated; means for determining whether the target TCI state is known or unknown based on at least one of: the at least one predicted beam and the measured first set of reference signal resources; and means for switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

Description

Technical Field

[0001] Example embodiments relate to beam prediction and determining a known or unknown state of a target transmission configuration indicator (TCI) state corresponding to at least one predicted beam. Background Art

[0002] The Third Generation Partnership Project (3GPP) has developed and is developing standards for fourth-generation (4G) (also known as Long Term Evolution (LTE)) and fifth-generation (5G) (also known as New Radio (NR)) communication systems. Regarding 5G NR communication systems and future communication systems, throughput and network capacity can be increased by using massive multiple-input multiple-output (MIMO) technology with beamforming, whereby network entities can establish and communicate via relatively narrow beams directed toward individual user terminals.

[0003] The process of determining which beam or beams to use is often referred to as beam management. Summary of the Invention

[0004] The scope of protection sought by various embodiments of the present invention is defined by the independent claims. Embodiments and features described in this specification that do not fall within the scope of the independent claims (if any) should be interpreted as examples that help understand the various embodiments of the present invention.

[0005] A first aspect provides an apparatus comprising: a component for measuring a first set of reference signal resources in a set of configured reference signal resources for a corresponding beam of a network entity for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; a component for receiving an indication from the network entity that a target transmission configuration indicator (TCI) state corresponding to at least one predicted beam is to be activated or indicated; a component for determining whether the target TCI state is known or unknown based on at least one of the following items: at least one predicted beam and the measured first set of reference signal resources; and a component for switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0006] In some example embodiments, the configured reference signal resource set is configured by a network entity.

[0007] In some example embodiments, the apparatus further comprises a component for receiving a message from a network entity for triggering the apparatus to perform prediction-based beam reporting or prediction-based beam switching, wherein the measurement is performed upon or after receipt of the message.

[0008] In some example embodiments, the component for measuring is further configured to measure a first set of reference signal resources for the first group of beams for determining at least one predicted beam from the second group of beams using a model, wherein the model performs an inference operation for providing data representing an indication of the at least one predicted beam.

[0009] In some example embodiments, the apparatus further comprises: means for inputting data representing the measured first set of reference signal resources into a model; and means for reporting data representing an indication of at least one predicted beam to a network entity, wherein at least one reference signal resource in the configured set of reference signal resources is associated with at least one predicted beam.

[0010] In some example embodiments, the apparatus further comprises: means for reporting the measured first set of reference signal resources to a network entity for input to the model by the network entity. In some example embodiments, the apparatus further comprises: means for receiving data representing an indication of at least one predicted beam.

[0011] In some example embodiments, the component for determining whether the target TCI state is known or unknown further includes: a component for determining that one or more predicted quality metrics of one or more reference signal resources associated with the corresponding reporting prediction beam are greater than or equal to a preconfigured or predefined quality threshold level; in response to determining that at least one of the one or more predicted quality metrics of the one or more reference signal resources associated with the corresponding reporting prediction beam is greater than or equal to the preconfigured or predefined quality threshold level, determining that the target TCI state is known when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state.

[0012] In some example embodiments, the quality metric comprises one or more of the following group: reference signal received power (RSRP); L1 physical layer RSRP (L1-RSRP); signal to noise ratio (SNR) of the reference signal resource; or any other quality metric associated with the reference signal resource.

[0013] In some example embodiments, the component for determining whether the target TCI state is known or unknown also includes: a component for determining that one or more current model performance metrics of one or more reported prediction beams corresponding to the reported prediction beams associated with one or more reference signal resources are greater than or equal to a preconfigured or predefined model performance threshold level; in response to determining that at least one current model performance metric of the one or more reported prediction beams corresponding to the reported prediction beams associated with the one or more reference signal resources is greater than or equal to the preconfigured or predefined model performance threshold level, when the one or more reference signal resources correspond to the reference signal resources associated with the target TCI state, determining that the target TCI state is known.

[0014] In some example embodiments, the model performance metric includes one or more of the following groups: a model prediction accuracy metric; a precision and recall metric; an area under the curve receiver operating curve (AUC-ROC) metric; a model confidence metric; a probability score metric; or any other model performance metric used to evaluate the predictive performance of the model.

[0015] In some example embodiments, wherein the means for switching to the target TCI state within the time period based on a determination that the target TCI state is known further comprises one or more of: performing the TCI state switching within the time period according to the known target TCI state, wherein the time period is calculated based on an L1-RSRP measurement time period required to omit L1-RSRP measurements of reference signal resources of the target TCI state; or performing the TCI state switching within the time period according to the known target TCI state, wherein the time period is calculated based on a determination that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0016] In some example embodiments, where the apparatus is configured to perform prediction-based beam reporting or before prediction-based beam switching, the component for measuring is further configured to: trigger measurement of at least one of the reference signal resources associated with at least one of the predicted beams after reporting data representing an indication of one or more predicted beams, wherein the measurement of at least one of the reference signal resources associated with at least one of the predicted beams is performed based on one or more of the following groups: measuring the reference signal resources associated with at least one predicted beam in a first time slot of a specific reference signal resource occurring after reporting the predicted beam associated with the reference signal resource; or measuring the reference signal resources associated with at least one predicted beam within a predetermined or specified time period after reporting the predicted beam associated with the reference signal resource.

[0017] In some example embodiments, wherein after measuring a reference signal associated with at least one predicted beam, prediction-based beam switching is performed based on whether a target TCI state is known.

[0018] In some example embodiments, the measurement of the at least one measured reference signal resource is an L1 physical layer measurement; and / or the L1 physical layer measurement is a reference signal resource received power (RSRP) measurement.

[0019] In some example embodiments, where the apparatus is configured to perform prediction-based beam reporting, the means for determining whether the target TCI state is known or unknown further comprises: means for determining that a first set of reference signal resources is associated with a corresponding reported predicted beam; and means for determining that the target TCI state is known when the reference signal resources in the first reference signal set correspond to reference signal resources associated with the target TCI state.

[0020] In some example embodiments, the means for determining whether the target TCI state is known or unknown further comprises means for determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of: at least a predicted beam and a measured first set of reference signal resources.

[0021] In some example embodiments, the one or more conditions further include: if it is determined that at least one measurement reference signal resource in the first reference signal resource set corresponds to: a reference signal included in or associated with the target TCI state; or a reference signal received from a source that is quasi-co-located with the source of the reference signal included in or associated with the target TCI state; or a reference signal corresponding to at least one reference signal included in or associated with the target TCI state and having a quality metric that is higher than a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0022] In some example embodiments, the one or more conditions further include: if a predicted quality metric of a reference signal corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0023] In some example embodiments, the apparatus is configured to: perform monitoring of prediction accuracy, wherein the monitoring of prediction accuracy is based on measurements of one or more reference signal resources corresponding to one or more previous prediction beams; and determine a prediction accuracy metric for at least one prediction beam based on said monitoring of the previous prediction beams and an indication of the corresponding at least one prediction beam; wherein the one or more conditions further include: if the prediction accuracy metric for at least one prediction beam is greater than or equal to a prediction accuracy threshold level, then the target TCI state is known.

[0024] In some example embodiments, the one or more conditions further include: if a current model performance metric corresponding to a model inference operation of at least one predicted beam having a reference signal included in or associated with the target TCI state is greater than or equal to a model performance threshold level, then the target TCI state is known.

[0025] In some example embodiments, the one or more conditions further include: if a current model confidence or probability score metric corresponding to a model inference operation of at least one predicted beam having a reference signal included in or associated with the target TCI state is greater than or equal to a corresponding confidence threshold level or probability score threshold level, then the target TCI state is known.

[0026] In some example embodiments, wherein: the apparatus is configured to set at least one of: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level; or the apparatus is configured to receive at least one of: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level from a network entity.

[0027] In some example embodiments, wherein: the apparatus is configured to report at least one of the following to a network entity: a monitoring accuracy metric, a model performance, a model confidence metric, or a probability score metric; and the apparatus is configured to use at least one of the following as a current monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric for comparison with a corresponding monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level: the reported monitoring accuracy metric, the model performance metric, the model confidence metric, or the probability score metric.

[0028] In some example embodiments, wherein: the apparatus is configured to receive at least one of the following items from a network entity: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric; and the apparatus is configured to use the currently received monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric as a current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level.

[0029] In some example embodiments, wherein prior to switching the beam based on the prediction, the means for measuring is further configured to: measure one or more reference signals in a second set of configured sets of reference signal resources associated with the at least one predicted beam.

[0030] In some example embodiments, measuring one or more reference signals in a second set of configured reference signal resource sets includes measuring a reference signal associated with at least one predicted beam or a reference signal associated with a beam received from a source quasi-co-located with a source of the reference signal associated with at least one predicted beam.

[0031] In some example embodiments, prediction-based beam switching is performed based on whether the target TCI state is known based on measuring at least one reference signal associated with at least one predicted beam or a reference signal associated with a beam received from a source quasi-co-located with a source of the reference signal associated with at least one predicted beam.

[0032] In some example embodiments, wherein: the measurement of the at least one measured reference signal is an L1 physical layer measurement; and as an option, the L1 physical layer measurement is a reference signal received power RSRP measurement.

[0033] In some example embodiments, the one or more conditions further include: if an indication or activation of the target TCI state is received within a predefined time period after completing reporting to the network entity data representing a second set of beams including at least one predicted beam, the target TCI state is known, and the reference signal associated with the at least one predicted beam corresponds to the reference signal associated with the target TCI state.

[0034] In some example embodiments, wherein: the predefined time period is less than a time period for performing inference operations; or the predefined time period is less than a time period since a most recent inference operation; or the predefined time period is greater than a time period for non-predictive TCI switching.

[0035] In some example embodiments, where one or more conditions further include: if an indication or activation of the target TCI state is received within X milliseconds, seconds, time slots, radio frames, or symbols after the device completes reporting to the network entity data representing a second set of beams including an indication of at least one predicted beam, then the target TCI state is known.

[0036] In some example embodiments, the one or more conditions further include: the target TCI state is known if the device has performed at least one measurement on a reference signal corresponding to a reference signal associated with the target TCI state or a reference signal received from a source quasi-co-located with the source of the reference signal associated with the target TCI state within N milliseconds, seconds, time slots, radio frames, or symbols before receiving an indication that the target TCI state is indicated or activated.

[0037] In some example embodiments, where the apparatus includes means for performing prediction-based beam reporting or prior to prediction-based beam switching, the means for measuring is further configured to: trigger measurement of at least one reference signal resource associated with at least one predicted beam after reporting data representing an indication of one or more predicted beams, wherein the measurement of at least one reference signal resource associated with at least one predicted beam is performed based on one or more of the following group:

[0038] Measuring the reference signal resource associated with at least one predicted beam in the first time slot of a specific reference signal resource that occurs after the predicted beam associated with the reference signal resource is reported; or measuring the reference signal resource associated with at least one predicted beam within a predetermined or specified time period after the predicted beam associated with the reference signal resource is reported.

[0039] In some example embodiments, the time period is in the range of milliseconds, seconds, time slots, radio frames, or symbols.

[0040] In some example embodiments, the apparatus comprises a user terminal.

[0041] A second aspect provides a method, comprising: measuring, by a terminal device, a first reference signal resource set in a configured reference signal resource set for a corresponding beam of a network entity for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; receiving an indication from the network entity that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; determining whether the target TCI state is known or unknown based on at least one of the following items: at least one predicted beam and the measured first reference signal resource set; and switching to the target TCI state within a time period based on the determination indicating that the target TCI state is known.

[0042] In some example embodiments, the configured reference signal resource set is configured by a network entity.

[0043] In some example embodiments, the method of the second aspect further comprises: receiving a message from the network entity for triggering the terminal device to perform prediction-based beam reporting or prediction-based beam switching, wherein the measurement is performed upon or after receiving the message.

[0044] In some example embodiments, the method of the second aspect, wherein the measuring further comprises: measuring a first set of reference signal resources for a first set of beams for determining at least one predicted beam from a second set of beams using a model, wherein the model performs an inference operation for providing data representing an indication of the at least one predicted beam.

[0045] In some example embodiments, the method of the second aspect further comprises: inputting data representing the measured first set of reference signal resources into a model; and reporting data representing an indication of at least one predicted beam to a network entity, wherein at least one reference signal resource in the configured set of reference signal resources is associated with the at least one predicted beam.

[0046] In some example embodiments, the method of the second aspect further comprises: reporting the measured first reference signal resource set to a network entity for input into the model by the network entity. In some example embodiments, the method of the second aspect further comprises: receiving data representing an indication of at least one predicted beam from the network entity.

[0047] In some example embodiments, the method of the second aspect further includes: sending measurements of a first set of reference signal resources for a first group of beams from a terminal device to a network entity for determining at least one predicted beam from a second group of beams using a model, wherein the model performs an inference operation for providing data representing an indication of at least one predicted beam; and receiving an indication of at least one predicted beam from the network entity for determining by the terminal device whether the target TCI state is known or unknown and switching to the target TCI state based on the determination based on one or more conditions associated with at least one of the following items: at least the predicted beam and the measured first set of reference signal resources.

[0048] In some example embodiments, the method of the second aspect, wherein determining whether the target TCI state is known or unknown further includes: determining that one or more predicted quality metrics of one or more reference signal resources associated with the corresponding reporting prediction beam are greater than or equal to a preconfigured or predefined quality threshold level; in response to determining that at least one of the one or more predicted quality metrics of the one or more reference signal resources associated with the corresponding reporting prediction beam is greater than or equal to the preconfigured or predefined quality threshold level, determining that the target TCI state is known when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state.

[0049] In some example embodiments, the method of the second aspect, wherein the quality metric comprises one or more of the following group: reference signal received power (RSRP); L1 physical layer RSRP (L1-RSRP); signal to noise ratio (SNR) of the reference signal resource; or any other quality metric associated with the reference signal resource.

[0050] In some example embodiments, the method of the second aspect, wherein determining whether the target TCI state is known or unknown further includes: determining that one or more current model performance metrics of one or more reported prediction beams corresponding to the reported prediction beams associated with one or more reference signal resources are greater than or equal to a preconfigured or predefined model performance threshold level; in response to determining that at least one current model performance metric of the one or more reported prediction beams corresponding to the reported prediction beams associated with the one or more reference signal resources is greater than or equal to the preconfigured or predefined model performance threshold level, determining that the target TCI state is known when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state.

[0051] In some example embodiments, the method of the second aspect, wherein the model performance metric includes one or more from the following group: a model prediction accuracy metric; a precision and recall metric; an area under the curve receiver operating curve AUC-ROC metric; a model confidence metric; a probability score metric; or any other model performance metric for evaluating the predictive performance of the model.

[0052] In some example embodiments, the method of the second aspect, wherein switching to the target TCI state within the time period based on a determination that the target TCI state is known further comprises one or more of: performing the TCI state switching within the time period according to the known target TCI state, wherein the time period is calculated based on an L1-RSRP measurement time period required to omit L1-RSRP measurements of reference signal resources of the target TCI state; or performing the TCI state switching within the time period according to the known target TCI state, wherein the time period is calculated based on a determination that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0053] In some example embodiments, the method of the second aspect, when the terminal device is configured to perform prediction-based beam reporting or before prediction-based beam switching, the measurement also includes: after reporting data representing an indication of one or more predicted beams, triggering measurement of at least one reference signal resource associated with at least one of the predicted beams, wherein the measurement of at least one reference signal resource associated with at least one of the predicted beams is performed based on one or more of the following groups: measuring the reference signal resource associated with at least one predicted beam in a first time slot of a specific reference signal resource that occurs after reporting the predicted beam associated with the reference signal resource; or measuring the reference signal resource associated with at least one predicted beam within a predetermined or specified time period after reporting the predicted beam associated with the reference signal resource.

[0054] In some example embodiments, the method of the second aspect, wherein after measuring a reference signal associated with at least one predicted beam, prediction-based beam switching is performed based on whether the target TCI state is known.

[0055] In some example embodiments, the method of the second aspect, wherein the measurement of the at least one measured reference signal resource is an L1 physical layer measurement; and / or the L1 physical layer measurement is a reference signal resource received power RSRP measurement.

[0056] In some example embodiments, the method of the second aspect, wherein when the terminal device is configured to perform prediction-based beam reporting, determining whether the target TCI state is known or unknown further includes: determining that a first reference signal resource set is associated with a corresponding reported predicted beam; and determining that the target TCI state is known when the reference signal resources in the first reference signal set correspond to reference signal resources associated with the target TCI state.

[0057] In some example embodiments, the method of the second aspect, wherein determining whether the target TCI state is known or unknown further comprises determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least a predicted beam and a measured first reference signal resource set.

[0058] In some example embodiments, the method of the second aspect further comprises: receiving, from a network entity, a configuration of a reference signal resource set for measuring a corresponding beam for the network entity, wherein the reference signal resource set is associated with prediction-based beam reporting or prediction-based beam switching.

[0059] In some example embodiments, the method of the second aspect, wherein determining whether the target TCI state is known or unknown further comprises determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items and switching to the target TCI state based on the determination: at least a predicted beam and a measured first reference signal resource set.

[0060] In some example embodiments, the method of the second aspect, wherein the one or more conditions further include: if it is determined that at least one measurement reference signal resource in the first reference signal resource set corresponds to: a reference signal included in or associated with the target TCI state; or a reference signal received from a source that is quasi-co-located with the source of the reference signal included in or associated with the target TCI state; or a reference signal corresponding to at least one reference signal included in or associated with the target TCI state and having a quality metric that is higher than a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0061] In some example embodiments, the method of the second aspect, wherein the one or more conditions further include: if a predicted quality metric of a reference signal corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0062] In some example embodiments, the method of the second aspect further includes: performing monitoring of prediction accuracy, wherein the monitoring of prediction accuracy is based on measurement of one or more reference signal resources corresponding to one or more previous prediction beams; and determining a prediction accuracy metric for at least one prediction beam based on said monitoring of the previous prediction beam and an indication of the corresponding at least one prediction beam; wherein the one or more conditions further include: if the prediction accuracy metric for at least one prediction beam is greater than or equal to a prediction accuracy threshold level, the target TCI state is known.

[0063] In some example embodiments, the method of the second aspect, wherein the one or more conditions further include: if a current model performance metric corresponding to a model inference operation of at least one predicted beam having a reference signal included in or associated with the target TCI state is greater than or equal to a model performance threshold level, then the target TCI state is known.

[0064] In some example embodiments, the method of the second aspect, wherein one or more conditions further include: if a current model confidence or probability score metric corresponding to a model inference operation of at least one predicted beam having a reference signal included in or associated with the target TCI state is greater than or equal to a corresponding confidence threshold level or probability score threshold level, then the target TCI state is known.

[0065] In some example embodiments, the method of the second aspect further includes setting at least one of the following items: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level; or receiving at least one of the following items from a network entity: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level.

[0066] In some example embodiments, the method of the second aspect further includes reporting at least one of the following items to a network entity: a monitoring accuracy metric, a model performance, a model confidence metric, or a probability score metric; and using at least one of the following items as a current monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric for comparison with a corresponding monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level: the reported monitoring accuracy metric, the model performance metric, the model confidence metric, or the probability score metric.

[0067] In some example embodiments, the method of the second aspect further includes receiving at least one of the following items from a network entity: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric; and using the currently received monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric as the current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level.

[0068] In some example embodiments, wherein prior to switching the beam based on the prediction, the means for measuring is further configured to: measure one or more reference signals in a second set of configured sets of reference signal resources associated with the at least one predicted beam.

[0069] In some example embodiments, the method of the second aspect, wherein measuring one or more reference signals in a second set of configured reference signal resource sets includes: measuring a reference signal associated with at least one predicted beam or a reference signal associated with a beam received from a source quasi-co-located with a source of the reference signal associated with at least one predicted beam.

[0070] In some example embodiments, the method of the second aspect, wherein: prediction-based beam switching is performed based on whether the target TCI state is known based on measuring at least one reference signal associated with at least one predicted beam or a reference signal associated with a beam received from a source quasi-co-located with the source of the reference signal associated with at least one predicted beam.

[0071] In some example embodiments, the method of the second aspect, wherein the one or more conditions further include: if an indication or activation of the target TCI state is received within a predefined time period after completing reporting to the network entity data representing a second set of beams including at least one predicted beam, the target TCI state is known, and the reference signal associated with the at least one predicted beam corresponds to the reference signal associated with the target TCI state.

[0072] In some example embodiments, the method of the second aspect, wherein: the predefined time period is less than the time period for performing inference operations; or the predefined time period is less than the time period since the most recent inference operation; or the predefined time period is greater than the time period for non-predictive TCI switching.

[0073] In some example embodiments, the method of the second aspect, wherein the one or more conditions further include: if an indication or activation of the target TCI state is received within X milliseconds, seconds, time slots, radio frames, or symbols after the device completes reporting to the network entity data representing a second set of beams including an indication of at least one predicted beam, then the target TCI state is known.

[0074] In some example embodiments, the method of the second aspect, wherein the one or more conditions further include: if the device has performed at least one measurement on a reference signal corresponding to a reference signal associated with the target TCI state or a reference signal received from a source quasi-co-located with the source of the reference signal associated with the target TCI state within N milliseconds, seconds, time slots, radio frames, or symbols before receiving an indication that the target TCI state is indicated or activated, then the target TCI state is known.

[0075] In some example embodiments, the method of the second aspect, wherein the time period is in the range of milliseconds, seconds, time slots, radio frames, or symbols.

[0076] In some example embodiments, the method of the second aspect, wherein the method may be performed by an apparatus comprising a user terminal. As an option, the method may be performed at the user terminal.

[0077] A third aspect provides a computer program comprising an instruction set which, when executed on an apparatus, is configured to cause the apparatus to perform a method comprising: measuring a first set of reference signal resource sets of a configured reference signal resource set for a corresponding beam of a network entity for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; receiving an indication from the network entity that a target transmission configuration indicator (TCI) state corresponding to at least one predicted beam is to be activated or indicated; determining whether the target TCI state is known or unknown based on at least one of the following items: at least one predicted beam and the measured first set of reference signal resources; and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0078] In some example embodiments, the third aspect may include any other features mentioned with respect to the method of the second aspect.

[0079] 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: measuring a first set of reference signal resource sets of a configured reference signal resource set for a corresponding beam of a network entity for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; receiving an indication from the network entity that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; determining whether the target TCI state is known or unknown based on at least one of the following items: at least one predicted beam and the measured first set of reference signal resources; and switching to the target TCI state within a time period based on the determination indicating that the target TCI state is known.

[0080] In some example embodiments, the fourth aspect may include any other features mentioned with respect to the method of the second aspect.

[0081] A fifth aspect of the present invention provides a device having at least one processor and at least one memory, the at least one memory having computer-readable code stored thereon, the computer-readable code controlling the at least one processor when executed to: measure a first reference signal resource set of a configured reference signal resource set for a corresponding beam of a network entity for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; receive an indication from the network entity that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; determine whether the target TCI state is known or unknown based on at least one of the following items: at least one predicted beam and the measured first reference signal resource set; and switch to the target TCI state within a time period based on the determination indicating that the target TCI state is known.

[0082] In some example embodiments, the fifth aspect may include any other features mentioned with respect to the method of the second aspect.

[0083] The sixth aspect provides a network device, comprising: a component for sending a configuration of a first reference signal resource set for measuring a corresponding beam for the network device to determine at least one predicted beam to a terminal device, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; and a component for sending an indication to the terminal device that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; wherein the target TCI state, whether known or unknown, is determined based on at least one of the following items: at least one predicted beam and a first reference signal resource set, and is switched to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0084] In some example embodiments, the network apparatus further comprises configuring the configured set of reference signal resources.

[0085] In some example embodiments, the network device further comprises: means for sending a message to the terminal device for triggering the terminal device to perform prediction-based beam reporting or prediction-based beam switching, wherein the measurement is performed upon or after receiving the message.

[0086] In some example embodiments, the network device further comprises means for receiving, from the terminal device, an indication of at least one predicted beam generated by the terminal device inputting measurements of the first set of reference signal resources into a model, wherein the model performs an inference operation to provide data representative of the indication of the at least one predicted beam.

[0087] In some example embodiments, the network device further comprises: means for receiving measurements of a first set of reference signal resources for a first group of beams from a terminal device for determining at least one predicted beam from a second group of beams using a model, wherein the model performs an inference operation for providing data representing an indication of the at least one predicted beam; and means for sending an indication of the at least one predicted beam to the terminal device for determining, by the terminal device, whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least the predicted beam and the measured first set of reference signal resources and / or the configured first set of reference signal resources, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0088] In some example embodiments, the network apparatus further comprises: means for inputting data representing the measured first set of reference signal resources to the model; and means for reporting data representing an indication of the at least one predicted beam to the terminal apparatus.

[0089] In some example embodiments, the network apparatus further comprises means for receiving a first set of reference signal resources of the configuration measurements from the terminal apparatus for input to the model at the network apparatus.

[0090] In some example embodiments, the network apparatus further comprises means for determining whether the target TCI state is known or unknown based on at least one of: at least one predicted beam and a configured first set of reference signal resources, and switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known.

[0091] In some example embodiments, the network device further comprises means for determining whether the target TCI state is known or unknown based on at least one of: at least one predicted beam and a measured set of first reference signal resources received from the terminal device, and switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known.

[0092] In some example embodiments, the component for determining whether the target TCI state is known or unknown further includes: a component for determining that one or more predicted quality metrics of one or more reference signal resources associated with the corresponding predicted beam are greater than or equal to a preconfigured or predefined quality threshold level; in response to determining that at least one of the one or more predicted quality metrics of the one or more reference signal resources associated with the corresponding reported predicted beam is greater than or equal to the preconfigured or predefined quality threshold level, determining that the target TCI state is known when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state.

[0093] In some example embodiments, the quality metric comprises one or more of the following group: reference signal received power (RSRP); L1 physical layer RSRP (L1-RSRP); signal to noise ratio (SNR) of the reference signal resource; or any other quality metric associated with the reference signal resource.

[0094] In some example embodiments, the network apparatus further comprises means for switching to the target TCI state within the time period based on the determination that the target TCI state is known.

[0095] In some example embodiments, the network apparatus, wherein the means for switching further comprises: means for performing the TCI state switching within the time period based on a known target TCI state, wherein the time period is calculated based on an L1-RSRP measurement time period required to omit L1-RSRP measurements of reference signal resources of the target TCI state; or means for performing the TCI state switching within the time period based on a known target TCI state, wherein the time period is calculated based on a determination that at least one L1-RSRP measurement or measurement sample related to reference signal resources of the target TCI state has been obtained.

[0096] In some example embodiments, the measurement of the reference signal resource of the at least one measurement of the first set of reference signal resources is an L1 physical layer measurement; and / or the L1 physical layer measurement is a reference signal resource received power (RSRP) measurement.

[0097] In some example embodiments, where the terminal device is configured to perform prediction-based beam reporting, the means for determining whether the target TCI state is known or unknown further comprises: means for determining that a reference signal resource in a first reference signal resource set is associated with a corresponding predicted beam or a reported predicted beam; and means for determining that the target TCI state is known when the reference signal resource in the first reference signal set corresponds to a reference signal resource associated with the target TCI state.

[0098] In some example embodiments, the network device includes components for and / or is configured to: determine whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items and switch to the target TCI state based on the determination: at least a predicted beam and a first set of reference signal resources.

[0099] In some example embodiments, the network device further includes means for determining whether the target TCI state is known or unknown, further including determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items and switching to the target TCI state within a time period based on the determination indicating that the target TCI state is known.

[0100] In some example embodiments, the network device, the terminal device is configured to determine whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items and switch to the target TCI state based on the determination: at least one predicted beam and the measured first reference signal resource set.

[0101] In some example embodiments, the network device further includes a component for determining whether the target TCI state is known or unknown, further including determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least one predicted beam and a measured first reference signal resource set received from the terminal device, and determining to switch to the target TCI state within the time period based on an indication that the target TCI state is known.

[0102] In some example embodiments, the component for determining whether the target TCI state is known or unknown further includes: a component for determining that one or more current model performance metrics corresponding to one or more predicted beams associated with one or more reference signal resources are greater than or equal to a preconfigured or predefined model performance threshold level; in response to determining that at least one current model performance metric of the one or more reported predicted beams associated with one or more reference signal resources is greater than or equal to a preconfigured or predefined model performance threshold level, determining that the target TCI state is known when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state.

[0103] In some example embodiments, the model performance metric includes one or more of the following groups: a model prediction accuracy metric; a precision and recall metric; an area under the curve receiver operating curve (AUC-ROC) metric; a model confidence metric; a probability score metric; or any other model performance metric used to evaluate the predictive performance of the model.

[0104] In some example embodiments, a network device, wherein: the network device includes components for calculating and / or is configured to calculate at least one of a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric associated with a model and at least one prediction beam for determining whether a target TCI state is known. For example, the network device may be configured to determine whether the target TCI state is known or potentially known to or by a terminal device.

[0105] In some example embodiments, a network device, wherein: the network device includes means for calculating and / or is configured to calculate at least one of a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric associated with a model and at least one prediction beam for use in determining by a terminal device whether a target TCI state is known.

[0106] In some example embodiments, a network device, wherein: the network device is configured to use a calculated monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric associated with a model and at least one prediction beam as a current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric when compared to a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level for determining whether at least one prediction beam for a target TCI state and / or the target TCI state is known or unknown.

[0107] In some example embodiments, the network device further includes: a component for sending a message to the terminal device for triggering the terminal device to perform prediction-based beam reporting or prediction-based beam switching, wherein the measurement is performed when or after receiving the message.

[0108] In some example embodiments, measurements of one or more reference signal resources corresponding to one or more previously predicted beams are performed by the terminal device and reported accordingly to the network device.

[0109] In some example embodiments, a network device, wherein the network device further comprises or is configured to: perform monitoring of prediction accuracy, wherein the monitoring of prediction accuracy is based on reported measurements of one or more reference signal resources corresponding to one or more previously predicted beams; and determine a prediction accuracy metric for at least one predicted beam based on said monitoring of the previously predicted beams and an indication of the corresponding at least one predicted beam; and a component for sending data representing the prediction accuracy metric for the at least one predicted beam to a terminal device for determining, by the terminal device, whether the target TCI state is known or unknown based on comparing whether the prediction accuracy metric for the at least one predicted beam is greater than or equal to a prediction accuracy threshold level.

[0110] In some example embodiments, a network device, wherein: the network device is configured to set at least one of the following items: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level; or the network device is configured to receive at least one of the following items from a terminal device: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level.

[0111] In some example embodiments, the network device, wherein: the network device is configured to report at least one of the following to the terminal device: a monitoring accuracy metric, a model performance, a model confidence metric, or a probability score metric.

[0112] In some example embodiments, the one or more conditions further include: if it is determined that at least one measurement reference signal resource in the first reference signal resource set corresponds to: a reference signal included in or associated with the target TCI state; or a reference signal received from a source that is quasi-co-located with the source of the reference signal included in or associated with the target TCI state; or a reference signal corresponding to at least one reference signal included in or associated with the target TCI state and having a quality metric that is higher than a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0113] In some example embodiments, the one or more conditions further include: if a predicted quality metric of a reference signal corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0114] In some example embodiments, the one or more conditions further include: if a current model performance metric of a model inference operation corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a model performance threshold level, then the target TCI state is known.

[0115] In some example embodiments, the one or more conditions further include: if a current model confidence or probability score metric of a model inference operation corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a corresponding confidence threshold level or probability score threshold level, then the target TCI state is known.

[0116] In some example embodiments, wherein: the network device is configured to report at least one of the following items to the terminal device: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric; and the network device is configured to use at least one of the following items as a current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level: the reported monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric.

[0117] In some example embodiments, wherein: the network device is configured to receive at least one of the following items from the terminal device: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric; and the network device is configured to use the currently received monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric as the current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level.

[0118] In some example embodiments, wherein, prior to switching the beam based on the prediction, means for receiving, from the terminal device, a measurement of a second set of reference signal resources of the configured sets of reference signal resources associated with the at least one predicted beam.

[0119] In some example embodiments, the means for receiving a second set of configured reference signal resource sets from a terminal device comprises receiving measurements of a reference signal associated with at least one predicted beam or measurements of a reference signal associated with a beam received from a source quasi-co-located with a source of the reference signal associated with at least one predicted beam.

[0120] In some example embodiments, prediction-based beam switching is performed based on whether the target TCI state is known based on received measurements of at least one reference signal associated with at least one predicted beam or received measurements of a reference signal associated with a beam received from a source quasi-co-located with the source of the reference signal associated with at least one predicted beam.

[0121] In some example embodiments, wherein: the at least one measurement of the reference signal is an L1 physical layer measurement; and, as an option, the L1 physical layer measurement is a reference signal received power (RSRP) measurement.

[0122] In some example embodiments, the one or more conditions further include: if an indication or activation of the target TCI state is received by the terminal device within a predefined time period after the terminal device completes reporting data representing the second set of beams including the at least one predicted beam to the network device, the target TCI state is known, and the reference signal associated with the at least one predicted beam corresponds to the reference signal associated with the target TCI state.

[0123] In some example embodiments, wherein: the predefined time period is less than a time period for performing inference operations; or the predefined time period is less than a time period since a most recent inference operation; or the predefined time period is greater than a time period for non-predictive TCI switching.

[0124] In some example embodiments, where one or more of the conditions further include: if an indication or activation of the target TCI state is received by the terminal device within X milliseconds, seconds, time slots, radio frames or symbols after the terminal device completes reporting data representing the second set of beams (including an indication of at least one predicted beam) to the network entity, then the target TCI state is known.

[0125] In some example embodiments, one or more of the conditions further include: the target TCI state is known if the terminal device has performed at least one measurement on a reference signal corresponding to a reference signal associated with the target TCI state or a reference signal received from a source quasi-co-located with the source of the reference signal associated with the target TCI state within N milliseconds, seconds, time slots, radio frames or symbols before receiving an indication that the target TCI state is indicated or activated.

[0126] In some example embodiments, the network device of the sixth aspect, wherein the network device comprises a network entity.

[0127] The seventh aspect provides a method, comprising: sending a configuration of a first reference signal resource set for measuring a corresponding beam for a network device to a terminal device for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; and sending an indication to the terminal device that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; wherein the target TCI state, whether known or unknown, is determined based on at least one of the following items: at least one predicted beam and a first reference signal resource set, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0128] In some example embodiments, the method of the seventh aspect further comprises configuring the configured reference signal resource set.

[0129] In some example embodiments, the method of the seventh aspect further comprises: sending a message to the terminal device for triggering the terminal device to perform prediction-based beam reporting or prediction-based beam switching, wherein the measurement is performed upon receipt of the message or after the point-to-point message is received.

[0130] In some example embodiments, the method of aspect seven further comprises receiving, from a terminal device, an indication of at least one predicted beam generated by the terminal device inputting measurements of a first set of reference signal resources into a model, wherein the model performs an inference operation to provide data representing an indication of at least one predicted beam.

[0131] In some example embodiments, the method of the seventh aspect further includes: receiving measurements of a first reference signal resource set for a first group of beams from a terminal device for determining at least one predicted beam from a second group of beams using a model, wherein the model performs an inference operation for providing data representing an indication of at least one predicted beam; and sending an indication of at least one predicted beam to the terminal device for determining by the terminal device whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least one predicted beam and the measured first reference signal resource set and / or the configured first reference signal resource set, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0132] In some example embodiments, the method of the seventh aspect further comprises: inputting data representing the measured set of first reference signal resources into the model; and reporting data representing an indication of the at least one predicted beam to the terminal device.

[0133] In some example embodiments, the method of the seventh aspect further comprises: receiving a measured first set of reference signal resources of the configured reference signal resources from the terminal device for input to the model at the network device.

[0134] In some example embodiments, the method of the seventh aspect further includes determining whether the target TCI state is known or unknown based on at least one of the following items: at least one predicted beam and a configured first reference signal resource set, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0135] In some example embodiments, the method of the seventh aspect further includes determining whether the target TCI state is known or unknown based on at least one of: at least one predicted beam and a measured set of first reference signal resources received from the terminal device, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0136] In some example embodiments, the method of aspect seven, wherein the determining whether the target TCI state is known or unknown further includes: determining that one or more predicted quality metrics of one or more reference signal resources associated with the corresponding predicted beam are greater than or equal to a preconfigured or predefined quality threshold level; in response to determining that at least one of the one or more predicted quality metrics of the one or more reference signal resources associated with the corresponding reported predicted beam is greater than or equal to a preconfigured or predefined quality threshold level, when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state, determining that the target TCI state is known.

[0137] In some example embodiments, the method of the seventh aspect, wherein the quality metric includes one or more of the following group: reference signal received power RSRP; L1 physical layer RSRP L1-RSRP; signal to noise ratio SNR of the reference signal resource; any other quality metric associated with the reference signal resource.

[0138] In some example embodiments, the method of the seventh aspect further comprises: switching to the target TCI state within the time period based on a determination that the target TCI state is known.

[0139] In some example embodiments, the method of the seventh aspect, wherein the switching further comprises: performing TCI state switching within the time period according to a known target TCI state, wherein the time period is calculated based on an L1-RSRP measurement time period required to omit L1-RSRP measurement of reference signal resources of the target TCI state; or performing TCI state switching within the time period according to a known target TCI state, wherein the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample related to the reference signal resources of the target TCI state has been obtained.

[0140] In some example embodiments, the method of the seventh aspect, wherein the measurement of the reference signal resource of at least one measurement of the first reference signal resource set is an L1 physical layer measurement; and / or the L1 physical layer measurement is a reference signal resource received power RSRP measurement.

[0141] In some example embodiments, where the terminal device is configured to perform prediction-based beam reporting, determining whether the target TCI state is known or unknown further comprises: determining that a reference signal resource in a first reference signal resource set is associated with a corresponding predicted beam or a reported predicted beam; and determining that the target TCI state is known when the reference signal resource in the first reference signal set corresponds to a reference signal resource associated with the target TCI state.

[0142] In some example embodiments, the method of the seventh aspect further includes determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items and switching to the target TCI state based on the determination: at least a predicted beam and a first reference signal resource set.

[0143] In some example embodiments, the method of aspect seven, wherein the determining whether the target TCI state is known or unknown further comprises determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least one predicted beam and a configured first reference signal resource set, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0144] In some example embodiments, the method of aspect seven, a network device, wherein the terminal device is configured to determine whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items and switch to the target TCI state based on the determination: at least one predicted beam and a measured first reference signal resource set.

[0145] In some example embodiments, the method of aspect seven, wherein the determining whether the target TCI state is known or unknown further comprises determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least one predicted beam and a measured first reference signal resource set received from the terminal device, and determining to switch to the target TCI state within the time period based on an indication that the target TCI state is known.

[0146] In some example embodiments, the method of aspect seven, wherein the determining whether the target TCI state is known or unknown further includes: determining that one or more current model performance metrics corresponding to one or more predicted beams in the predicted beams associated with one or more reference signal resources are greater than or equal to a preconfigured or predefined model performance threshold level; in response to determining that at least one current model performance metric of the one or more reported predicted beams in the reported predicted beams associated with one or more reference signal resources is greater than or equal to a preconfigured or predefined model performance threshold level, when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state, determining that the target TCI state is known.

[0147] In some example embodiments, the method of the seventh aspect, wherein the model performance metric includes one or more of the following groups: a model prediction accuracy metric; a precision and recall metric; an area under the curve receiver operating curve AUC-ROC metric; a model confidence metric; a probability score metric; or any other model performance metric for evaluating the predictive performance of the model.

[0148] In some example embodiments, the method of the seventh aspect further includes calculating at least one of a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric associated with the model and at least one prediction beam for determining whether the target TCI state is known. For example, the method of the seventh aspect may further include determining whether the target TCI state is known or potentially known to or by the terminal device.

[0149] In some example embodiments, the method of the seventh aspect also includes calculating and / or being configured to calculate at least one of a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric associated with the model and at least one prediction beam for determining by the terminal device whether the target TCI state is known.

[0150] In some example embodiments, the method of the seventh aspect also includes using a calculated monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric associated with the model and at least one prediction beam as a current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric when compared to a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level for determining whether at least one prediction beam and / or target TCI state for the target TCI state is known or unknown.

[0151] In some example embodiments, the method of the seventh aspect further includes: sending a message to the terminal device for triggering the terminal device to perform prediction-based beam reporting or prediction-based beam switching, wherein the measurement is performed when or after the message is received.

[0152] In some example embodiments, the method of the seventh aspect, wherein measurements of one or more reference signal resources corresponding to one or more previously predicted beams are performed by the terminal device and reported accordingly to the network device.

[0153] In some example embodiments, the method of the seventh aspect further includes: performing monitoring of prediction accuracy, wherein the monitoring of prediction accuracy is based on reported measurements of one or more reference signal resources corresponding to one or more previously predicted beams; and determining a prediction accuracy metric for at least one predicted beam based on the monitoring of the previously predicted beams and an indication of the corresponding at least one predicted beam; and sending data representing the prediction accuracy metric for the at least one predicted beam to a terminal device for use by the terminal device in determining whether the target TCI state is known or unknown based on comparing whether the prediction accuracy metric for at least one predicted beam is greater than or equal to a prediction accuracy threshold level.

[0154] In some example embodiments, the method of the seventh aspect further includes setting at least one of the following items: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level; or receiving at least one of the following items from a terminal device: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level.

[0155] In some example embodiments, the method of the seventh aspect further comprises reporting to the terminal device at least one of: a monitoring accuracy metric, a model performance, a model confidence metric, or a probability score metric.

[0156] In some example embodiments, the method of the seventh aspect, wherein the one or more conditions further include: if it is determined that at least one measurement reference signal resource in the first reference signal resource set corresponds to: a reference signal included in or associated with the target TCI state; or a reference signal received from a source that is quasi-co-located with the source of the reference signal included in or associated with the target TCI state; or a reference signal corresponding to at least one reference signal included in or associated with the target TCI state and having a quality metric that is higher than a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0157] In some example embodiments, the method of aspect seven, wherein the one or more conditions further include: if a predicted quality metric of a reference signal corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a preconfigured or predefined quality threshold level, then the target TCI state is known.

[0158] In some example embodiments, the method of aspect seven, wherein one or more conditions further include: if a current model performance metric of a model inference operation corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a model performance threshold level, then the target TCI state is known.

[0159] In some example embodiments, the method of aspect seven, wherein one or more conditions further include: if a current model confidence or probability score metric of a model inference operation corresponding to at least one predicted beam of a reference signal included in or associated with the target TCI state is greater than or equal to a corresponding confidence threshold level or probability score threshold level, then the target TCI state is known.

[0160] In some example embodiments, the method of the seventh aspect also includes reporting at least one of the following items to the terminal device: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric; and using at least one of the following items by the network device as the current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with a corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level: the reported monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric.

[0161] In some example embodiments, the method of the seventh aspect also includes receiving at least one of the following items from the terminal device: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric; and using the currently received monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric by the network device as the current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with the corresponding monitoring accuracy threshold level, model performance threshold level, confidence threshold level, or probability score threshold level.

[0162] In some example embodiments, the method of the seventh aspect, wherein, prior to the predicted-based beam switching, the method further comprises receiving, from the terminal device, a measurement of a second set of reference signal resources in the configured set of reference signal resources associated with at least one predicted beam.

[0163] In some example embodiments, the method of aspect seven, wherein the second set of reference signal resource sets configured for reception from the terminal device further comprises: receiving measurements of reference signals associated with at least one predicted beam or measurements of reference signals associated with a beam received from a source quasi-co-located with a source of the reference signal associated with at least one predicted beam.

[0164] In some example embodiments, the method of aspect seven, wherein: based on reception measurements of at least one reference signal associated with at least one predicted beam or reception measurements of a reference signal associated with a beam received from a source quasi-co-located with the source of the reference signal associated with at least one predicted beam, prediction-based beam switching is performed based on whether the target TCI state is known.

[0165] In some example embodiments, the method of the seventh aspect, wherein: the measurement of at least one measurement reference signal is an L1 physical layer measurement; and, as an option, the L1 physical layer measurement is a reference signal received power RSRP measurement.

[0166] In some example embodiments, the method of aspect seven, wherein the one or more conditions further include: if an indication or activation of the target TCI state is received by the terminal device within a predefined time period after the terminal device completes reporting data representing a second set of beams including at least one predicted beam to the network device, then the target TCI state is known and the reference signal associated with the at least one predicted beam corresponds to the reference signal associated with the target TCI state.

[0167] In some example embodiments, the method of the seventh aspect, wherein: the predefined time period is less than the time period used to perform the inference operation; or the predefined time period is less than the time period since the most recent inference operation; or the predefined time period is greater than the time period used for non-predictive TCI switching.

[0168] In some example embodiments, the method of aspect seven, wherein the one or more conditions further include: if an indication or activation of the target TCI state is received by the terminal device within X milliseconds, seconds, time slots, radio frames, or symbols after the terminal device completes reporting data representing the second set of beams (including an indication of at least one predicted beam) to the network entity, then the target TCI state is known.

[0169] In some example embodiments, the method of aspect seven, wherein one or more conditions further include: if the terminal device has performed at least one measurement on a reference signal corresponding to a reference signal associated with the target TCI state or a reference signal received from a source quasi-co-located with the source of the reference signal associated with the target TCI state within N milliseconds, seconds, time slots, radio frames, or symbols before receiving an indication that the target TCI state is indicated or activated, then the target TCI state is known.

[0170] In some example embodiments, the network device of the seventh aspect, wherein the network device comprises a network entity.

[0171] The eighth aspect provides a computer program comprising an instruction set, which, when executed on a device, is configured to cause the device to perform a method comprising: sending a configuration of a first reference signal resource set for measuring a corresponding beam for a network device to a terminal device for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; and sending an indication to the terminal device that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; wherein the target TCI state, whether known or unknown, is determined based on at least one of the following items: at least one predicted beam and a first reference signal resource set, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0172] In some example embodiments, the eighth aspect may include any other features mentioned with respect to the method of the seventh aspect.

[0173] 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: sending a configuration of a first reference signal resource set for measuring a corresponding beam for a network device to a terminal device for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; and sending an indication to the terminal device that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; wherein the target TCI state, whether known or unknown, is determined based on at least one of the following items: at least one predicted beam and a first reference signal resource set, and switching to the target TCI state in a time period based on a determination indicating that the target TCI state is known.

[0174] In some example embodiments, the ninth aspect may include any other features mentioned with respect to the method of the seventh aspect.

[0175] The tenth aspect of the present invention provides a device having at least one processor and at least one memory, on which computer-readable code is stored, which controls the at least one processor when the computer-readable code is executed: sending a configuration of a first reference signal resource set for measuring a corresponding beam for a network device to a terminal device for determining at least one predicted beam, wherein the configured reference signal resource set is associated with a prediction-based beam report or a prediction-based beam switching; and sending an indication to the terminal device that a target transmission configuration indicator TCI state corresponding to at least one predicted beam is to be activated or indicated; wherein the target TCI state, whether known or unknown, is determined based on at least one of the following items: at least one predicted beam and a first reference signal resource set, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

[0176] In some example embodiments, the tenth aspect may include any other features mentioned with respect to the method of the seventh aspect.

[0177] An eleventh aspect of the present invention provides a computer readable medium comprising instruction codes stored thereon, which, when executed by one or more processors, cause the one or more processors to perform the method according to the second aspect.

[0178] A twelfth aspect of the present invention provides a computer-readable medium comprising instruction codes stored thereon, which, when executed by one or more processors, cause the one or more processors to perform the method according to the seventh aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0179] The invention will now be described by way of non-limiting example with reference to the accompanying drawings, in which:

[0180] Figure 1 A network scenario including a network entity and a user terminal is shown;

[0181] Figure 2 is a flowchart illustrating operations that may be performed at a user terminal according to one or more example embodiments;

[0182] Figure 3 is a diagram illustrating a method according to one or more example embodiments Figure 2 a schematic diagram showing an example of how the operation is performed;

[0183] Figure 4a is a signaling and processing diagram according to one or more example embodiments where beam prediction is performed at a user terminal;

[0184] Figure 4b is a diagram showing a method for performing a multi-processor circuit according to one or more example embodiments. Figure 4aA flowchart of operations performed at a network entity;

[0185] Figure 5a is a signaling and flow diagram according to one or more example embodiments where beam prediction is performed at a network entity;

[0186] Figure 5b is a diagram showing a method for performing a multi-processor circuit according to one or more example embodiments. Figure 5a A flowchart of operations performed at a network entity;

[0187] Figure 6a is a flowchart illustrating operations performed at a user terminal according to one or more other example embodiments;

[0188] Figure 6b is a flowchart illustrating operations performed at a user terminal according to one or more other example embodiments;

[0189] Figure 6c is a flowchart illustrating operations performed at a user terminal according to one or more other example embodiments;

[0190] Figure 6d is a flow chart illustrating operations performed at a network entity according to one or more other example embodiments;

[0191] Figure 7 illustrates an apparatus that may be configured according to one or more example embodiments; and

[0192] Figure 8 Non-transitory computer-readable media that can be configured in accordance with one or more example embodiments are shown. DETAILED DESCRIPTION

[0193] Example embodiments relate to beam prediction, and in particular, apparatus and methods for determining a known or unknown state of a target transmission configuration indicator (TCI) state corresponding to an indication of at least one predicted beam.

[0194] With respect to 5G NR, massive multiple-input multiple-output (MIMO) and beamforming technologies are used to provide enhanced throughput and capacity. Massive MIMO employs multiple antenna arrays and spatial multiplexing to transmit independent and separate coded data signals or data streams. When employed in a network entity that may comprise part of a radio access network (RAN), this may allow simultaneous communication with multiple user terminals on the same time and frequency resources. Beamforming is used together with massive MIMO technology to provide directional beams for transmitting data streams to respective user terminals with good connectivity and reduced interference. Beam management may refer to the process by which a particular device (such as a network entity) determines which beam(s) to use for a particular user terminal(s). Beam management may involve beamforming, beam selection, and beam switching, with the goal of maintaining good connectivity despite issues such as path loss and / or changes in user terminal position and / or orientation. Beam management is particularly applicable to frequency range 2 (FR2) communications for 5G NR, but is not exclusive to it.

[0195] According to 3GPP Release 17, beam management may employ a so-called unified TCI state framework applicable to both downlink and uplink channels, whereby the beam to be switched to (hereinafter referred to as the target beam) may be indicated to the user terminal by the network entity using an indicated TCI state corresponding to the target beam. The channel may include at least one of a physical downlink shared channel (PDSCH), a physical downlink control channel (PDCCH), a channel state information reference signal (CSI-RS), a physical uplink shared channel (PUSCH), a physical uplink control channel (PUCCH), or a sounding reference signal (SRS) channel. For example, the network entity may indicate a target TCI state to the user terminal, which may include an indication of at least one specific source reference signal (RS) associated with the target beam, which the user terminal may use for subsequent transmission and / or reception hypotheses. The indicated target TCI state may also include quasi-co-location (QCL) information to indicate one or more relationships between a specific RS and one or more other RSs, for example, the quasi-co-location (QCL) information may indicate whether the source (e.g., port) of the specific RS is in QCL with the source of one or more other RSs. If the properties of the channel through which the symbols on one source are transmitted can be inferred from the channel through which the symbols on the other source are transmitted, the two sources are said to be QCL. The quasi co-location relationship is configured by the higher layer parameter qcl-Type1 for the first DL RS and qcl-Type2 for the second DL RS (if configured). For the case of two DL RSs, the QCL type may be different. The quasi co-location type corresponding to each DL RS is given by the higher layer parameter qcl-Type in QCL-Info and can take one of the following values:

[0196] - "typeA": {Doppler shift, Doppler spread, average delay, delay spread} - "typeB": {Doppler shift, Doppler spread}

[0197] - "typeC": {Doppler shift, average delay}

[0198] - "typed": {spatial Rx parameters}

[0199] The network entity may activate the target TCI state and the user terminal may switch to the indicated target TCI state after a TCI state switch delay period associated with the target TCI state, such that both the network entity and the user terminal are synchronized with respect to when they switch to the target beam.

[0200] The TCI state switching delay period may depend on whether the target TCI state is "known" or "unknown" by the user terminal.

[0201] For example, according to 3GPP TS 38.133, the target TCI state may be known if (in one case) the user terminal has sent at least one physical layer (L1) reference signal received power (RSRP) report for an RS included in or associated with the target TCI state before the network node has indicated the target TCI state. If the target TCI state is unknown by the user terminal, the TCI state switching delay period may be longer than if the target TCI state is known.

[0202] Beam prediction that provides one or more target beams has been proposed in 3GPP Release 19. Beam prediction can be performed using a model and therefore methods associated with artificial intelligence (AI) or machine learning (ML) can be employed. Beam prediction can enable faster and / or more accurate beam switching. For example, it is proposed to enable spatial domain downlink beam prediction for beam set A based on measurements of beam set B. Beam set A can be a beam set associated with the configuration of a reference signal resource set. Beam set B can be associated with a first configuration set of reference signal resource sets for measurement, where the first reference signal resource set can be measured for input into the model. In some examples, beam set B (also referred to as a first RS resource set associated with a first group of beams) can include a subset of beam set A (also referred to as a beam set associated with a configured RS resource set). The prediction can be performed using a model provided at a user terminal or a network entity. The model may receive as input measurements associated with beam set B (or a first reference signal set associated with the corresponding beam), wherein the model performs an inference operation to output data representing an indication of at least one predicted beam from beam set A. That is, at least one beam in beam set A may be predicted using the model based on receiving as input measurements associated with beam set B.

[0203] Therefore, after receiving an indication of a target TCI state corresponding to at least one predicted beam from beam set A, the user terminal may not have yet measured the RS associated with the at least one predicted beam (or the RS from the QCL source, where the source of the RS is associated with the at least one predicted beam). Therefore, the indicated target TCI state may be unknown, which may cause a longer TCI state switching delay period, thereby reducing the advantages associated with the beam prediction method.

[0204] Example embodiments may avoid or mitigate such issues by ensuring or increasing the likelihood that the indicated target TCI state will be known at the user terminal based on one or more conditions associated with at least one predicted beam and / or measured first reference signal set and / or its associated beam (e.g., beam set B). The user terminal and / or network entity may perform a switch to the target TCI state based on a determination that the target TCI state is known from these one or more conditions. In some example embodiments, the switch to the target TCI state is performed within a time period based on a determination that the target TCI state is known.

[0205] Figure 1 A network scenario 100 comprising a network entity 120 and a user terminal 122 is shown.

[0206] Although one user terminal 122 is shown, in other example embodiments there may be multiple user terminals that may be considered separately.

[0207] The network entity 120 may include a network node such as, but not limited to, a gNB (in the context of 5G / NR or future generations of technology) or a transmission and reception point (TRP) or a relay node. The network entity 120 may be connected to one or more other network functions (not shown), which may include part of a core network or the like.

[0208] The user terminal 122 may include a terminal device or user equipment (UE) such as, but not limited to, a mobile phone, a tablet, a laptop, a personal computer, an Internet of Things (IoT) device, a digital assistant, or a wearable terminal (such as a smart watch).

[0209] The network entity 120 may be configured to transmit data to the user terminal 122 over the air interface 124 using one or more downlink channels.

[0210] The user terminal 122 may be configured to receive data sent by the network entity 120 over the air interface 124 using one or more downlink channels.

[0211] The network entity 120 may also be configured to receive data sent by the user terminal 122 over the air interface 124 using one or more uplink channels.

[0212] Examples of downlink and uplink channels are mentioned above.

[0213] The air interface 124 may include a cellular air interface associated with the radio access technology (RAT) that both the network entity 120 and the user terminal 122 are configured to support, such as 5G / NR or a future generation RAT.

[0214] Network entity 120 and user terminal 122 may be configured so that the network entity transmits data for at least one of the above downlink channels using a directional beam. For example, network entity 120 may process and transmit data via an antenna array using MIMO technology. This may involve forming a first beam 126. User terminal 122 may configure its receiver to receive data transmitted using first beam 126.

[0215] Later, the network entity 120 may switch beams, eg, from a current beam to a new beam or target beam 128 .

[0216] This process is called beam switching and may need to handle issues such as path loss experienced by the user terminal 122 and / or changes in the user terminal's position and / or orientation. To this end, a model may be used to predict at least one "predicted beam" to which the network entity 120 and the user terminal 122 may switch.

[0217] The term model may mean a machine learning model or a computational model or similar.

[0218] The model may include, but is not limited to, an artificial neural network (ANN). Various known types of ANNs exist, including feedforward neural networks, perceptron neural networks, convolutional neural networks, recurrent neural networks, deep neural networks, and the like. Each type may be more suitable for a specific application or task. The model may be trained over one or more training iterations to predict at least one predicted beam based on applying measurements of a first set of reference signal resources associated with one or more beams as input to the model. The one or more training iterations may involve supervised or unsupervised learning, or the like.

[0219] For example, for supervised learning, the training dataset can be a labeled training dataset that includes multiple labeled training data instances. Each training data instance includes measurement data associated with a specific beam set from beam set B or an associated reference signal resource, wherein the training data instance is annotated or labeled with data representing one or more beams from beam set A that are the most suitable beams to be switched to based on the measurement data of the training data instance. Supervised learning adjusts the weights of the model based on the model, and when performing its inference operation on the measurement data of the one or more training data instances, provides an indication of predicting one or more beams from beam set A that correspond to the corresponding labeled or most suitable beams from beam set A for each training data instance.

[0220] For example, for unsupervised learning, training iterations can be based on a set of measurement data for beam set B and / or associated reference signal resources (e.g., the first set) and a set of measurement data associated with beams of beam set A and / or associated reference signal resources, and the unsupervised learning adjusts the weights of the model based on its model, and in its inference operation, when data correspondingly representing the measurement data associated with beam set B (or the first set) is input into the model, provides an indication of predicting the most appropriate beam from beam set A.

[0221] It should be understood that the specific training method and / or model type is not the focus of the example embodiments disclosed herein.

[0222] By predicting data representing at least one predicted beam, it may be avoided to perform an exhaustive search of all possible beams that the network entity is capable of forming.

[0223] The model may be provided at the network entity 120 or the user terminal 122 .

[0224] Figure 2 is a flow chart indicating operation 200 according to one or more example embodiments. Operation 200 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 200 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the operation to be performed. For example, operation 200 may be performed by a user terminal or device, such as a Figure 1 122) is executed by the user terminal.

[0225] The first operation 201 may include receiving a configuration of a set of reference signal resources for measuring a corresponding beam for a network entity, wherein the set of reference signal resources is associated with prediction-based beam reporting or prediction-based beam switching.

[0226] The second operation 202 may include measuring a first set of reference signal resources of a configured set of reference signal resources for a first group (e.g., set B) of beams to obtain a value for the reference signal resource using a model (e.g., as referenced). Figure 1

[0014] In another embodiment, a model (described in detail in the accompanying drawings) is used to determine at least one predicted beam from a second set of beams (e.g., set A), wherein the model performs an inference operation to provide data representative of an indication of the at least one predicted beam. The data representative of the indication of the at least one predicted beam output by the model can form a set of predicted beams or associated RS resources.

[0227] The third operation 203 may include receiving an indication from the network entity that a target transmission configuration indicator (TCI) state corresponding to at least one predicted beam is to be activated or indicated.

[0228] The fourth operation 204 may include determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: at least one predicted beam in the prediction set and the measured first reference signal resource set and / or corresponding beam.

[0229] A fifth operation 205 may include switching to a target TCI state based on the determination.

[0230] In some example embodiments, a configured set of reference signal resources (hereinafter referred to as RS resources) may include a set of one or more RSs associated with corresponding beams that a network entity may transmit.

[0231] In some example embodiments, the configuration may be received from a network entity.

[0232] In some example embodiments, the configuration may include resource sets that respectively indicate RS sets associated with respective beams.In some example embodiments, the resource sets may indicate time and frequency resources, enabling a user terminal to perform measurements on the configured RS resource sets.

[0233] In some example embodiments, the configured RS resource set may be associated with prediction-based beam reporting or prediction-based beam switching, as measurements on the RS resource set may be used for the purpose of prediction-based beam reporting or prediction-based beam switching.

[0234] In some example embodiments, another operation may include receiving a message from a network entity for triggering prediction-based beam reporting or prediction-based beam switching, wherein measurement of the configured RS resource set is performed upon or after receiving the message.

[0235] In some example embodiments, measuring the configured set of RS resources may include measuring at least one configured set of RS resources in the configured set of RS resources. In some example embodiments, measuring the configured set of RS resources may include performing at least one physical layer measurement, i.e., L1 measurement, such as, but not limited to, RSRP measurement.

[0236] In some example embodiments, in operation 204, one or more conditions are used to determine whether the target TCI state is known based on, but not limited to, the following: for example, a) whether any RS resources included in or associated with the target TCI state correspond to quasi-co-located (QCL) source RS resources related to the first set; b) whether any one or more RSs included in or associated with the target TCI state correspond to one or more RSs of the first set and corresponding measured quality metrics for the RS resources in the first set; c) whether the RS resources included in or associated with the target TCI state correspond to RSs of at least one predicted beam in the predicted set of predicted beams. resources and corresponding prediction quality metrics or measured quality metrics for RS resources in the prediction set; d) whether the RS resources included in or associated with the target TCI state correspond to the RS resources of at least one prediction beam in the prediction set and the corresponding monitoring metrics for the RS associated with the prediction set; and / or e) whether the RS resources included in or associated with the target TCI state correspond to the RS resources of at least one prediction beam in the prediction set and the corresponding prediction or performance metrics, such as but not limited to, for example, model performance metrics, model performance accuracy metrics, model confidence or probability score metrics and corresponding thresholds for model inference operations for the RS in the prediction set.

[0237] In some example embodiments, in operation 205, based on determining that switching to a target TCI state includes determining that the target TCI state is known, switching to the target TCI state within a time period. In some example embodiments, switching to the target TCI state within a time period based on determining that the target TCI state is known further includes one or more of the following: performing a TCI state switch within the time period according to the known target TCI state, where the time period is calculated based on an L1-RSRP measurement time period required to omit reference signal resources of the target TCI state for L1-RSRP measurement; or performing a TCI state switch within the time period according to the known target TCI state, where the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample related to the reference signal resources of the target TCI state has been obtained.

[0238] Figure 3 shows Figure 2 operation and / or Figures 4a to 6d an example of how the operation of can be performed.

[0239] According to the first operation 201, the user terminal 122 may receive a configuration for measuring a first set of RS resources 310 (e.g., RS#i, RS#j) for corresponding beams (e.g., beam #i, beam #j) of the network entity 120. The configuration may indicate a set of M RSs (RS#1 to RS#M) associated with the corresponding beams (beam #1 to beam #M), where the first set of RS resources 310 is used for measurement by the user terminal 122. The first set of RS resources 310 may include a subset of the M RSs. For example, the first set of RS resources 310 may include N RSs (RS#1,..., RS#i, RS#j,..., to RS#N) associated with the corresponding beams (beam #1 to beam #N), where N < M. As an example, the beam may be identified or associated with a reference signal.

[0240] The configuration may be received from the network entity 120.

[0241] The user terminal 122 may measure at least one RS resource in the first set of RS resources 310 based on the configuration.

[0242] According to the second operation 302, the user terminal 122 may measure at least one RS resource in the first set of RS resources 310 at a specific time.

[0243] In some example embodiments, the user terminal 122 may be configured to measure the first set of RS resources 310 in response to receiving a message from the network entity 120 for triggering a prediction-based beam report or a prediction-based beam switch.

[0244] In some example embodiments, the message may indicate or separately configure the first set of RS resources 310 .

[0245] Measuring the first set of RS resources 310 may include performing channel and / or interference related measurements, such as performing at least one physical layer (L1) measurement on the N RSs, for example, at least one RSRP measurement. Such measurements are used in connection with prediction-based beam reporting or prediction-based beam switching.

[0246] The measurement of the first RS resource set 310 may be performed within a first time period.

[0247] The measurement of the first RS resource set 310 may start when or after receiving the above-mentioned message for triggering prediction-based beam reporting or prediction-based beam switching.

[0248] The measurements on the first set of RS resources 310 (ie, the measurement results) may be applied or input to a model 312 (eg, as referenced Figure 1 or the model described in 2).

[0249] In some example embodiments, the measurements of the first set 310 are not reported to the network entity 120 or do not need to be (or are configured not to be) reported to the network entity 120. For example, the RS resource may not be associated with a reporting configuration. For example, the RS resource is associated with a reporting configuration that configures measurements on the RS not to be reported.

[0250] The model 312 may be provided at the user terminal 122 or the network entity 120 .

[0251] In the event that the model 312 is provided at the user terminal 122 , the user terminal may apply or input the measurement results on the first set of RS resources 310 to the model 312 .

[0252] In case the model 312 is provided at the network entity 120 and the user terminal 122 performs measurements on the first set 310, the user terminal 122 may report the measurement results to the network entity 120. The network entity 120 may apply or input the received measurement results to the model 312.

[0253] In an alternative embodiment, if the model 312 is provided at the network entity 120 , the network entity 120 may be configured to perform measurements of the first set of RS resources 310 , in which case the user terminal does not perform measurements of the first set of RS resources 310 .

[0254] Model 312 may be configured to perform inference operations to provide an indication of at least one predicted beam 314 in a second set of predicted RS resources 314 (predicted set). In the illustrated example, multiple predicted beams (e.g., beam #4: RS#4, ..., beam #k: RS#k, ..., beam #N: RS#N) of predicted set 314 are predicted. While multiple predicted beams are shown, this is by way of example only, and one skilled in the art will appreciate that any number of predicted beams associated with a configuration may be predicted, e.g., one beam: beam #4, or more than one beam: beam #4, ..., beam #k, ..., beam #N associated with a particular RS resource RS#4 may be predicted based on application requirements.

[0255] The inference operation of model 312 can be configured to provide corresponding prediction metrics associated with one or more prediction beams of the prediction set 314, including but not limited to, for example, the prediction quality for each prediction beam (e.g., predicted RSRP or signal-to-noise ratio (SNR)) and / or prediction model performance associated with the prediction of one or more prediction beams (e.g., mean squared error, area under the curve, probability score / confidence score for the prediction (e.g., the model's confidence that the predicted value is correct), etc.).

[0256] Instead of user terminal 122 reporting measurement results for first RS set 310, a second predicted RS resource set (predicted set 314) (or a second set of predicted RS resources) based on first RS resource set 310 may be associated with the reporting configuration. User terminal 122 may be configured to report data representing at least one predicted beam to network entity 120. Such predicted RS resources are used in connection with prediction-based beam reporting or prediction-based beam switching. In some example embodiments, the report by user terminal 122 may also include data representing the predicted metrics output by model 312.

[0257] User terminal 122 may report at least one predicted beam 314 to network entity 120. In the illustrated example, user terminal 122 may report data representing predicted set 314 to network entity 120, where predicted set 314 includes at least one or more predicted beams: Beam #4, ... Beam #k, ... Beam #N, and possibly associated specific RS resources RS #4, ... RS #k, ... RS #N. In some example embodiments, the report by user terminal 122 may also include data representing a predicted metric output by model 312.

[0258] In the case where the model 312 is provided at the network entity 120, at least one predicted beam of the prediction set 314 is known by the network entity 120. The network entity 120 may report prediction metrics to the user terminal 122 explicitly or implicitly, including but not limited to, for example, prediction quality for one or more predicted beams and / or model performance metrics associated with one or more predicted beams.

[0259] Once the network entity 120 has the prediction set 314, the network entity 120 may generate a target TCI state corresponding to at least one predicted beam in the prediction set 314 to be activated or indicated. The network entity 120 may send an indication that the target TCI state corresponding to the at least one predicted beam in the prediction set 314 is to be activated or indicated. The target TCI state may indicate a specific RS resource associated with the at least one predicted beam in the prediction set 314. In the example shown, the target TCI state may indicate that RS #4 is associated with beam #4. The target TCI state may include one or more RSs.

[0260] In some example embodiments, the predicted beams of the prediction set 314 (or RSs corresponding to the predicted beams) may correspond to RSs included in the TCI state or the target TCI state.

[0261] In some example embodiments, the predicted beam (or the RS corresponding to the predicted beam) of the predicted set 314 may correspond to one of the RSs included in the TCI state or the target TCI state. In one example, if multiple RSs are included in the TCI state (or the target TCI state), the RS associated with the predicted beam (or the predicted RS of the predicted beam) may correspond to the RS providing QCL-type D information in the TCI state.

[0262] According to the third operation 203, the user terminal 122 may receive an indication from the network entity 120 that a TCI state (or a target TCI state) corresponding to at least one predicted beam is to be activated or indicated. In the case where the model 312 is provided at the network entity 120, the network entity 120 may also provide the user terminal 122 with data representing one or more predicted beams of the prediction set 314 and / or corresponding prediction metrics associated with, for example, prediction quality (e.g., predicted RSRP or signal-to-noise ratio (SNR)) and / or model performance (e.g., mean square error, or other model performance metrics).

[0263] According to the fourth operation 204, the user terminal 122 may determine whether the indicated target TCI state is known or unknown based on the measured first RS resource set 310, the predicted set 314 of predicted beams of RS resources, and / or at least one prediction metric associated with at least one predicted beam of the predicted set 314. This operation is indicated by reference numeral 318.

[0264] When determining whether the indicated target TCI state is known or unknown, the user terminal 122 considers various one or more conditions 316 associated with at least the measured first set of RS resources 310, the predicted set 314 of predicted beams of RS resources and / or the predicted metrics associated with at least one predicted beam of the predicted set 314.

[0265] These one or more conditions 316 are used to determine the target TCI state based on, but not limited to, for example: a) whether any RS resources included in or associated with the target TCI state correspond to quasi-co-located source RS resources associated with the first set 310; b) whether any one or more RSs included in or associated with the target TCI state correspond to one or more RSs of the first set 310 and corresponding measured quality metrics for the RS resources in the first set 310; c) whether the RS resources included in or associated with the target TCI state correspond to RS resources of at least one predicted beam of the predicted set 314 and corresponding measured quality metrics for the RS resources in the predicted set 314. A prediction quality metric or a measurement quality metric; d) whether the RS resources included in or associated with the target TCI state correspond to the RS resources of at least one prediction beam in the prediction set 314 and the corresponding monitoring metric for the RS in the prediction set; and / or e) whether the RS resources included in or associated with the target TCI state correspond to the RS resources of at least one prediction beam in the prediction set 314 and the corresponding prediction or performance metric, such as but not limited to, for example, a model performance metric, a model performance accuracy metric, a model confidence or a probability score metric for the RS in the prediction set, where the prediction metric or performance metric can be compared with a corresponding threshold.

[0266] Although the user terminal 122 refers to Figure 3Described as being configured to use or apply the above-mentioned one or more conditions 316, wherein the one or more conditions 316 may include or be associated with one or more of the following conditions described below, for determining whether the target TCI state is known or unknown, is merely by way of example, and the present invention is not limited thereto. It should be understood by those skilled in the art that other devices (such as, for example but not limited to, the network entity 120) may also be configured or adapted to use or apply the above-mentioned one or more conditions 316 and / or the following one or more conditions, for determining whether the target TCI state is known or unknown, or whether the target TCI state is known or unknown to the user terminal 122 before switching, and so on, according to the requirements of the application.

[0267] For example, alternatively or in addition to the above condition 316, the user terminal 122 may be configured to determine that the indicated target TCI state is known using one or more of the following conditions if at least one RS resource in the measured first set of RS resources 310 corresponds to:

[0268] (i) RS resources included in or associated with the target TCI state; or

[0269] (ii) RS resources associated with at least one predicted beam in the prediction set 314, or

[0270] (iii) RS received from a source (e.g., port) of a source QCL of RS resources associated with at least one predicted beam; or

[0271] (iv) RS received from a source that is quasi-co-located with the source of the RS included in or associated with the target TCI state.

[0272] Therefore, the user terminal 122 may determine that the indicated target TCI state is known. Otherwise, it may be determined that the target TCI state is unknown.

[0273] In some example embodiments, alternatively or in addition to the above-described condition(s) 316, if the measured quality metric of the first set of RS resources 310 corresponding to the RS resources included in or associated with the target TCI state is greater than or equal to a preconfigured or predefined measured quality threshold level, the user terminal 122 may be configured to use the following condition to determine that the indicated target TCI state is known. Otherwise, it may be determined that the target TCI state is unknown. In some example embodiments, the measured quality metric may include, but is not limited to, one or more of the following: RS received power (RSRP); L1 physical layer RSRP (L1-RSRP); signal-to-noise ratio (SNR) of the RS resource; or any other quality metric associated with the RS resource that can be measured.

[0274] In some example embodiments, alternatively or in addition to the above-described condition(s) 316, user terminal 122 may be configured to use the following condition to determine that the indicated target TCI state is known if a predicted quality metric of the RS resource corresponding to the predicted beam of the RS resource included in or associated with the target TCI state is greater than or equal to a preconfigured or predefined quality threshold level. Otherwise, it may be determined that the target TCI state is unknown. In some example embodiments, the predicted quality metric may include, but is not limited to, one or more of the following: predicted RSRP; predicted L1-RSRP; predicted SNR of the RS resource; any other quality metric associated with the RS resource that can be predicted with respect to at least one predicted beam of the prediction set 314.

[0275] For example, determining whether the target TCI state is known or unknown may further include determining that one or more predicted quality metrics of one or more RS resources associated with the corresponding reported predicted beam of the prediction set 314 are greater than or equal to a preconfigured or predefined quality threshold level. In response to determining that at least one of the one or more predicted quality metrics of the one or more RS resources associated with the corresponding reported predicted beam is greater than or equal to the preconfigured or predefined quality threshold level, when the one or more RS resources correspond to RS resources included in or associated with the target TCI state, it is determined that the target TCI state is known. Otherwise, it may be determined that the target TCI state is unknown.

[0276] In some example embodiments, alternatively or in addition to the above-described condition(s) 316, the user terminal 122 may be configured to use the following condition to determine that the target TCI state is known if the prediction accuracy metric for at least one predicted beam is greater than or equal to a prediction accuracy threshold level. Otherwise, it may be determined that the target TCI state is unknown. For example, the user terminal 122 may be configured to derive the prediction accuracy metric from performing monitoring of the prediction accuracy of the model 312 with respect to the at least one predicted beam. For example, the monitoring of the prediction accuracy may be based on measurements of one or more RS resources corresponding to one or more previously predicted beams output by the model 312. The prediction accuracy metric for the at least one predicted beam may be based on the monitoring of the previously predicted beams and an indication of the corresponding at least one predicted beam of the prediction set 314.

[0277] In another example embodiment, alternatively or in addition to the above-described condition(s) 316, the user terminal 122 may be configured to monitor one or more RS resources corresponding to one or more previously predicted beams. The user terminal 122 may be configured to determine a monitoring accuracy metric between the (previous) predicted beam of the RS resources and the corresponding current predicted beam of the RS resources. If the monitoring accuracy metric for the current predicted beam having RS resources included in or associated with the target TCI state is greater than or equal to a monitoring accuracy threshold level, the user terminal 122 may determine that the target TCI state is known using the following condition. Otherwise, it may be determined that the target TCI state is unknown.

[0278] In some embodiments, the user terminal 122 is configured to perform monitoring of prediction accuracy, wherein the monitoring of prediction accuracy is based on measurements of one or more RS resources corresponding to one or more previously predicted beams. The user terminal 122 is configured to determine a prediction accuracy metric for at least one predicted beam based on said monitoring of the previously predicted beams and an indication of the corresponding at least one predicted beam. Alternatively or in addition to the above-mentioned (multiple) conditions 316, if the prediction accuracy metric for at least one predicted beam is greater than or equal to a prediction accuracy threshold level, the user terminal 122 may be configured to determine that the target TCI state is known using the following condition. Otherwise, it may be determined that the target TCI state is unknown.

[0279] In some embodiments, alternatively or in addition to the above-described condition(s) 316, the user terminal 122 may use the following conditions to determine that the target TCI state is known if the current model performance metric corresponding to the model inference operation of at least one predicted beam having RS resources included in or associated with the target TCI state is greater than or equal to the model performance threshold level. Otherwise, it may be determined that the target TCI state is unknown. For example, the model performance metric includes, but is not limited to, one or more of the following group: a model prediction accuracy metric; a precision and recall metric; an area under the curve receiver operating curve (AUC-ROC) metric; a confidence metric or a probability score metric; and any other model performance metric (or ML performance statistic / metric or measurement) for evaluating the performance of the model 312.

[0280] In one example, the confidence metric or probability score metric can be a confidence or probability value / score that can indicate the degree of confidence that the model 312 has in the accuracy of the predicted value or attribute. The probability or confidence can be expressed as, for example, a percentage value or a decimal value (e.g., in a range between [0..1]) and / or based on any other suitable value, etc.

[0281] In some embodiments, alternatively or in addition to the above-described condition(s) 316, the user terminal 122 may use the following condition to determine that the target TCI state is known if the current model confidence or probability score metric of the model inference operation corresponding to at least one predicted beam of the RS included in or associated with the target TCI state is greater than or equal to the corresponding confidence threshold level or probability score threshold level. Otherwise, the target TCI state may be determined to be unknown.

[0282] In some embodiments, alternatively or in addition to the above-described condition(s) 316, if an indication or activation of the target TCI state is received within a predefined time period after completing reporting data representing the second set of beams or the predicted set 314 including at least one predicted beam to the network entity 120, and the RS associated with the at least one predicted beam corresponds to the RS included in or associated with the target TCI state, the user terminal 122 may use the following conditions to determine that the target TCI state is known. Otherwise, it may be determined that the target TCI state is unknown. For example, the predefined time period is less than the time period used to perform the inference operation. In another example, the predefined time period is less than the time period since the most recent inference operation. In another example, the predefined time period is greater than the time period used for non-predicted TCI switching.

[0283] In some embodiments, alternatively or in addition to the above-described condition(s) 316, the user terminal 122 may use the following condition to determine that the target TCI state is known if an indication or activation of the target TCI state is received within X milliseconds, seconds, time slots, radio frames, or symbols after the user terminal 122 completes reporting data representing a predicted set 314 (e.g., a second set (or group)) of beams including data representing an indication of at least one predicted beam to the network entity 120. Otherwise, the target TCI state may be determined to be unknown.

[0284] In some embodiments, alternatively or in addition to the above-mentioned condition(s) 316, if the user terminal 122 has performed at least one measurement on an RS corresponding to an RS included in or associated with the target TCI state or on an RS received from a source quasi-co-located with an RS included in or associated with the target TCI state within N milliseconds, seconds, time slots, radio frames or symbols before receiving an indication that the target TCI state is indicated or activated, the user terminal 122 can be configured to use the following condition to determine that the target TCI state is known.

[0285] In some embodiments, user terminal 122 may be configured to set at least one of the following: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level. In other embodiments, network entity 120 may be configured to instruct user terminal 122 to set at least one of the following: a monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level. For example, user terminal 122 may receive the monitoring accuracy threshold level, the model performance threshold level, the confidence threshold level, or the probability score threshold level from network entity 120.

[0286] In some embodiments, the user terminal 122 may be configured to report at least one of the following: a monitoring accuracy metric, a model performance metric, a confidence threshold level, or a probability score threshold level to the network entity 120. The user terminal 122 and / or the network entity 120 are configured to use at least one of the following as a current monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric for comparison with a corresponding monitoring accuracy threshold level, a model performance threshold level, a confidence threshold level, or a probability score threshold level: the reported monitoring accuracy metric, the model performance metric, the model confidence metric, or the probability score metric.

[0287] In other embodiments, the user terminal 122 may be configured to receive at least one of the following items from the network entity 120: a monitoring accuracy metric, a model performance metric, a model confidence metric, or a probability score metric. The user terminal 122 is configured to use at least one of the currently received monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric as the current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with the corresponding monitoring accuracy threshold level, model performance threshold level, model confidence threshold level, or probability score threshold level. In another example, the network entity 120 is configured to use the received monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric as the current monitoring accuracy metric, model performance metric, model confidence metric, or probability score metric for comparison with the monitoring accuracy threshold level, model performance threshold level, model confidence threshold level, or probability score threshold level.

[0288] In some embodiments, the target TCI state is considered known if the indication or activation of the target TCI state of operation 203 is received within X milliseconds, seconds, time slots, radio frames, or symbols after the user terminal 122 completes reporting data representing the predicted set 314 of beams including an indication of at least one predicted beam to the network entity. In some embodiments, for prediction-based TCI state switching, the target TCI state is considered known if the beam switching command / configuration of operation 203 from the network entity 120 is received within a time period of X time units after the user terminal 122 has reported one or more predicted beams of the prediction set 314. The time unit may be any suitable time unit, including but not limited to, for example, milliseconds, time slots, seconds, etc.

[0289] In some embodiments, before the prediction-based beam switching in operation 205, the user terminal 122 may be further configured to measure one or more RSs in the second set of configured RS resource sets associated with at least one predicted beam of the prediction set 314. For example, measuring the one or more RSs in the second set of configured RS resource sets includes measuring RSs associated with the at least one predicted beam or RSs associated with a beam received from a source of a source QCL of the RS associated with the at least one predicted beam of the prediction set 314. As an option, after or based on measuring the at least one RS associated with the at least one predicted beam or RSs associated with a beam received from a source of a source QCL of the RS associated with the at least one predicted beam of the prediction set 314, the prediction-based beam switching is performed based on whether the target TCI state is known.

[0290] In some example embodiments, the user terminal 122 may be configured to perform prediction-based beam reporting or prior to prediction-based beam switching, wherein the user terminal 122 is configured to: trigger measurement of at least one RS resource associated with at least one predicted beam of the predicted beam after reporting data representing an indication of one or more predicted beams of the prediction set 314, wherein the measurement of the at least one RS resource associated with the at least one predicted beam of the predicted beam is performed based on one or more of the following groups: measuring the RS resource associated with the at least one predicted beam in a first time slot of a specific RS resource occurring after reporting the predicted beam associated with the RS resource; or measuring the RS resource associated with the at least one predicted beam within a predetermined or specified time period after reporting the predicted beam associated with the RS resource. In another embodiment, after measuring the RS associated with the at least one predicted beam, prediction-based beam switching is performed based on whether the target TCI state is known.

[0291] For example, if the user terminal 122 has performed at least one measurement on the RS of at least one RS in the target TCI state or is in QCL with the at least one RS in the target TCI state, the user terminal 122 may determine that the target TCI state is a known TCI state. In another example, if the user terminal 122 has performed at least one measurement on the RS of at least one RS in the target TCI state or is in QCL with the at least one RS in the target TCI state within a predefined or predetermined time period before the user terminal 122 receives a TCI state switching indication or before the user terminal 122 is to perform TCI state switching, the user terminal 122 may determine that the target TCI state is a known TCI state.

[0292] The measurement of the at least one measured RS may be, for example, an L1 physical layer measurement. For example, the L1 physical layer measurement may be an RSRP measurement.

[0293] In some example embodiments, in addition to the above-described condition(s), user terminal 122 may determine that the indicated target TCI state is known if at least one measurement of a particular RS resource is an L1 physical layer measurement (e.g., an RSRP measurement).

[0294] In some embodiments, the target TCI state is known if an indication or activation of the target TCI state is received within X milliseconds, seconds, time slots, radio frames, or symbols after the user terminal 122 completes reporting data representing the prediction set 314 including at least one predicted beam to the network entity 120.

[0295] In another embodiment, in addition to one or more conditions 316, the user terminal 122 may determine that the target TCI state is known if the user terminal 122 has performed at least one measurement on an RS corresponding to an RS included in or associated with the target TCI state or on an RS received from the source of a source QCL of the RS included in or associated with the target TCI state within N milliseconds, seconds, time slots, radio frames, or symbols prior to operation 203 of receiving an indication or activation of the target TCI state.

[0296] According to a fifth operation 205, if the indicated target TCI is known, the user terminal 122 may switch to the target TCI state (eg, within or after a period of a known TCI state switching delay).

[0297] In some example embodiments, switching to the target TCI state may be performed within (or after) a determined time period based on an indication that the target TCI state is known. For example, the time period may be based on a known TCI state switching delay associated with the target TCI state when the target TCI state is known. In one example, switching to the target TCI state within a determined time period based on the target TCI state being known may further include one or more of the following: performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on omitting an L1-RSRP measurement time period required for L1-RSRP measurements of RS resources of the target TCI state; or performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0298] In some example embodiments, switching to the target TCI state may be performed within (or after) a time period of a known TCI state switch delay associated with the target TCI state, such as based on a time period specified in 3GPP TS 38.133. For example, the calculated time period may be derived from, based on, or include the downlink TCI state switch delay specified in 8.15.3, "MAC-CE based downlink TCI state switch delay," wherein when the target TCI state is known, the user terminal 122 may delay the TCI state switch until the first time slot after the delay period: time slot n + T HARQ ++TO uk *(T first-SSB +T SSB-proc ) / NR time slot length.

[0299] In some embodiments, the user terminal 122 may perform the determination of whether the target TCI state is unknown or known as described in the above embodiments. However, in response to determining that the target TCI state is unknown, the user terminal 122 may modify the predicted beam switching delay based on one or more of the following: when the predicted beam switching delay uses a parameter value related to the L1 RSRP measurement time (e.g., T L1-RSRP ) is calculated, the parameter value is set to 0 (for example, T L1-RSRP =0); when the predicted beam switching delay is used with TO uk When the relevant parameter values are calculated, for L1-RSRP measurement based on channel state information reference signal (CSI-RS), TO uk Set to 1; when the predicted beam switching delay is used with TO ukWhen the relevant parameter values are calculated, when the TCI state switching involves QCL-TypeD, for L1-RSRP measurement based on synchronization signal block (SSB) or synchronization signal physical broadcast channel block (SS / PBCH), TO uk Set to 0.

[0300] For example, according to 3GPP 38.133 8.15.3 “MAC-CE based downlink TCI state switch delay”, when the target TCI state is unknown, the user terminal 122 may delay the TCI state switch until the first time slot after the delay period of: time slot The predicted beam switching delay can be based on this delay period, where when the target TCI state is unknown in the predicted beam switching, the L1 RSRP measurement time (e.g., T L1-RSRP ) can be set to zero (e.g., T L1-RSRP = 0). In this example, the predicted beam switching delay may be the first time slot after the following delay period: time slot In another example, the predicted beam switching delay can be based on the above delay period, where when the target TCI state is unknown in the predicted beam switching, then for the L1-RSRP measurement based on CSI-RS, the parameter TO uk is set to 1. The predicted beam switching delay may be the first time slot after the following delay period: time slot In another example, the predicted beam switching delay can be based on the above delay period, where when the target TCI state is unknown in the predicted beam switching, when the TCI state switching involves QCL-Type D, for L1-RSRP measurement based on SSB, the delay is the same as TO uk The relevant parameter value is set to 0. The predicted beam switching delay may be the first time slot after the following delay period: time slot Time slot length.

[0301] The network entity 120 may be configured to switch to the target TCI state simultaneously so that both the user terminal 122 and the network entity 120 are synchronized.

[0302] Therefore, by measuring the first set of reference signal resources 310, predicting quality metrics and / or model performance metrics of the predicted beams in the prediction set 314, determining monitoring accuracy metrics associated with previous and current predicted beams in the prediction set 314, and / or measuring the reference signal resources of the predicted beams in the prediction set 314, the likelihood that the user terminal 122 will robustly and accurately determine that the indicated target TCI state is known is increased in the case of prediction-based beam reporting or prediction-based beam switching.

[0303] Although the one or more conditions 316 described above are described as being used or applied by the user terminal 122 to determine whether the target TCI state is known, this is by way of example only and the present invention is not limited thereto. Those skilled in the art will appreciate that the one or more conditions 316 and other conditions described herein may be used by the user terminal 122, the network entity 120, and / or both to determine the target TCI state (whether known or unknown) based on the one or more conditions 316 associated with at least one of the following: at least one predicted beam and the first set of reference signal resources 310. For example, the network entity 120 may implement a model 312 in which the user terminal 122 reports measurements of the first set of RS resources 310, causing the network entity 120 to generate a predicted set 314 of at least one predicted beam and may use the one or more conditions to determine whether the target TCI state associated with the predicted beam is known or unknown. The network entity 120 may receive the measurements of the first set of RS resources 310 and, accordingly, use the model 312 and / or other models to calculate or predict a quality metric and / or a model performance metric for the predicted beam in the predicted set 314. The network entity 122 may also determine monitoring accuracy metrics associated with previous and current predicted beams of the prediction set 314 when implementing the model 312, and / or receive additional measurements of RS resources for the predicted beams in the prediction set 314. Thus, the network entity 120 may be configured to determine whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of: at least the predicted beam, the configuration of the first set of RS resources 310, and / or the measured first set of RS resources 310, and switch to the target TCI state based on the determination.

[0304] In another example, the user terminal 122, the network entity 120, and / or both may be configured to determine whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least the predicted beam, the configuration of the first set of RS resources 310, and / or the measured first set of RS resources 310, and switch to the target TCI state based on the determination. In another example, as described above, the user terminal 122 may be configured to determine whether the target TCI state is known based on one or more conditions 316 associated with at least one of the following: at least the predicted beam of the prediction set 314 and the measured first set of RS resources 310, and switch to the target TCI state based on the determination. At the same time, the network entity 120 can be configured to determine whether the target TCI state is known based on one or more conditions 316 associated with at least one of the following items: at least the predicted beam of the prediction set 314, the configuration of the first RS resource set 310 and / or the measured first RS resource set 310 (for example, if reported to the network entity 120 by the user terminal 122), and when it is determined that the target TCI state is known, switch to the target TCI state accordingly.

[0305] In another example, the network entity 120 can be configured to determine whether the target TCI state is expected to be known or unknown by the user terminal 122 based on one or more conditions 316 associated with at least one of the following items: at least the predicted beam and configuration of the first RS resource set 310 and / or the measured first RS resource set 310 (e.g., if reported to the network entity 120 by the user terminal 122), and if known, the network entity 120 can switch to the target TCI state based on the determination.

[0306] Furthermore, in some embodiments, the user terminal 122, the network entity 120, and / or both may be configured to, in the case where the target TCI state is known or unknown, determine whether the target TCI is known based on at least one of the following: at least one predicted beam, the configuration of the first set of RS resources 310, and / or the measured first set of RS resources 310 (e.g., if reported by the user terminal 122 to the network entity 120). The user terminal 122, the network entity 120, and / or both may be configured to determine to switch to the target TCI state within the time period based on an indication that the target TCI state is known.

[0307] Switching to the target TCI state may be performed by both the user terminal 122 and / or the network entity 120 within (or after) a time period determined based on an indication that the target TCI state is known. For example, when the target TCI state is known, the time period may be based on a known TCI state switching delay associated with the target TCI state. In an example, switching to the target TCI state within a time period determined based on the target TCI state being known may further include one or more of: performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on omitting an L1-RSRP measurement time period required for L1-RSRP measurements of RS resources of the target TCI state; or performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0308] Therefore, by measuring the first RS resource set 310, predicting quality metrics and / or model performance metrics of the predicted beams in the prediction set 314, determining the monitoring accuracy metrics associated with the previous and current predicted beams in the prediction set 314, and / or measuring the RS resources of the predicted beams in the prediction set 314, the user terminal 122 and / or the network entity 120 can use one or more conditions 316 to robustly and accurately determine that the target TCI state indicated in the case of prediction-based beam reporting or prediction-based beam switching is known and can be switched to the target TCI state synchronously, wherein the user terminal 122 and / or the network entity 120 receive appropriate report messages from each other (for example, report messages associated with measurements of the first set 310, predictions of the prediction set 314, monitoring accuracy metrics, prediction accuracy metrics, model performance metrics, quality metrics, etc., but not limited to these) depending on the location where the model 312 is implemented.

[0309] Figure 4a A signaling and processing diagram is shown for a case where the user terminal 122 uses the model 312 to perform inference operations, according to some example embodiments.

[0310] The first operation 4.1 may include the network entity 120 configuring the user terminal 122 with a set of RS resources for performing measurements related to the first set of RS resources 310 and / or predicting beams from the set of RS resources to form a predicted set 314. The predicted set 314 may include beams associated with the first set of RS resources 310 and / or beams from the set of RS resources. This may also include the network entity 120 indicating (or separately configuring) to the user terminal 122 the first set of RS resources 310 to be measured for use in prediction-based beam switching or prediction-based beam reporting.

[0311] The second operation 4.2 may include the user terminal 122 starting to measure the first set of RS resources 310 for use in predicting at least one beam and associated RS resources from the set of RS resources to form a predicted set 314. The predicted set 314 includes one or more predicted beams and associated RS resources, as described in reference to FIG. Figure 3 As stated.

[0312] The third operation 4.3 may include the user terminal 122 performing an inference operation using the measurement result of the third operation 4.3. The inference operation uses a model configured to predict at least one RS resource set predicted beam and / or associated RS resources in the RS resource set for use in prediction-based beam reporting and / or prediction-based beam switching when the network entity 120 provides a target TCI state. The prediction result of the model may include, but is not limited to, at least one of the following: data representing an indication of at least one predicted beam and / or associated RS resources; at least one of data representing a predicted quality metric associated with the at least one predicted beam; or data representing a predicted model performance metric associated with the at least one predicted beam.

[0313] The fourth operation 4.4 may include the user terminal 122 reporting data representing at least one predicted beam to the network entity 120, which data may include an indication of (multiple) specific RS resources associated with the at least one predicted beam, and / or a prediction quality metric or a prediction model performance metric associated with the at least one predicted beam.

[0314] After reporting, as an option, operation 4.2 may be extended to optional operation 4.2', which may include user terminal 122 starting measurement of RS resources associated with at least one predicted beam of prediction set 314. Operations 4.4 and 4.2' may be performed in any order or simultaneously. In some example embodiments, fourth operation 4.4 may be deferred until after optional operation 4.2' and / or performed before optional operation 4.2'.

[0315] The fifth operation 4.5 may include the network entity 120 determining a target TCI state corresponding to at least one predicted beam (and / or a quality metric or a model performance accuracy metric), wherein the target TCI state includes an indication of (multiple) specific RS resources associated with the at least one predicted beam.

[0316] A sixth operation 4.6 may include the network entity 120 sending an indication of a target TCI state to the user terminal 122, including an indication of specific RS resource(s) associated with at least one predicted beam.

[0317] A seventh operation 4.7 may include the user terminal 122 receiving an indication of a target TCI state.

[0318] An eighth operation 4.8 may include the user terminal 122 determining whether the indicated target TCI state is known or unknown. This may include the user terminal determining whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least the predicted beam and the measured first RS resource set.

[0319] For example, the user terminal 122 may implement reference to the target TCI state in determining whether the target TCI state is known or unknown. Figure 3 The conditions 316 described use (but are not limited to) for example, measurements of the first reference signal resource set 310, quality metrics of the predicted beams in the prediction set 314 and / or predictions of model performance (or accuracy) metrics; determining monitoring accuracy metrics associated with previous and current predicted beams in the prediction set 314; and / or measurements of reference signal resources for the predicted beams in the prediction set 314; and the like, combinations thereof, modifications thereof and / or as required by the application.

[0320] Thus, using condition 316, user terminal 122 can determine that the target TCI state is known.

[0321] A ninth operation 4.9 may include the user terminal 122 switching to the target TCI state if the target TCI state is known.

[0322] Switching to the target TCI state may be performed within a period of a known TCI state switching delay associated with the target TCI state.The network entity 120 may be configured to switch to the target TCI state simultaneously.

[0323] As an optional embodiment or example embodiment of operations 4.1 to 4.9, a modification may be performed, wherein, in addition to the eighth operation 4.8, the optional modified eighth operation 4.8' may include the network entity 120 being configured to determine whether the indicated target TCI state is known or unknown. This may include the network entity 120 determining whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following items: at least the predicted beam, the configured first set of RS resources 310, and / or the measured first set of RS resources 310 if reported by the user terminal 122.

[0324] For example, the network entity 120 may implement Figure 3The one or more conditions 316 are described. Determining whether the target TCI state is known or unknown, or may be known or unknown to the user terminal 122, using (but not limited to), for example, the predicted beams of the prediction set 314, the measurement of the first RS resource set 310 (if reported by the user terminal 122), the prediction of the quality indicator and / or model performance (or accuracy) indicator of the predicted beams in the prediction set 314 (if reported by the user terminal 122); determining the monitoring accuracy indicator associated with the previous and current predicted beams in the prediction set 314 (if reported by the user terminal); and / or the measurement of the RS resources of the predicted beams in the prediction set 314 (e.g., operation 4.2') (if reported by the user terminal 122); the like, combinations thereof, modifications thereof, and / or as required by the application. Thus, using the condition 316, the network entity 120 can determine that the target TCI state is known.

[0325] In addition to optional operation 4.8', an optional ninth operation 4.9 may also be included, wherein if the target TCI state is known, the network entity 120 switches to the target TCI state. Optionally, switching to the target TCI state may be performed within a time period of a known TCI state switching delay associated with the target TCI state. Thus, the network entity 120 may be configured to switch to the target TCI state simultaneously with the user terminal 122.

[0326] In some embodiments, if the indicated target TCI is known, the user terminal 122 and / or the network entity 120 may switch to the target TCI state (eg, within or after a period of known TCI state switching delay).

[0327] In some example embodiments, switching to the target TCI state may be performed within (or after) a time period based on a determination indicating that the target TCI state is known. For example, when the target TCI state is known, the time period may be based on a known TCI state switching delay associated with the target TCI state. In an example, switching to the target TCI state within the time period based on the determination that the target TCI state is known may further include one or more of: performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on omitting an L1-RSRP measurement time period required for L1-RSRP measurements of RS resources of the target TCI state; or performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0328] Figure 4bis a flow diagram indicating operation 410 according to one or more example embodiments. Operation 410 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 410 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the operation to be performed. For example, operation 410 may be performed by a network device such as Figure 1 The network entity 120) is executed.

[0329] The first operation 411 may include sending a configuration of an RS resource set for measuring a corresponding beam for the network entity 120 to the user terminal 122 for determining at least one predicted beam, wherein the configured RS resource set is associated with prediction-based beam reporting or prediction-based beam switching.

[0330] The second operation 412 may include sending an indication to the user terminal 122 that a target TCI state corresponding to at least one predicted beam is to be activated or indicated.

[0331] Operation 410 may also include the network entity 120 receiving, from the user terminal 122, an indication of at least one predicted beam generated by the user terminal 122 inputting measurements of the first set of RS resources into a model, wherein the model performs an inference operation to provide data representing the indication of the at least one predicted beam.

[0332] In some example embodiments, whether the target TCI state is known or unknown is determined based on at least one of: at least one predicted beam and the first set of RS resources 310, and based on the determination indicating that the target TCI state is known, the user terminal 122 and / or the network entity 120 switches to the target TCI state within the time period.

[0333] In some embodiments, the network entity 120 may be configured to include determining whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least one predicted beam and a configured first reference signal resource set (and / or if reported by the user terminal 122, a measured first RS resource set), and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known. The one or more conditions may be based on, for example, a reference signal resource set. Figure 3 The various one or more conditions 316 are described.

[0334] As described herein, the user terminal 122 may be configured to determine whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following: at least one predicted beam and the measured first set of RS resources, and switch to the target TCI state within a time period based on a determination indicating that the target TCI state is known. The one or more conditions may be based on reference to Figure 3 The various one or more conditions 316 are described.

[0335] Figure 5a A signaling and processing diagram is shown for a case where the network entity 120 uses the model 312 to perform inference operations, according to some example embodiments.

[0336] The first to second operations 5 . 1 to 5 . 2 may correspond to the first to second operations 4 . 1 to 4 . 2 described above with respect to FIG. 4 .

[0337] The third operation 5.3 may include the user terminal 122 reporting to the network entity 120 measurements on the first set of RS resources 310 for use in the prediction / inference operation.

[0338] The fourth operation 5.4 may include: the network entity 120 uses the measurement results of the first RS resource set 310 reported in the third operation 5.3 to perform an inference operation. The inference operation is performed by the network entity 120 using a model configured to predict at least one predicted beam and / or associated RS resources in the RS resource set, for use by the network entity 120 in determining a target TCI state associated with at least one predicted beam of the prediction set 314. The prediction result of the model may include, but is not limited to, at least one of the following items: data representing an indication of at least one predicted beam and / or associated RS resources; at least one of data representing a prediction quality metric associated with at least one predicted beam, a prediction model performance metric associated with at least one predicted beam, a monitoring accuracy metric associated with at least one predicted beam, and / or any other desired metric, so that the user terminal 122 can use the reference data as shown in FIG. Figure 3 One or more conditions 316 are described for determining whether the target TCI state is known or unknown.

[0339] A fifth operation 5.5 may include the network entity 120 determining a target TCI state corresponding to the at least one predicted beam, wherein the target TCI state includes an indication of specific RS resource(s) associated with the at least one predicted beam.

[0340] The sixth operation 5.6′ may include the network entity 120 reporting data representing the prediction result to the user terminal 122. This may include, but is not limited to, data representing at least one of the following items: at least one predicted beam reported to the user terminal 122, an indication of (a plurality of) specific RS resources associated with the at least one predicted beam, a prediction quality metric or a prediction model performance metric associated with the at least one predicted beam, a monitoring accuracy metric associated with the at least one predicted beam, and / or any other desired metric and / or data, so that the user terminal 122 can use the data as described in the reference. Figure 3 The one or more conditions 316 are used to determine whether the target TCI state is known or unknown.

[0341] The sixth operation 5.6' may be performed before and / or after operation 5.5. The sixth operation 5.6' may be performed by the network entity 120 using explicit messaging using control signaling and / or other types of messaging, or implicitly within another messaging format or type, such as a TCI status indication.

[0342] The seventh operation 5.7 may include the network entity 120 sending an indication of the target TCI state to the user terminal 122, including an indication of the specific RS resource(s) associated with the at least one predicted beam.

[0343] As an option, the sixth operation 5.6′ may be performed explicitly or implicitly as a modified sixth operation 5.6″, wherein the network entity 120 reports data representing at least one predicted beam to the user terminal 122, which may include but is not limited to, for example, data representing an indication of at least one predicted beam and / or associated RS resources; at least one of data representing a prediction quality metric associated with the at least one predicted beam, a prediction model performance metric associated with the at least one predicted beam, a monitoring accuracy metric associated with the at least one predicted beam, and / or any other desired metric, so that the user terminal 122 can use the data as described in reference to FIG. Figure 3 One or more conditions 316 are described for determining whether the target TCI state is known or unknown. A modified sixth operation 5.6" may be performed by the network entity 120 when sending the TCI state indication to the user terminal 122 within the messaging used by operation 5.7.

[0344] Alternatively, the sixth operation 5.6' or the modified sixth operation may be performed after operation 5.7.

[0345] The eighth, ninth and tenth operations 5.8, 5.9, 5.10 may correspond to the seventh, eighth and ninth operations 4.7, 4.8, 4.9 performed by the user terminal as described above with respect to FIG. 4 .

[0346] After operations 5.6′, 5.6″, as an option, once the user terminal 122 has received at least one predicted beam and associated RS resources of the prediction set 314 from the network entity 120, operation 5.2 may be extended to operation 5.2′, which may include the user terminal 122 starting to measure RS resources associated with the at least one predicted beam of the prediction set 314. Operation 4.2′ may be performed at any time after operation 5.6′ or 5.6″.

[0347] Switching to the target TCI state may be performed within a period of a known TCI state switching delay associated with the target TCI state.The network entity 120 may be configured to switch to the target TCI state simultaneously.

[0348] As an optional embodiment or example embodiment of operations 5.1 to 5.9, a modification may be made, wherein, in addition to the ninth operation 5.9, the optional modified ninth operation 5.9' may include the network entity 120 being configured to determine whether the indicated target TCI state is known or unknown. This is because the network entity 120 implements the model 312, and wherein the network entity 120 may have data available therefor, which represents (but is not limited to) measurements of the first set of RS resources 310 reported to the network entity 120 in operation 5.3; and data representing an indication of at least one predicted beam and / or associated RS resources of the predicted set 314 predicted by the model 312 at the network entity 120 in operation 5.4; and other data that may be calculated in operation 5.4 or any other operation at the network entity 120, including but not limited to, for example, at least one of data representing a predicted quality metric associated with at least one predicted beam, a predicted model performance metric associated with at least one predicted beam, a monitoring accuracy metric associated with at least one predicted beam, and / or any other desired metric, so that the network entity 120 can use the data as described in reference to FIG. Figure 3 One or more conditions 316 described for determining whether the target TCI state is known or unknown. In operation 5.9', the network entity 120 may be configured to determine whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following items: at least the predicted beam of the prediction set 316 in operation 5.4, the configured first RS resource set 310 (e.g., operation 5.1), and / or the measured first RS resource set 310 reported by the user terminal 122 in operation 5.3.

[0349] For example, the network entity 120 may determine whether the target TCI state is known or unknown or may be known or unknown by the user terminal 122 using (but not limited to) the predicted beams of the prediction set 314 predicted by the model 312 in operation 5.4, the measurement of the first RS resource set 310 reported by the user terminal 122 in operation 5.3, the quality metric and / or model performance (or accuracy) metric of the predicted beams in the prediction set 314 calculated by the network entity 120 in operation 5.5 as necessary, and / or the prediction reported by the user terminal 122; determining the monitoring accuracy metric associated with the previous and current predicted beams in the prediction set 314 calculated in operation 5.5 and / or the measurement of RS resources using the predicted beams in the prediction set 314 (e.g., operation 5.2') (if reported by the user terminal 122 after or during operation 5.2'); and the like, combinations thereof, modifications thereof, and / or according to application requirements. Thus, using condition 316, the network entity 120 may determine that the target TCI state is known.

[0350] In addition to optional operation 5.9', an optional tenth operation 5.10 may also be included, wherein if the target TCI state is known, the network entity 120 switches to the target TCI state. Optionally, switching to the target TCI state may be performed within a time period of a known TCI state switching delay associated with the target TCI state. Thus, the network entity 120 may be configured to switch to the target TCI state simultaneously with the user terminal 122.

[0351] In some embodiments, if the indicated target TCI is known, the user terminal 122 and / or the network entity 120 may switch to the target TCI state (eg, within or after a period of known TCI state switching delay).

[0352] In some example embodiments, switching to the target TCI state may be performed within (or after) a time period based on a determination indicating that the target TCI state is known. For example, when the target TCI state is known, the time period may be based on a known TCI state switching delay associated with the target TCI state. In an example, switching to the target TCI state within the time period based on the determination that the target TCI state is known may further include one or more of: performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on an L1-RSRP measurement time period required to omit L1-RSRP measurements of RS resources of the target TCI state; or performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on a determination that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0353] Figure 5b is a flow chart indicating operation 510 according to one or more example embodiments. Operation 510 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 510 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the operation to be performed. For example, operation 510 may be performed by a network device such as Figure 1 The network entity 120) is executed.

[0354] In a first operation 511, a configuration of a reference signal resource set for measuring a corresponding beam for the network entity 120 is sent to the user terminal 122 for determining at least one predicted beam, wherein the configured reference signal resource set is associated with prediction-based beam reporting or prediction-based beam switching.

[0355] In a second operation 512, measurements of a first set of RS resources 310 for a first set of beams are received from the user terminal 122 for use in determining at least one predicted beam from a second set of beams using a model, wherein the model performs an inference operation to provide data representative of an indication of the at least one predicted beam. The network entity 120 may input the measurements of the first set of RS resources 310 into the model 312, which performs a model inference operation to provide data representative of an indication of the at least one predicted beam.

[0356] In a third operation 513, an indication of at least one predicted beam is sent to the user terminal 122 for use by the user terminal 122 in determining whether the target TCI state (when activated or indicated) is known or unknown based on one or more conditions associated with at least one of the following items: at least the predicted beam and the measured first RS resource set 310.

[0357] In a fourth operation 514, an indication is sent to the user terminal that the target TCI state corresponding to the at least one predicted beam is to be activated or indicated.

[0358] In some example embodiments, whether the target TCI state is known or unknown is determined based on at least one of: at least one predicted beam and the first set of RS resources 310, and is switched to the target TCI state by the user terminal 122 and / or the network entity 120 based on determining that the target TCI state is known.

[0359] In some example embodiments, whether the target TCI state is known or unknown is determined based on at least one of: at least one predicted beam and a first set of RS resources, and switching to the target TCI state within a time period by the user terminal 122 and / or the network entity 120 is based on determining that the target TCI state is known.

[0360] In some embodiments, the network entity 120 may be configured to include determining whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least one predicted beam and a configured first reference signal resource set (and / or a first RS resource set measured when reported by the user terminal 122 in operation 512), and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known. The one or more conditions may be based on, for example, a reference signal resource set. Figure 3 The various one or more conditions 316 are described.

[0361] In some embodiments, the user terminal 122 may be configured to determine whether to enable a user to enable a service based on one or more conditions 316 (e.g., reference number 316) associated with at least one of the following: Figure 3 and / or Figure 5a The method further comprises: determining whether the target TCI state is known or unknown by determining whether the target TCI state is known, and switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known.

[0362] As described herein, the user terminal 122 may be configured to determine whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least one predicted beam and a first set of measured RS resources, and switch to the target TCI state within a time period based on a determination indicating that the target TCI state is known. The one or more conditions may be based on reference to Figure 3 The various one or more conditions 316 are described.

[0363] Figure 6a is a flow chart indicating operation 600 according to one or more example embodiments. Operation 600 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 600 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the operation to be performed. For example, operation 600 may be performed by a user terminal 122 (such as Figure 1 122) is executed by the user terminal.

[0364] The first operation 601 may include measuring a first RS resource set 310 of a configured RS source set for a corresponding beam of the network entity 120 for determining at least one predicted beam. The configured RS resource set is associated with prediction-based beam reporting or prediction-based beam switching.

[0365] The second operation 602 may include receiving an indication from the network entity 120 that a target TCI state corresponding to at least one predicted beam is to be activated or indicated.

[0366] The third operation 603 may include determining whether the target TCI state is known or unknown based on at least one of the following: at least one predicted beam and the measured first RS resource set.

[0367] A fourth operation 604 may include switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known.

[0368] In some example embodiments, a configured set of reference signal resources (hereinafter referred to as RS resources) may include a set of one or more RSs associated with corresponding beams that a network entity may transmit.

[0369] In some example embodiments, the feature(s) executed based on a time period or time interval are as follows: Figure 1 – As described in Example 5b, it also applies to Figure 6a .

[0370] In some example embodiments, switching to the target TCI state within (or after) a time period is performed based on a determination that the target TCI state is known. For example, when the target TCI state is known, the time period may be based on a known TCI state switching delay associated with the target TCI state. In examples, switching to the target TCI state within the time period based on a determination that the target TCI state is known may further include one or more of: performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on omitting an L1-RSRP measurement time period required for L1-RSRP measurements of RS resources of the target TCI state; or performing a TCI state switch within the time period based on the known target TCI state, wherein the time period is calculated based on a determination that at least one L1-RSRP measurement or measurement sample associated with reference signal resources of the target TCI state has been obtained.

[0371] In some example embodiments, for example, additional operations may include determining a predefined duration as described above and updating it if necessary. The predefined duration may be based on, but is not limited to, milliseconds, time slots, frames, or any other time unit suitable for the target TCI state switch. Prior to receiving the TCI state switch indication in operation 602, the predefined duration may be set to N milliseconds, time slots, or frames, where N>0.

[0372] In some embodiments, the operation 600 may further include receiving a configuration of an RS resource set for measuring a corresponding beam for the network entity 120. The configuration may indicate a first RS resource set among the configured RS resource sets for measurement.

[0373] In some embodiments, operation 600 may further include the user terminal 122 being configured to generate a predicted beam set for the network entity 120 based on applying (or inputting) the measured first RS resource set 310 to a model. The model includes an inference operation for predicting an indication of at least one predicted beam that can be used for TCI state switching for the network entity 120. The inference operation may also output one or more predicted quality metrics or model performance metrics associated with the at least one predicted beam.

[0374] In some embodiments, the operations 600 may further include the user terminal 122 being configured to report the predicted set 314 including the at least one predicted beam to the network entity 120 .

[0375] In some embodiments, operation 600 may further include the user terminal 122 being configured to determine whether the target TCI state is known or unknown based on the RS of the predicted beam being in the first set and the RS being associated with the RS of the target TCI state. The determination may be based on one or more conditions associated with at least one of the following: at least one predicted beam and / or the measured first RS resource set. The one or more conditions may include, for example, reference Figure 3 and / or one or more conditions 316 as described herein. Other determinations of whether the target TCI state is known may include comparing one or more predicted quality metrics or model performance metrics associated with at least one predicted beam to be above or below a predefined quality metric threshold or model performance metric threshold, respectively.

[0376] In some example embodiments, Figure 1 –5b The features described in the example also apply to Figure 6a .

[0377] Figure 6bis a flow diagram indicating operation 610 according to one or more example embodiments. Operation 610 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 610 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the performance of the operations. For example, after user terminal 122 reports a predicted beam from prediction set 314 to network entity 120, operation 610 may be performed by user terminal 122 (such as Figure 1 122) is executed by the user terminal.

[0378] The first operation 611 may include the user terminal 122 starting a timer after reporting the last predicted beam in the predicted beam set 314 to the network entity 120 .

[0379] The second operation 612 may include receiving an indication from the network entity 120 that a TCI state corresponding to at least one predicted beam is to be activated or indicated.

[0380] The third operation 613 may include determining whether the target TCI state is known or unknown based on the TCI state received within a predefined duration after the start timer and the RS of the predicted beam corresponding to the RS of the target TCI state. Figure 3 The associated conditions 316 perform one or more determinations.

[0381] A fourth operation 614 may include switching to a target TCI state based on the TCI state received within the predefined duration and a determination that the TCI state is known.

[0382] In some example embodiments, Figure 1 –6a The features described in relation to the example also apply to Figure 6b .

[0383] In some example embodiments, for example, additional operations may include determining a predefined duration as described above and updating it if necessary. The predefined duration may be, but is not limited to, based on milliseconds, time slots, frames, or any other time unit suitable for the target TCI state switch. Prior to receiving the TCI state switch indication in operation 602, the predefined duration may be set to N milliseconds, time slots, or frames, where N>0.

[0384] Figure 6c6 is a flow diagram indicating operation 620 according to one or more example embodiments. Operation 620 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 620 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the operation to be performed. For example, after user terminal 122 learns at least one predicted beam from prediction set 314, user terminal 122 may perform operation 620.

[0385] The first operation 621 may include the user terminal 122 starting a timer after reporting at least one predicted beam to the network entity.

[0386] The second operation 621 may include triggering measurement of at least one RS resource corresponding to at least one predicted beam within a predefined measurement duration using a timer.

[0387] In some embodiments, the second operation 621, the measurement of at least one RS resource associated with at least one predicted beam can be performed based on one or more of the following groups: measuring the RS resource associated with at least one predicted beam in the first time slot of a specific RS resource that occurs after reporting the predicted beam associated with the RS resource; or measuring the RS resource associated with at least one predicted beam within a predetermined or specified time period after reporting the predicted beam associated with the RS resource.

[0388] In some embodiments, the second operation 621 , after measuring RSs associated with at least one predicted beam, performs prediction-based beam switching based on whether the target TCI state is known.

[0389] In some embodiments, the measurement of the at least one measured RS resource is an L1 physical layer measurement.Alternatively, or additionally, the L1 physical layer measurement is an RSRP measurement.

[0390] In some example embodiments, Figures 1 to 6b The features described in relation to the examples also apply to Figure 6c .

[0391] In some example embodiments, for example, additional operations may include determining a predefined measurement duration as described above and updating it if necessary. In some example embodiments, the predefined measurement duration is based on the first time slot of the RS after reporting the corresponding predicted beam, or based on a predetermined or specified time period after reporting the predicted beam of the prediction set 314. The predefined measurement duration may have, but is not limited to, a unit of time based on milliseconds, time slots, frames, or any other unit of time suitable for the target TCI state transition. The predefined measurement duration may be set to X milliseconds, time slots, or frames, where X>0. The measurement is an L1 measurement or an L1RSRP measurement.

[0392] In some example embodiments, Figures 1 to 6b The features described in relation to the examples also apply to Figure 6c .

[0393] Figure 6d is a flow diagram indicating operation 630 according to one or more example embodiments. Operation 630 may be performed in hardware, software, firmware, or a combination thereof. For example, operation 630 may be performed individually or collectively by components, wherein the components may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the operation to be performed. For example, operation 630 may be performed by a network device such as Figure 1 The network entity 120) is executed.

[0394] The first operation 631 may include the network entity 120 sending a configuration of a first reference signal resource set for measuring a corresponding beam for a network device to the user terminal 122 for determining at least one predicted beam. The configured reference signal resource set is associated with prediction-based beam reporting or prediction-based beam switching.

[0395] The second operation 632 may include the network entity 120 sending an indication to the user terminal 122 that a target TCI state corresponding to at least one predicted beam is to be activated or indicated.

[0396] The third operation 633 may include the network entity 120 switching to a target TCI state within a time period. This may be within a time period determined based on an indication that the target TCI state is known. The target TCI state (whether known or unknown) is determined based on at least one of the following: at least one predicted beam and a first set of reference signal resources.

[0397] In some embodiments of operation 630, the user terminal 122 is configured to determine whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least one predicted beam in the prediction set 314 and the measured first RS set 310, and switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known. The one or more conditions 316 may be based on, for example, reference Figures 3 to 5b Those various one or more conditions 316 , combinations thereof, modifications thereof, as described herein and / or as required by the application.

[0398] The second or third operation 632 or 633 may also include the network entity 120 determining whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least one predicted beam and the configured first set of RS resources 310 and / or the measured first set of RS resources (if reported by the user terminal 122) and switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known. The one or more conditions 316 may be based on the reference Figures 3 to 5b The various one or more conditions 316 described above, wherein the network entity 120 is configured to implement the one or more conditions 316. Similarly, the user terminal 122 may use the one or more conditions 316 as described in reference Figures 3 to 5b One or more conditions 316 are described to determine whether the TCI status is known or unknown.

[0399] In some example embodiments, operation 633 may further include network entity 120 switching to the target TCI state within the time period based on the determination that the target TCI state is known. User terminal 122 may also switch to the target TCI state within the same time period based on its determination that the target TCI state is known.

[0400] For example, operation 633 may also include the network entity 120 switching based on the following items: performing TCI state switching within the time period according to a known target TCI state, wherein the time period is calculated based on omitting the L1-RSRP measurement time period required for L1-RSRP measurement of RS resources for the target TCI state; or performing TCI state switching within the time period according to the known target TCI state, wherein the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample related to the reference signal resources of the target TCI state has been obtained.

[0401] In some example embodiments, operation 630 may include model 312 implemented by user terminal 122, wherein operation 630 further includes network entity 120 receiving from user terminal 122 an indication of at least one predicted beam generated by user terminal 122 and inputting measurements of the first set of RS resources 310 into model 312, wherein model 312 performs inference operations to provide data representing the indication of the at least one predicted beam.

[0402] In other example embodiments, operation 630 may include model 312 implemented by network entity 120, wherein operation 630 also includes network entity 120 receiving measurements of a first set of RS resources 310 for a first group of beams from user terminal 122 for determining at least one predicted beam from a second group of beams using model 312, wherein model 312 performs an inference operation to provide data representing an indication of at least one predicted beam of the predicted set 314.

[0403] In some embodiments, operation 630 may further include network entity 120 inputting data representing the measured first set of RS resources 310 (e.g., received from user terminal 122 or even measured by network entity 120) into the model. Operation 630 may also include network entity 120 reporting data representing an indication of at least one predicted beam to user terminal 122.

[0404] Operation 630 of the network entity 120 may also include sending data representing an indication of the at least one predicted beam to the user terminal 122 for use by the user terminal 122 in determining whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following items and switching to the target TCI state based on the determination. The one or more conditions 316 may be based on the reference Figures 3 to 5b Those various one or more conditions 316 are described.

[0405] Operation 630 of the network entity 120 may also include performing a determination of whether the target TCI state is known or unknown based on one or more conditions 316 associated with at least one of the following: at least one predicted beam and the configured first set of RS resources 310 and / or the measured first set of RS resources (if reported by the user terminal 122), and switching to the target TCI state within the time period based on the determination indicating that the target TCI state is known. The one or more conditions 316 may be based on, for example, reference Figures 3 to 5b The various one or more conditions 316 described above, wherein the network entity 120 is configured to implement the one or more conditions 316. Similarly, the user terminal 122 may use the one or more conditions 316 as described in reference Figures 3 to 5bThe one or more conditions 316 determine whether the TCI status is known or unknown.

[0406] In some example embodiments, Figures 1 to 6c The features described in relation to the examples also apply to Figure 6d .

[0407] Example device

[0408] Figure 7 An example apparatus 700 capable of supporting at least some embodiments is shown. Device 700 is illustrated, which may include the user terminal 122 described above or a device configured to control its functionality. Alternatively, device 700 may include the network entity 120 described above. Device 700 includes a processor 710, which may include, for example, a single-core or multi-core processor, wherein a single-core processor includes one processing core and a multi-core processor includes more than one processing core. Processor 710 may generally include a control device. Processor 710 may include more than one processor. Processor 710 may be a control device. A processing core may include, for example, a Cortex-A8 processing core manufactured by ARM Holdings or a Steamroller processing core manufactured by Advanced Micro Devices. Processor 710 may include at least one Qualcomm Snapdragon and / or Intel Atom processor. Processor 710 may include at least one application-specific integrated circuit (ASIC). Processor 710 may include at least one field-programmable gate array (FPGA). Processor 710 may be a component for executing the method steps in device 800. Processor 710 may be configured, at least in part, by computer instructions to perform actions.

[0409] The processor may include circuitry, or may be constructed as one or more circuitry configured to perform the stages of the method according to the embodiments described herein. As used in this application, the term "circuitry" may refer to one or more or all of the following: (a) a hardware circuitry implementation only, such as an implementation of analog and / or digital circuitry only, and (b) a combination of hardware circuitry and software, such as, as applicable: (i) a combination of analog and / or digital hardware circuitry with software / firmware, and (ii) any portion of a hardware processor(s) with software (including digital signal processor(s), software, and memory(s) that work together to enable an apparatus (such as network entity 120) or a device configured to control its functionality to perform various functions) and (c) hardware circuitry and / or processor(s), such as microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) to operate, but where the software is not required for operation, the software may not be present.

[0410] This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term "circuitry" also covers an implementation of merely a hardware circuit or processor (or multiple processors) or a portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term "circuitry" also covers, for example, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in a server, cellular network device, or other computing or network device, if applicable to the particular claim element.

[0411] Device 700 may include memory 720. Memory 720 may include random access memory and / or permanent memory. Memory 720 may include at least one RAM chip. Memory 720 may include, for example, solid-state, magnetic, optical, and / or holographic memory. Memory 720 may be at least partially accessible to processor 710. Memory 720 may be at least partially included in processor 710. Memory 720 may be a component for storing information. Memory 720 may include computer instructions that processor 710 is configured to execute. When computer instructions configured to cause processor 710 to perform certain actions are stored in memory 720, and device 700 is generally configured to operate under the direction of processor 710 using computer instructions from memory 720, processor 710 and / or at least one of its processing cores may be considered to be configured to perform the certain actions. Memory 720 may be at least partially included in processor 710. Memory 720 may be at least partially external to device 700, but accessible to device 700.

[0412] The device 700 may include a transmitter 730. The device 700 may include a receiver 740. The transmitter 730 and the receiver 740 may be configured to transmit and receive information according to at least one cellular or non-cellular standard, respectively.

[0413] The transmitter 730 may include more than one transmitter. The receiver 740 may include more than one receiver. The transmitter 730 and / or the receiver 740 may be configured to operate according to, for example, Global System for Mobile Communications (GSM), Wideband Code Division Multiple Access (WCDMA), 5G / NR, Advanced 5G (i.e., NR Rel-18, 19 and later), Long Term Evolution (LTE), IS-95, Wireless Local Area Network (WLAN), Ethernet, and / or Worldwide Interoperability for Microwave Access (WiMAX) standards.

[0414] Device 700 may include a near field communication (NFC) transceiver 750. NFC transceiver 750 may support at least one NFC technology, such as NFC, Bluetooth, Wibree, or similar technology.

[0415] Device 700 may include a user interface (UI) 760. UI 760 may include at least one of a display, a keyboard, a touch screen, a vibrator arranged to send a signal to the user by vibrating device 700, a speaker, and a microphone. The user can operate device 700 via UI 760, for example, answer an incoming call, make a phone or video call, browse the internet, manage digital files stored in memory 720 or in a cloud accessible via transmitter 730 and receiver 740 or via NFC transceiver 750, and / or play games.

[0416] Device 700 may include or be arranged to accept a user identity module 770. User identity module 770 may include, for example, a subscriber identity module (SIM) card that may be installed in device 700. User identity module 770 may include information identifying a subscription of a user of device 700. User identity module 770 may include encrypted information that may be used to verify the identity of the user of device 700 and / or facilitate encryption of transmitted information and billing of the user of device 700 for communications conducted via device 700.

[0417] The processor 710 may be equipped with a transmitter that is arranged to output information from the processor 710 to other devices included in the device 700 via electrical leads within the device 700. Such a transmitter may include a serial bus transmitter, arranged to output information to the memory 720 for storage therein, for example, via at least one electrical lead. In addition to a serial bus, the transmitter may include a parallel bus transmitter.

[0418] Likewise, the processor 710 may include a receiver arranged to receive information in the processor 710 from other devices included in the device 700 via electrical leads internal to the device 700. Such a receiver may include a serial bus receiver arranged to receive information from the receiver 740, for example, via at least one electrical lead, for processing in the processor 710. In addition to a serial bus, the receiver may include a parallel bus receiver.

[0419] The device 700 may include Figure 7 700 may include additional devices not shown. For example, when device 700 comprises a smartphone, it may include at least one digital camera. Some devices 700 may include a rear-facing camera and a front-facing camera, wherein the rear-facing camera may be intended for digital photography and the front-facing camera is used for video calling. Device 700 may include a fingerprint sensor that is arranged to at least partially authenticate the user of device 700. In some embodiments, device 700 lacks at least one of the aforementioned devices. For example, some devices 700 may lack an NFC transceiver 750 and / or a user identity module 770.

[0420] The processor 710, memory 720, transmitter 730, receiver 740, NFC transceiver 750, UI 760, and / or user identity module 770 can be interconnected in a variety of different ways via electrical leads within the device 700. For example, each of the above devices can be separately connected to a main bus within the device 700 to allow the devices to exchange information. However, as will be understood by those skilled in the art, this is merely an example, and depending on the embodiment, various methods may be selected to interconnect at least two of the above devices without departing from the scope of the present invention.

[0421] Figure 8 A non-transitory medium 800 is shown according to some embodiments. The non-transitory medium 800 is a computer-readable storage medium. It can be, for example, a CD, a DVD, a USB stick, a Blu-ray disc, etc. The non-transitory medium 800 stores computer program instructions that cause an apparatus to perform any of the aforementioned methods, such as those disclosed in connection with the flowcharts and related features in this specification.

[0422] The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the foregoing description, many specific details are provided, such as examples of lengths, widths, shapes, etc., to provide a thorough understanding of the embodiments of the present invention. However, those skilled in the relevant art will recognize that the present invention can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other cases, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the present invention.

[0423] Although the above examples illustrate the principles of the embodiments in one or more specific applications, it will be apparent to those skilled in the art that many modifications can be made to the form, use, and implementation details without inventiveness and without departing from the principles and concepts of the invention. Therefore, the present invention is not intended to be limited except by the following claims.

[0424] The verbs "to comprise" and "to include" are used in this document as open limitations that neither exclude nor require the presence of unrecited features. Unless expressly stated otherwise, the features recited in the dependent claims are freely combinable with each other. Furthermore, it should be understood that the use of "a" or "an" in this document, i.e., in the singular, does not exclude a plural reference.

Claims

1. A device for communication, comprising: means for measuring a first set of reference signal resources of a configured set of reference signal resources for a corresponding beam of a network entity for determining at least one predicted beam, wherein the configured set of reference signal resources is associated with prediction-based beam reporting or prediction-based beam switching; means for receiving an indication from the network entity that a target transmission configuration indicator (TCI) state corresponding to the at least one predicted beam is to be activated or indicated; means for determining whether the target TCI state is known or unknown based on at least one of: the at least one predicted beam and the measured first set of reference signal resources; as well as Means for switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

2. The device according to claim 1, wherein The configured reference signal resource set is configured by the network entity.

3. The apparatus according to claim 1 or 2, further comprising: means for receiving a message from the network entity for triggering the apparatus to perform the prediction-based beam reporting or the prediction-based beam switching, wherein the measurement is performed upon or after receipt of the message.

4. The device according to any one of claims 1 to 3, wherein The components for measuring are configured to: The first set of reference signal resources for a first set of beams is measured for use in determining the at least one predicted beam from a second set of beams using a model, wherein the model performs inference operations for providing data representative of an indication of the at least one predicted beam.

5. The apparatus according to claim 4, further comprising: means for inputting data representing said first set of measured reference signal resources into said model; as well as means for reporting data representative of said indication of said at least one predicted beam to said network entity, At least one reference signal resource in the configured reference signal resource set is associated with the at least one predicted beam.

6. The apparatus according to claim 4, further comprising: Means for reporting the measured first set of reference signal resources to the network entity for input to the model by the network entity.

7. The apparatus according to claim 5 or 6, wherein the means for determining whether the target TCI state is known or unknown further comprises: means for determining that one or more predicted quality metrics of one or more reference signal resources associated with a corresponding reported predicted beam are greater than or equal to a preconfigured or predefined quality threshold level; as well as In response to determining that at least one of the one or more predicted quality metrics of the one or more reference signal resources associated with the corresponding reported predicted beam is greater than or equal to the preconfigured or predefined quality threshold level, when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state, it is determined that the target TCI state is known.

8. The apparatus of claim 7, wherein the quality metric comprises one or more of the following group: Reference signal received power RSRP; L1 physical layer RSRP L1-RSRP; The signal-to-noise ratio (SNR) of the reference signal resource; Any other quality metric associated with the reference signal resource.

9. The apparatus according to claim 7 or 8, wherein the means for determining whether the target TCI state is known or unknown further comprises: means for determining that one or more current model performance metrics corresponding respectively to one or more of said reported predicted beams associated with one or more reference signal resources are greater than or equal to a preconfigured or predefined model performance threshold level; In response to determining that at least one of the one or more current model performance metrics of the one or more reported predicted beams correspondingly associated with one or more reference signal resources is greater than or equal to the preconfigured or predefined model performance threshold level, when the one or more reference signal resources correspond to reference signal resources associated with the target TCI state, it is determined that the target TCI state is known.

10. The apparatus of claim 9, wherein the model performance metric comprises one or more of the following group: Model prediction accuracy measures; precision and recall metrics; Area under the curve receiver operating curve AUC-ROC metric; Model confidence measures; Probability score measure; Any other model performance metric used to evaluate the predictive performance of the model.

11. The apparatus of any preceding claim, wherein the means for switching to the target TCI state within the time period based on the determination that the target TCI state is known further comprises one or more of: performing the TCI state switching within the time period according to the known target TCI state, wherein the time period is calculated based on an L1-RSRP measurement time period required for omitting L1-RSRP measurement of the reference signal resource of the target TCI state; or The TCI state switching is performed within the time period according to the known target TCI state, wherein the time period is calculated based on determining that at least one L1-RSRP measurement or measurement sample related to the reference signal resource of the target TCI state has been obtained.

12. The apparatus of any preceding claim, wherein when the apparatus is configured to perform prediction-based beam reporting or prior to prediction-based beam switching, the means for measuring is further configured to: triggering, after reporting data representing an indication of the one or more predicted beams, measurement of at least one of the reference signal resources associated with at least one of the predicted beams, wherein the measurement of the at least one of the reference signal resources associated with the at least one of the predicted beams is performed based on one or more from the group consisting of: measuring a reference signal resource associated with at least one predicted beam in a first time slot of the specific reference signal resource occurring after reporting the predicted beam associated with the reference signal resource; or Reference signal resources associated with at least one predicted beam are measured within a predetermined or specified time period after reporting the predicted beam associated with the reference signal resource.

13. The device according to claim 12, wherein After measuring the reference signal associated with the at least one predicted beam, prediction-based beam switching is performed based on whether the target TCI state is known.

14. An apparatus according to any preceding claim, wherein: The measurement of the at least one measured reference signal resource is an L1 physical layer measurement; or The L1 physical layer measurement is a reference signal resource received power RSRP measurement.

15. An apparatus according to any preceding claim, wherein: When the apparatus is configured to perform prediction-based beam reporting, the means for determining whether the target TCI state is known or unknown further comprises: means for determining that the first set of reference signal resources is associated with a corresponding reported predicted beam; and means for determining that the target TCI state is known when the reference signal resources in the first reference signal set correspond to reference signal resources associated with the target TCI state.

16. A network device comprising: means for sending, to a terminal device, a first set of reference signal resources for measuring a corresponding beam for the network device for determining a configuration of at least one predicted beam, wherein the configured set of reference signal resources is associated with prediction-based beam reporting or prediction-based beam switching; as well as means for sending an indication to the terminal device that a target transmission configuration indicator (TCI) state corresponding to the at least one predicted beam is to be activated or indicated; The target TCI state, whether known or unknown, is determined based on at least one of the following items: the at least one predicted beam and the first reference signal resource set, and switching to the target TCI state within a time period is based on a determination indicating that the target TCI state is known.

17. The network device according to claim 16, further comprising: The terminal device is configured to determine whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: the at least one predicted beam and the measured first reference signal resource set, and switch to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

18. The network device according to claim 16 or 17, further comprising: means for determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items, and switching to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

19. The network device according to claims 16 to 18, further comprising: Means for switching to the target TCI state within the time period based on the determination that the target TCI state is known.

20. A method for communication, comprising: measuring a first set of reference signal resources from among a set of configured reference signal resources for a corresponding beam of a network entity for determining at least one predicted beam, wherein the set of configured reference signal resources is associated with prediction-based beam reporting or prediction-based beam switching; receiving an indication from the network entity that a target transmission configuration indicator (TCI) state corresponding to the at least one predicted beam is to be activated or indicated; determining whether the target TCI state is known or unknown based on at least one of: the at least one predicted beam and the measured first reference signal resource set; as well as Switching to the target TCI state within a time period is based on a determination indicating that the target TCI state is known.

21. A method for communication, comprising: sending, to a terminal device, a first set of reference signal resources for measuring a corresponding beam for a network device for determining a configuration of at least one predicted beam, wherein the configured set of reference signal resources is associated with prediction-based beam reporting or prediction-based beam switching; as well as sending an indication to the terminal device that a target transmission configuration indicator (TCI) state corresponding to the at least one predicted beam is to be activated or indicated; The target TCI state, whether known or unknown, is determined based on at least one of the following items: the at least one predicted beam and the first reference signal resource set, and switching to the target TCI state within a time period is based on a determination indicating that the target TCI state is known.

22. The method according to claim 21, further comprising: The terminal device is configured to determine whether the target TCI state is known or unknown based on one or more conditions associated with at least one of the following items: the at least one predicted beam and the measured first reference signal resource set, and switch to the target TCI state within a time period based on a determination indicating that the target TCI state is known.

23. The method according to claim 21 or 22, further comprising: determining whether the target TCI state is known or unknown based on one or more conditions associated with at least one of: the at least one predicted beam and the configured first set of reference signal resources, and switching to the target TCI state within the time period based on a determination indicating that the target TCI state is known; as well as As an option, switching to the target TCI state within the time period is performed based on the determination that the target TCI state is known.

24. A computer-readable medium comprising instruction codes stored thereon, which, when executed by one or more processors, cause the one or more processors to perform the method according to claim 20.

25. A computer readable medium comprising instruction codes stored thereon, which, when executed by one or more processors, cause the one or more processors to perform the method according to any one of claims 21 to 23.

Images