Method and apparatus for AI/ML based beam management

1 . A method of wireless communication of a user equipment (UE), comprising:

receiving, from a base station, a first monitoring configuration for monitoring an artificial intelligence/machine learning (AI/ML) model for managing a set of beams;

measuring a first subset of the set of beams, wherein the first subset is selected based on the first monitoring configuration;

performing inference using the AI/ML model based on measurements of the first subset to determine prediction values of the set of beams;

filtering the prediction values of the set of beams to obtain filtered prediction values corresponding to a second subset of the set of beams, wherein the second subset is configured based on the first monitoring configuration and comprises fewer beams than the set of beams;

measuring the second subset of the set of beams to determine measured values of the second subset; and

calculating one or more performance metrics based on the filtered prediction values and the measured values of the second subset, wherein the one or more performance metrics are selected based on the first monitoring configuration.

2 . The method of claim 1 , wherein the one or more performance metrics comprise at least one of:

a beam prediction accuracy metric, a beam prediction ranking accuracy metric, a reference signal received power (RSRP) prediction accuracy metric, a reference signal received quality (RSRQ) prediction accuracy metric, and a signal to interference plus noise ratio (SINR) prediction accuracy.

3 . The method of claim 1 , wherein calculating the one or more performance metrics comprises:

comparing each of the filtered prediction values with a corresponding measured value of the second subset; and

determining the one or more performance metrics based on a result of the comparing.

4 . The method of claim 1 , further comprising:

reporting the one or more performance metrics to the base station.

5 . The method of claim 1 , further comprising:

determining whether the one or more performance metrics meet one or more metrics thresholds; and

reporting, to the base station, one or more indications indicating whether the one or more performance metrics meet the one or more metrics thresholds.

6 . The method of claim 1 , wherein the first monitoring configuration comprise at least one of:

the one or more performance metrics;

a container method parameter specifying a reporting format to be used by the UE when reporting the one or more performance metrics or indications of whether the one or more performance metrics meet one or more metrics thresholds;

a number of resources parameter specifying a number of resources to be measured for calculating the one or more performance metrics;

a monitoring frequency parameter specifying a frequency at which the UE performs monitoring and collects samples;

a metrics threshold parameter specifying values of the one or more metrics thresholds;

a statistical threshold parameter specifying a percentage of monitoring instances that must meet an occurrence criteria over a number of samples to be collected in order to determine model performance;

a number of samples parameter specifying the number of samples to be collected before applying a statistical threshold;

a monitoring duration parameter specifying a duration over which the monitoring is performed; and

an indication of a resource set that contains at least one of the first subset or the second subset.

7 . The method of claim 1 , further comprising:

transmitting, to the base station, UE capabilities indicating support for monitoring the AI/ML model.

8 . The method of claim 7 , wherein the UE capabilities comprise at least one of:

an indication of support for monitoring reporting,

a list of supported performance metrics,

a list of supported reporting formats,

a number of supported resources for the first subset,

a number of supported resources for the second subset, and

a list of supported monitoring frequencies.

9 . The method of claim 1 , further comprising:

receiving at least one additional monitoring configuration, each monitoring configuration associated with a different monitoring frequency, a different resource set, and a different performance metric;

performing monitoring at the different monitoring frequencies by measuring the associated resource sets; and

calculating performance metrics based on the monitoring.

10 . An apparatus for wireless communication, the apparatus being a user equipment (UE), comprising:

a memory; and

at least one processor coupled to the memory and configured to:

receive, from a base station, a first monitoring configuration for monitoring an artificial intelligence/machine learning (AI/ML) model for managing a set of beams;

measure a first subset of the set of beams, wherein the first subset is selected based on the first monitoring configuration;

perform inference using the AI/ML model based on measurements of the first subset to determine prediction values of the set of beams;

filter the prediction values of the set of beams to obtain filtered prediction values corresponding to a second subset of the set of beams, wherein the second subset is configured based on the first monitoring configuration and comprises fewer beams than the set of beams;

measure the second subset of the set of beams to determine measured values of the second subset; and

calculate one or more performance metrics based on the filtered prediction values and the measured values of the second subset, wherein the one or more performance metrics are selected based on the first monitoring configuration.

11 . The apparatus of claim 10 , wherein the one or more performance metrics comprise at least one of: a beam prediction accuracy metric, a beam prediction ranking accuracy metric, a reference signal received power (RSRP) prediction accuracy metric, a reference signal received quality (RSRQ) prediction accuracy metric, and a signal to interference plus noise ratio (SINR) prediction accuracy.

12 . The apparatus of claim 10 , wherein to calculate the one or more performance metrics, the at least one processor is further configured to:

compare each of the filtered prediction values with a corresponding measured value of the second subset; and

determine the one or more performance metrics based on a result of the comparing.

13 . The apparatus of claim 10 , wherein the at least one processor is further configured to:

report the one or more performance metrics to the base station.

14 . The apparatus of claim 10 , wherein the at least one processor is further configured to:

determine whether the one or more performance metrics meet one or more metrics thresholds; and

report, to the base station, one or more indications indicating whether the one or more performance metrics meet the one or more metrics thresholds.

15 . The apparatus of claim 10 , wherein the first monitoring configuration comprise at least one of:

the one or more performance metrics;

a container method parameter specifying a reporting format to be used by the UE when reporting the one or more performance metrics or indications of whether the one or more performance metrics meet one or more metrics thresholds;

a number of resources parameter specifying a number of resources to be measured for calculating the one or more performance metrics;

a monitoring frequency parameter specifying a frequency at which the UE performs monitoring and collects samples;

a metrics threshold parameter specifying values of the one or more metrics thresholds;

a statistical threshold parameter specifying a percentage of monitoring instances that must meet an occurrence criteria over a number of samples to be collected in order to determine model performance;

a number of samples parameter specifying the number of samples to be collected before applying a statistical threshold;

a monitoring duration parameter specifying a duration over which the monitoring is performed; and

an indication of a resource set that contains at least one of the first subset or the second subset.

16 . The apparatus of claim 10 , wherein the at least one processor is further configured to:

transmit, to the base station, UE capabilities indicating support for monitoring the AI/ML model.

17 . The apparatus of claim 16 , wherein the UE capabilities comprise at least one of:

an indication of support for monitoring reporting;

a list of supported performance metrics;

a list of supported reporting formats;

a number of supported resources for the first subset;

a number of supported resources for the second subset; and

a list of supported monitoring frequencies.

18 . The apparatus of claim 10 , wherein the at least one processor is further configured to:

receive at least one additional monitoring configuration, each monitoring configuration associated with a different monitoring frequency, a different resource set, and a different performance metric;

perform monitoring at the different monitoring frequencies by measuring the associated resource sets; and

calculate performance metrics based on the monitoring.

19 . A non-transitory computer-readable medium storing computer executable code for wireless communication of a user equipment (UE), comprising code to:

receive, from a base station, a first monitoring configuration for monitoring an artificial intelligence/machine learning (AI/ML) model for managing a set of beams;

measure a first subset of the set of beams, wherein the first subset is selected based on the first monitoring configuration;

perform inference using the AI/ML model based on measurements of the first subset to determine prediction values of the set of beams;

filter the prediction values of the set of beams to obtain filtered prediction values corresponding to a second subset of the set of beams, wherein the second subset is configured based on the first monitoring configuration and comprises fewer beams than the set of beams;

measure the second subset of the set of beams to determine measured values of the second subset; and

calculate one or more performance metrics based on the filtered prediction values and the measured values of the second subset, wherein the one or more performance metrics are selected based on the first monitoring configuration.

20 . The non-transitory computer-readable medium of claim 19 , wherein the one or more performance metrics comprise at least one of: a beam prediction accuracy metric, a beam prediction ranking accuracy metric, a reference signal received power (RSRP) prediction accuracy metric, a reference signal received quality (RSRQ) prediction accuracy metric, and a signal to interference plus noise ratio (SINR) prediction accuracy.