Methods of set a beam configuration for channel state information reporting in artificial intelligence / machine learning model based beam management

Abstract

Embodiments herein describe systems, methods, and apparatuses for configuration of beam sets for AI / ML models. In some embodiments a UE may receive a first resource configuration including a configuration for a first set of beams and an association identifier (ID). The UE may perform training or performance monitoring for a model based on the first resource configuration. The UE may receive a second resource configuration with the same association ID indicating that the set of beams are linked.

Description

METHODS OF SET A BEAM CONFIGURATION FOR CHANNEL STATEINFORMATION REPORTING IN ARTIFICIAL INTELLIGENCE / MACHINELEARNING MODEL BASED BEAM MANAGEMENTTECHNICAL FIELD

[0001] This application relates generally to wireless communication systems implementing artificial intelligence / machine learning (AI / ML) model(s).BACKGROUND

[0002] Wireless mobile communication technology uses various standards and protocols to transmit data between a base station and a wireless communication device. Wireless communication system standards and protocols can include, for example, 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) (e.g., 4G), 3GPP New Radio (NR) (e.g., 5G), and Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard for Wireless Local Area Networks (WLAN) (commonly known to industry groups as Wi-Fi®).

[0003] As contemplated by the 3GPP, different wireless communication systems’ standards and protocols can use various radio access networks (RANs) for communicating between a base station of the RAN (which may also sometimes be referred to generally as a RAN node, a network node, or simply a node) and a wireless communication device known as a user equipment (UE). 3GPP RANs can include, for example, Global System for Mobile communications (GSM), Enhanced Data Rates for GSM Evolution (EDGE) RAN (GERAN), Universal Terrestrial Radio Access Network (UTRAN), Evolved Universal Terrestrial Radio Access Network (E-UTRAN), and / or Next-Generation Radio Access Network (NG-RAN).

[0004] Each RAN may use one or more radio access technologies (RATs) to perform communication between the base station and the UE. For example, the GERAN implements GSM and / or EDGE RAT, the UTRAN implements Universal Mobile Telecommunication System (UMTS) RAT or other 3GPP RAT, the E-UTRAN implements LTE RAT (sometimes simply referred to as LTE), and NG-RAN implements NR RAT (sometimes referred to herein as 5G RAT, 5G NR RAT, or simply NR). In certain deployments, the E-UTRAN may also implement NR RAT. In certain deployments, NG-RAN may also implement LTE RAT.1P68965WO1 4924-2349-0130U

[0005] A base station used by a RAN may correspond to that RAN. One example of an E-UTRAN base station is an Evolved Universal Terrestrial Radio Access Network (E- UTRAN) Node B (also commonly denoted as evolved Node B, enhanced Node B, eNodeB, or eNB). One example of an NG-RAN base station is a next generation Node B (also sometimes referred to as a g Node B or gNB).

[0006] A RAN provides its communication services with external entities through its connection to a core network (CN). For example, E-UTRAN may utilize an Evolved Packet Core (EPC) while NG-RAN may utilize a 5G Core Network (5GC).BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0007] To easily identify the discussion of any particular element or act, the most significant digit or digits in a reference number refer to the figure number in which that element is first introduced.

[0008] FIG. 1 illustrates a flow diagram of a UE side model training stage and a flow diagram of a UE side model inferencing stage, according to embodiments herein.

[0009] FIG. 2 illustrates an example of a CSI report reported after UE side model inferencing, according to embodiments herein.

[0010] FIG. 3 illustrates a flow diagram of a UE side model monitoring stage and a flow diagram of a UE side model inferencing stage, according to embodiments herein.

[0011] FIG. 4 illustrates a diagram of periodic and / or semi-persistent measurement and CSI reporting, according to embodiments herein.

[0012] FIG. 5 illustrates an example Set B beam and Set A beam reference signal transmission timing, according to embodiments herein.

[0013] FIG. 6 illustrates an example of a CSI report reported after UE side model inferencing, according to embodiments herein.

[0014] FIG. 7 illustrates a diagram of periodic and semi-persistent measurement and reporting for BM-Case2, according to embodiments herein.

[0015] FIG. 8 illustrates a method performed by a UE, according to embodiments herein.

[0016] FIG. 9 illustrates a method performed by a network node, according to embodiments herein.2P68965WO1 4924-2349-0130U

[0017] FIG. 10 illustrates an example architecture of a wireless communication system, according to embodiments disclosed herein.

[0018] FIG. 11 illustrates a system for performing signaling between a wireless device and a network device, according to embodiments disclosed herein.DETAILED DESCRIPTION

[0019] Various embodiments are described with regard to a UE. However, reference to a UE is merely provided for illustrative purposes. The example embodiments may be utilized with any electronic component that may establish a connection to a network and is configured with the hardware, software, and / or firmware to exchange information and data with the network. Therefore, the UE as described herein is used to represent any appropriate electronic component.

[0020] In certain wireless communication systems, for artificial intelligence / machine learning (AI / ML)-based beam management, support is provided for beam management (BM)-Casel and BM-Case2 for characterization and baseline performance evaluations. BM-Casel includes spatial-domain downlink (DL) beam prediction for Set A beams based on measurement results of Set B beams. BM-Case2 includes temporal DL beam prediction for Set A beams based on the historic measurement results of Set B beams. For BM-Casel and BM-Case2, beams in Set A and Set B can be in the same frequency range. Set A beams includes beams for which the AI / ML model generates prediction. Set B beams includes beams that are measured, and the measurements used as inputs to the AI / ML model.

[0021] In some wireless communication systems, for UE-sided AI / ML model(s) for BM-Case-1, a CSI-ReportConfig information element (IE) may be used for the configuration of inference result reporting. However, the contents and details of the CSI- ReportConfig IE may be further considered. In some examples, one CSI- ResourceConfigld IE may be configured for Set B beams only (referred to herein as “Alt 1”). However, in “Alt 1,” how the UE may determine information about the Set A beams may be further considered. In some other examples, one CSI-ResourceConfigld E may be configured for both the Set A beams and the Set B beams (referred to herein as “Alt 2”), however how to configure resource set(s) for Set A beams and Set B beams in the CSI-ResourceConfig IE may be further considered. In yet some other examples, two CSI- ResourceConfigld IES may be configured for Set A beams and Set B beams separately3P68965WO1 4924-2349-0130U(referred to herein as “Alt 3”). In yet some other examples, one CSI-ResourceConfigld IE may be configured for the Set B beams and the Set A beams are configured using separate / different resource set(s) other than by the CSI-ResourceConfigld E (referred to herein as “Alt 4”), however how to configure / indicate separate resource set(s) for Set A beams may be further considered. Note that separate CSI-ReportConfig IES for Set A beams and Set B beams are not precluded, and a measurement may not be performed for Set A beams and a measurement may be performed for Set B beams subject to the CSI- ReportConfig IE. Additionally, the association between Set A beams and Set B beams with or without the use of additional IE(s) may be further considered and other configurations may not be precluded. Note that “Alt 4” may be a special case of “Alt 1” as one CSI-ResourceConfigld E is configured for the Set B beams in both cases.

[0022] In certain wireless communication mechanisms, for UE-sided model(s) developed (e.g., trained, updated) at the UE, the following procedure (“AI-Examplel of Ml-Optionl”) may be further considered (including the feasibility / necessity). The procedure may include the network signaling data collection related configuration(s) and the associated identifiers(s) (ID(s)), where the associated ID(s) for each sub use case may be related to network-side additional conditions (referred to herein as “Step A”). Note that the associated ID(s) may be used to extract the network-side additional conditions as to maintain consistency in the training stage, the performance monitoring stage, and the inference stage of the AI / ML model. The UE(s) may collect the data corresponding to the associated ID(s) (referred to herein as “Step B”). Accordingly, the AI / ML model(s) may be developed (e.g., trained, updated) at the UE based on the collected data corresponding to the associated ID(s) (referred to herein as “Step C”). In some cases, “Step B” and “Step C” may be performed offline. The UE may report information of its AI / ML models corresponding to associated IDs to the network and a model ID may be determined / assigned for each AI / ML model (referred to herein as “Step D”).

[0023] In some cases, in “Step D”, the relationship between the model ID(s) and the associated ID(s) may be reported. However, how model ID(s) are determined / assigned may be further considered. In some examples, the network may assign the model ID(s). In some other examples, the UE assigns / reports the model ID(s). In yet some other examples, the associated ID(s) may be assumed as the model ID(s). In such instances, the determination / assignment of model ID(s) may not be needed. In yet some other4P68965WO1 4924-2349-0130Uexamples, the model ID(s) may be determined by pre-defined rule(s) provided in the 3 GPP specification. Additionally, how to report the model ID(s) may be further considered. Note that “Step D” discussed herein may be used to facilitate AI / ML model inference, and “Steps A / B / C” and additional interact! on(s) of the associated IDs between UE and network may be considered for resolving model consistency without model identification.

[0024] Additionally, regarding the associated ID(s), the UE may assume that the network-side additional conditions with the same associated ID are consistent at least within a cell / base station. However, whether / how the UE assumption can be applicable for multiple cells (including feasibility) may be further considered.

[0025] Some embodiments herein introduce a Set A beam indication for beam management inference in channel state information (CSI) report content. In some embodiments, for “Alt 1” and “Alt 4” discussed herein where only Set B beams are configured by the CSI-ResourceConfigld IE, the Set A beams may be linked to the training configuration by an association ID. In some cases, the training configuration may be configured by a CSI-MeasConfig IE and no report may be associated for the UE side training. In some such cases, a dedicated training data collection for the UE side model may be introduced.

[0026] In some other embodiments, for “Alt 1” and “Alt 4” discussed herein where only Set B beams are configured by the CSI-ResourceConfigld IE, the Set A beams may be linked to a performance monitoring configuration for a ground truth measurement. For example, the UE may report an inference report based on the channel state information reference signal (CSI-RS) resource configuration for performance monitoring.

[0027] Additionally or alternately, a combination of the embodiments discussed herein for “Alt 1” and “Alt 4” may be utilized.

[0028] It should be understood that embodiments discussed herein may apply to both BM-Casel (i.e., in the spatial domain) and BM-Case2 (i.e., in the time domain). Further, embodiments herein improve the consistency between the training stage and the inference stage of the AI / ML model and between the inference stage and the performance monitoring stage of the AI / ML model, as these stages may occur a certain amount of time between each other.

[0029] Embodiments herein relate to data collection for the training of a UE side AI / ML model. In some embodiments, during the data collection for training, for UE side5P68965WO1 4924-2349-0130Umodel, the network may configure a synchronization signal block (SSB) or a CSI-RS measurement without any UE feedback.

[0030] In some embodiments, for UE side model training, the base station may configure a CSI-MeasConfig IE which may include a CSI-SSB-ResourceSet field and NZP-CSI-RS-ResourceSet field. Accordingly, the Set A beams may be configured by an NZP-CSI-RS-ResourceSet field, and the Set B beams may be configured by a CSI-SSB- ResourceSet field or an NZP-CSI-RS-ResourceSet field. In some cases, an association ID that links the Set A beams and the Set B beams may be included in the CSI-measConfig IE.

[0031] FIG. 1 illustrates a flow diagram of a UE side model training stage 102 and a flow diagram of a UE side model inferencing stage 104, according to embodiments herein. The flow diagram of a UE side model training stage 102 may begin with the UE 108 optionally transmitting 116 a request for data collection for beam management to the network 110. The network 110 may send 118 a CSI-MeasConfig IE for data collection to the UE 108 including a Set A beams resourceSet, a Set B beams resourceSet, and association ID. In some cases, the UE 108 may store the Set A beams and Set B beams resourceSet and association ID in memory for use in training and / or inference of the AI / ML model (optionally performed offline). It should be understood that there may be no associated reporting performed in the training stage. Accordingly, the network 110 may send 120 SSB and / or CSI-RS transmissions to the UE 108. Then, the UE 108 may buffer 122 the SSB / CSI-RS measurement. Optionally, the UE 108 may send 124 a stop request to the network 110 and the network 110 may respond 126 by stopping data collection. The UE 108 may forward 128 the data collected into the data collection to the server 106. This may be 3GPP transparent (i.e., done using Bluetooth or WIFI). The server 106 may perform 130 AI / ML model design, AI / ML model training, and / or AI / ML model testing and may forward 132 the AI / ML model to the UE 108.

[0032] Accordingly, in some embodiments, for UE side AI / ML model inferencing, the UE may perform AI / ML model inferencing based on the Set B beam measurement(s). In the measurement resource configuration (e.g., CSLResourceConfig for Set B beams), various additional information may be included. For example, a CSI-SSB-ResourceSet field or an NZP-CSI-RS-ResourceSet field, and an association ID (to associate the Set B beams back to the Set A beams) may be included in the measurement resource configuration. In some cases, a quasi co-location (QCL) type D relationship may be6P68965WO1 4924-2349-0130Uassumed by the UE between the set-B beams configuration in inferencing and the Set B beams configuration in the training dataset collection procedure / configuration with the same association ID. In some such cases, for the ResourceSet field configuration, each resource of the Set B beams is QCL-type D with the resource(s) of the Set B beams during data collection.

[0033] For example, the flow diagram of a UE side model inferencing stage 104 illustrated in FIG. 1 may begin with the network 114 transmitting 134, to the UE 112, a CSI-ReportConfig IE configuring a CSI report format for beam management inferencing and a CSI-ResourceConfig IE for the configuration of the Set B beams. The network 114 may transmit 136 the Set B beam reference signals to the UE 112. The UE 112 may perform 138 AI / ML model inferencing based on, for example, the Set B beam reference signals. The UE 112 may report 140, to the network 114, a CSI report including the predicted set of beams where, in some cases, the beam index of the beams may map 142 to the Set A beams resource index during training with the same associated ID as, in some examples, the CSI-ReportConfig IE and the CSI-ResourceConfig IE may not include information on the Set A beams.

[0034] In some cases, a large amount of time may pass between performance of the inference stage and the training stage. As a result, according to embodiments herein, it is beneficial for the beam index of the beams to map 142 to the Set A beams resource index during training with the same associated ID as to maintain consistency between the training stage and the inference stage.

[0035] In some embodiments, for UE side AI / ML model inferencing, the UE may perform AI / ML model inferencing based on the Set B beam measurements. For example, for CSI reporting, the N beams reported in the CSI report may refer to a resource index configured during data collection for the Set A beams with the same association ID. In some cases, the CSI report may include the association ID, a resource indicator (e.g., resource indicator #z), and / or optionally the reference signal received power (RSRP) value corresponding to the resource indicator.

[0036] The resource indicator #i (e.g., #z=l ,2,...,N) is the indicator of the reported resource. The bitwidth of the resource indicator is [log_2 (KRS)], where KRS is the number of reference signal ports for the Set A beams in data collection configuration. Note that due to offline training the data collection procedures may happen a certain amount of time (e.g., days and / or weeks) ahead of the performed inference. In some7P68965WO1 4924-2349-0130Uexamples, the resource indicator may depend on the Set A beams configuration as the resource ID may correspond to NZP-CSI-RS-Resourceld field in the data collection configuration.

[0037] In some embodiments, the resource indicator(s) in the CSI report may be ordered based on AI / ML model inference results starting from best candidate beam in the CSI report.

[0038] FIG. 2 illustrates an example of a CSI report 200 reported after UE side model inferencing, according to embodiments herein. The CSI report 200 may include resource indicators (i.e., first resource indicator 202 and second resource indicator 204 up to an Nth resource indicator 206) and optionally, an RSRP value for each resource indicator (i.e., a first RSRP value 208 corresponding to the first resource indicator 202, a second differential RSRP value 210 corresponding to the second resource indicator 204 up to an Nth differential RSRP value 212 corresponding to the Nth resource indicator 206). Note that a differential RSRP is reported for the second resource indicator 204 up to the Nth resource indicator 206 as differential encoding (e.g., reporting the differential of the RSRP value of the first resource indicator to the RSRP value of the second resource indicator) uses less headroom compared to reporting the entire RSRP value.

[0039] In some embodiments, for UE side AI / ML based beam model inferencing, the UE may perform AI / ML model inferencing based on Set B beam measurements. The measurement resource configuration for the measurements may include, for example, a CSI-SSB-ResourceSet field or an NZP-CSI-RS-ResourceSet field, and an association ID.

[0040] FIG. 3 illustrates a flow diagram of a UE side model monitoring stage 302 and a flow diagram of a UE side model inferencing stage 304, according to embodiments herein. The flow diagram of a UE side model monitoring stage 302 may begin with the network 308 transmitting 314, to the UE 306, a CSI-ReportConfig IE for beam management performance monitoring including a CSI-ResourceConfig IE for configuring the Set A beams and the Set B beams. Then, the network 308 may transmit 316, to the UE 306, Set A beam reference signals. Optionally, the network 308 may transmit 318, to the UE 306, the Set B beam reference signals as, in some cases, the transmitted Set A beam reference signals may be reused as the Set B beam reference signals if the Set A beams includes the resources of the Set B beams.

[0041] In some cases, the set B beam resource can reuse the inference set B RS configuration and measurement. An implicit link between the set A RS in performance8P68965WO1 4924-2349-0130Umonitoring and the set B RS in inference can be exploited. For example, the set B transmission that is the closest in time to set A RS transmission can be used for performance monitoring. In some cases, when a set B beam is a subset of set A beam, the RS configuration for set A can include the set A beams that are different than the set B beam.

[0042] The UE 306 may measure 320 the Set A beam reference signals depending on the network side performance monitoring and / or depending on the UE assisted monitoring. In some examples, the UE 306 may generate a CSI report and may report 322 the CSI report to the network 308. In some examples, the UE 306 may generate a variety of performance monitoring metrics for the model inferencing and may report 322 the performance monitoring metrics to the UE 306.

[0043] In some embodiments, for UE side model performance monitoring, in the measurement resource configuration for Set A beams for performance monitoring, a CSI- SSB -Re source Set field or NZP-CSI-RS-ResourceSet field, the association ID, and / or a Set A beams resource set may be included. In some cases, the Set A beams resource set may have a different number of ports and a different periodicity compared to the Set B beams resource set used for inference. In some embodiments a Set B beams resource may be configured for performance monitoring in the same CSI-ResourceConfig IE (e.g., the inference resource configuration may not be reused).

[0044] For example, the flow diagram of a UE side model inferencing stage 304 illustrated in FIG. 3 may begin with the network 312 transmitting 324 a CSI- ReportConfig IE for beam management inferencing including a CSI-ResourceConfig IE for the Set B beams, to the UE 310. Then, the network 312 may transmit 326 the Set B beam reference signals to the UE 310. Accordingly, the UE 310 may perform 328 AI / ML model inferencing based on the Set B beam reference signals. Then, the UE 310 may report 330 a CSI report including a set of predicted beams where the beam index of the predicted beams may map 332 to the Set A beams resource index configured for performance monitoring.

[0045] FIG. 4 illustrates a diagram 400 of periodic and / or semi-persistent measurement and CSI reporting, according to embodiments herein. As shown, Set B beam reference signals 402 may be transmitted by a network node periodically for the UE to use in the AI / ML inference of Set A beams. The UE may generate a report 404 based on9P68965WO1 4924-2349-0130Umeasurements of the Set B beam reference signals 402 and on inferencing the Set A beams. The UE may send the report 404 to the network.

[0046] The network may also periodically transmit Set A beam reference signals 406 for performance monitoring. As shown, the periodicity for Set A beam reference signals 406 may be longer than the periodicity for Set B beam reference signals 402. Further, the UE report 404 of network beams with RS resource ID configured for monitoring may include the same association ID as the Set A beam reference signals 406.

[0047] Additionally, in some embodiments, a Set A beam of the Set A beam reference signals 406 may overlap with one of the Set B beams of the Set B beam reference signals 402 used for performance monitoring. As a result, the Set B beam may be reused 412 as a Set A beam and a new measurement is not necessary to perform.

[0048] The UE may transmit a report for performance monitoring to the network node based on the set A RS transmission. Note that the UE reports 408 for performance monitoring may be transmitted less often as compared to the UE report 404 with inferencing, as beams may be predicted based on the Set B beam reference signals 402 which are transmitted more often as compared to the Set A beam reference signals 406.

[0049] In some embodiments, for UE side AI / ML model performance monitoring, when a Set B beams resourceSet is used to perform inference and a separate resource for Set A beams are configured for performance monitoring, an implicit association may be used for the association of the Set B beams resource set and the corresponding Set A beams resource set. For example, when different CSI measurement resource sets are tagged with the same association ID, the UE may assume that the closest Set B beam resource set transmission may be used for performance monitoring.

[0050] FIG. 5 illustrates an example Set B beam and Set A beam reference signal transmission timing, according to embodiments herein. In some cases, the Set B beams reference signal may be transmitted a certain amount of time ahead of the Set A beams reference signal measurement (e.g., at least a processing delay 502 Tproc ahead of the Set A beam). In some other cases, the Set B beam reference signal may be transmitted in the same timing window 504 as the timing window 504 of when the Set A beam reference signal is transmitted for performance monitoring (i.e., the Set B beams reference signal and the Set A beams reference signal are transmitted in the same timing window 504).

[0051] In some embodiments, for UE side model inferencing, the UE may perform AI / ML model inference based on the Set B beams measurement. For example, as10P68965WO1 4924-2349-0130Uillustrated in FIG. 2 for CSI reporting, the N beams reported in the CSI report may refer to a resource index configured for Set A beams resource set performance monitoring with the same association ID. The report may include, for example, the association ID and / or a resource indicator #i. The resource indicator may take various values (e.g., z=l,2,...,N) and may be understood as the indicator of the reported resource. The bitwidth of the resource indicator may be [log_2 (KRS)], where KRS is the number of reference signal ports for the Set A beams used for performance monitoring. Additionally, the resource indicator may be ordered based on the AI / ML model inferencing results starting from best candidate beam. Optionally, the report may further include an RSRP value of the corresponding resource indicator. Note that a differential RSRP is reported for the second resource indicator (i.e., resource indicator #2) up to the Nth resource indicator (i.e., resource indicator #N) using differential encoding (e.g., reporting the differential of the RSRP value of the first resource indicator to the RSRP value of the second resource indicator). The use of differential encoding reporting may use less headroom compared to reporting the entire RSRP value. In some cases, the resource indicator(s) in the CSI report may be ordered based on AI / ML model inference results starting from best candidate beam, in the CSI report.

[0052] In some embodiments, a combination of the embodiments discussed herein may be utilized. In some cases, for a UE side model in Al based beam management, for data collection, inferencing, and performance monitoring, the UE may assume a QCL type D relationship for Set B beams configuration during the training stage, the inference stage and performance monitoring stage. In some examples, a same resourceSet may be configured for training, inferencing, and performance monitoring. In some other examples, different resourceSets may be configured for training, inferencing, and performance monitoring and the UE may assume that each resource ID has a QCL type D relationship.

[0053] In some other cases, the UE may assume a QCL type D relationship for the Set A beams configuration during training data collection procedure and performance monitoring. In some examples, a same resourceSet may be configured for training and performance monitoring. In some other examples, different resourceSets may be configured for training, inferencing, and performance monitoring and the UE may assume each resource has a QCL type D relationship.11P68965WO1 4924-2349-0130U

[0054] The UE may report the predicted network beam(s) based on the resource ID of the Set A beams in either the AI / ML model training stage or the AI / ML model performance monitoring stage.

[0055] Embodiments herein may extend and apply to BM-Case2, where BM-Case2 may be understood as a time domain prediction whereas BM-Casel may be understood as a spatial domain prediction. As a result, in some embodiments, in BM-Case 2 sub-case 1, the Set A beams and the Set B beams are a part of the same resourceSet (e.g., UE performing time domain prediction of a future measurement). Additionally, in BM-Case 2 sub-case 2, the Set A beams may be a bigger set of beams compared to the set of Set B beams. Note that this may be the same for BM-Casel.

[0056] In some embodiments, for BM-Case 2 sub-case 1, the Set A beam configuration may reuse the same resource ID as the Set B beam measurement resource configuration. Embodiments previously discussed herein additionally apply to BM-Case2 sub-case 2, where a larger beam set is predicted (e.g., embodiments discussed with reference to FIGS. 1-5). In some cases, the report bits may scale with a pre-configured prediction instance by the network.

[0057] Additionally, in some embodiments for BM-Case2, for UE side model inferencing, the UE may perform AI / ML model inference based on the Set B beams measurements, similar to that in BM-Casel. For example, for CSI reporting, the N beams reported may refer to a resource index configured during data collection for the Set A beams with the same association ID for each prediction instance. In some cases, the report content may include the association ID, a resource indicator #z, and / or optionally the RSRP value corresponding to the resource indictor.

[0058] The resource indicator #i may various values (e.g., z=l ,2,...,N) and is the indicator of the reported resource for each prediction instance. The bitwidth of the resource indicator is [log_2 (KRS)], where KRS is the number of reference signal ports for the Set A beams in data collection configuration. Note that due to offline training, the data collection procedures can happen days / weeks ahead of actual inference. In some examples, the resource indicator may depend on the Set A beams configuration as the resource ID may correspond to the NZP-CSI-RS-Resourceld field in the data collection configuration per prediction instance.

[0059] FIG. 6 illustrates an example of a CSI report 600 reported after UE side model inferencing, according to embodiments herein. The CSI report 600 may include resource12P68965WO1 4924-2349-0130Uindi ctors and optionally, a RSRP value for each resource indicator for each prediction instance. For example, for a first prediction instance, the CSI report 600 may include a first resource indicator 602, a second resource indicator 604 up to an Nth resource indicator 606 and optionally a first RSRP value 608 corresponding to the first resource indicator 602, a second differential RSRP value 610 corresponding to the second resource indicator 604 and an Nth differential RSRP value 612 corresponding to the Nth resource indicator 606. For a second prediction instance, the CSI report 600 may include a first resource indicator 614, a second resource indicator 616, up to an Nth resource indicator 618 and optionally a first RSRP value 620 corresponding to the first resource indicator 614, a second differential RSRP value 622 corresponding to the second resource indicator 616 and an Nth differential RSRP value 624 corresponding to the Nth resource indicator 618.

[0060] Note that a differential RSRP is reported for the second RSRP value, the third RSRP value, up to the Nth RSRP value as differential encoding (e.g., reporting the differential of the RSRP value of the first resource indicator to the RSRP value of the second resource indicator) may use less headroom compared to reporting the entire RSRP value. In some cases, the resource indicator(s) in the CSI report 600 may be ordered based on AI / ML model inference results starting from best candidate beam, in the CSI report. Alternatively, the differential RSRP can be reported based on the first RSRP value of the prediction instance 1.

[0061] FIG. 7 illustrates a diagram 700 of periodic and semi-persistent measurement and reporting for BM-Case2, according to embodiments herein. The network may periodically transmit set B reference signals 708. The UE may measure the set B reference signals 708 and use the measurements for inferencing. The UE may use inferencing to send a UE report 702 more frequently than the set B reference signals 708. This may save resources in the time domain.

[0062] The network may also periodically send Set A beam reference signals 704 for performance monitoring. The UE may send a report 710 for performance monitoring of the model to the network based on Set A beam reference signals 704 and the inferences. The UE report of network beams with RS resource ID configured for monitoring and the Set A beam reference signals 704 may have the same association ID.

[0063] FIG. 8 illustrates a method 800 performed by a UE, according to embodiments herein. The illustrated method 800 includes receiving 802 a first resource configuration13P68965WO1 4924-2349-0130Uincluding a configuration for a first set of beams and an association ID. The method 800 further includes performing 804 training or performance monitoring for a model based on measurements of a first set of reference signals corresponding to the first set of beams. The method 800 further includes receiving 806 a second resource configuration including a configuration for a second set of beams and the association ID. The method 800 further includes identifying 808 that the first set of beams are linked to the second set of beams based on the association ID being included in both the first resource configuration and the second resource configuration. The method 800 further includes inferencing 810 predicted measurements for the first set of beams using the model based on measurements of a second set of reference signals corresponding to the second set of beams.

[0064] In some embodiments, the method 800 further comprises generating a CSI report based on the predicted measurements for the first set of beams, wherein index values of beams in the CSI report correlate to resource index values configured during data collection for the first set of beams with the association ID. In some such embodiments, the CSI includes a resource indicator ordered based on inference results starting from best candidate beam.

[0065] In some embodiments of the method 800, the UE performs training for the model, wherein the first resource configuration is a CSI-MeasConfig information element and further comprises the second set of beams, wherein the first set of beams is configured by NZP-CSI-RS-ResourceSet, and wherein the second set of beams is configured by CSI-SSB-ResourceSet or NZP-CSI-RS-ResourceSet.

[0066] In some embodiments of the method 800, the second resource configuration is a CSI-ReportConfig information element and further includes CSI-SSB-ResourceSet or NZP-CSI-RS-ResourceSet to configure the second set of beams.

[0067] In some embodiments, the method 800 further comprises in response to the second set of beams being used for inference and the first set of beams being configured for performance monitoring, determining that transmissions of the first set of beams and the second set of beams that are the closest in time are related.

[0068] In some embodiments, the method 800 further comprises generating a CSI report based on the predicted measurements for the first set of beams, wherein index values of beams in the CSI report correlate to resource index values configured for performance monitoring of the first set of beams with the same association ID.14P68965WO1 4924-2349-0130U

[0069] In some embodiments of the method 800, there is a QCL type D relationship for the second resource configuration during training procedure, inference and performance monitoring.

[0070] In some embodiments of the method 800, there is a QCL type D relationship for the first resource configuration during training procedure and performance monitoring.

[0071] FIG. 9 illustrates a method 900 performed by a network node, according to embodiments herein. The illustrated method 900 includes receiving 902, from a UE, a first resource configuration including a configuration for a first set of beams and an association ID. The method 900 further includes sending 904, to the UE, the first set of beams for training or performance monitoring for a model. The method 900 further includes sending 906, to the UE, a second resource configuration including a configuration for a second set of beams and the association ID, wherein the first set of beams are linked to the second set of beams based on the association ID being included in both the first resource configuration and the second resource configuration. The method 900 further includes sending 908 a second set of reference signals corresponding to the second set of beams. The method 900 further includes receiving 910 a CSI based on inferences using the model and measurements of the second set of reference signals.

[0072] In some embodiments of the method 900, index values of beams in the CSI report correlate to resource index values configured during data collection for the first set of beams with the association ID. In some such embodiments, the CSI includes a resource indicator ordered based on inference results starting from best candidate beam.

[0073] In some embodiments of the method 900, the first resource configuration is a CSLMeasConfig information element and further comprises the second set of beams, wherein the first set of beams is configured by NZP-CSLRS-ResourceSet, and wherein the second set of beams is configured by CSLSSB-ResourceSet or NZP-CSLRS- ResourceSet.

[0074] In some embodiments of the method 900, the second resource configuration is a CSI-ReportConfig information element and further includes CSLSSB-ResourceSet or NZP-CSLRS-ResourceSet to configure the second set of beams.

[0075] In some embodiments of the method 900, index values of beams in the CSI report correlate to resource index values configured for performance monitoring of the first set of beams with the same association ID.15P68965WO1 4924-2349-0130U

[0076] In some embodiments of the method 900, there is a QCL type D relationship for the second resource configuration during training procedure, inference and performance monitoring.

[0077] In some embodiments of the method 900, there is a QCL type D relationship for the first resource configuration during training procedure and performance monitoring.

[0078] FIG. 10 illustrates an example architecture of a wireless communication system 1000, according to embodiments disclosed herein. The following description is provided for an example wireless communication system 1000 that operates in conjunction with the LTE system standards and / or 5G or NR system standards as provided by 3 GPP technical specifications.

[0079] As shown by FIG. 10, the wireless communication system 1000 includes UE 1002 and UE 1004 (although any number of UEs may be used). In this example, the UE 1002 and the UE 1004 are illustrated as smartphones (e.g., handheld touchscreen mobile computing devices connectable to one or more cellular networks), but may also comprise any mobile or non-mobile computing device configured for wireless communication.

[0080] The UE 1002 and UE 1004 may be configured to communicatively couple with a RAN 1006. In embodiments, the RAN 1006 may be NG-RAN, E-UTRAN, etc. The UE 1002 and UE 1004 utilize connections (or channels) (shown as connection 1008 and connection 1010, respectively) with the RAN 1006, each of which comprises a physical communications interface. The RAN 1006 can include one or more base stations (such as base station 1012 and base station 1014) that enable the connection 1008 and connection 1010.

[0081] In this example, the connection 1008 and connection 1010 are air interfaces to enable such communicative coupling, and may be consistent with RAT(s) used by the RAN 1006, such as, for example, an LTE and / or NR.

[0082] In some embodiments, the UE 1002 and UE 1004 may also directly exchange communication data via a sidelink interface 1016. The UE 1004 is shown to be configured to access an access point (shown as AP 1018) via connection 1020. By way of example, the connection 1020 can comprise a local wireless connection, such as a connection consistent with any IEEE 802.11 protocol, wherein the AP 1018 may comprise a Wi-Fi® router. In this example, the AP 1018 may be connected to another network (for example, the Internet) without going through a CN 1024.16P68965WO1 4924-2349-0130U

[0083] In embodiments, the UE 1002 and UE 1004 can be configured to communicate using orthogonal frequency division multiplexing (OFDM) communication signals with each other or with the base station 1012 and / or the base station 1014 over a multicarrier communication channel in accordance with various communication techniques, such as, but not limited to, an orthogonal frequency division multiple access (OFDMA) communication technique (e.g., for downlink communications) or a single carrier frequency division multiple access (SC-FDMA) communication technique (e.g., for uplink and ProSe or sidelink communications), although the scope of the embodiments is not limited in this respect. The OFDM signals can comprise a plurality of orthogonal subcarriers.

[0084] In some embodiments, all or parts of the base station 1012 or base station 1014 may be implemented as one or more software entities running on server computers as part of a virtual network. In addition, or in other embodiments, the base station 1012 or base station 1014 may be configured to communicate with one another via interface 1022. In embodiments where the wireless communication system 1000 is an LTE system (e.g., when the CN 1024 is an EPC), the interface 1022 may be an X2 interface. The X2 interface may be defined between two or more base stations (e.g., two or more eNBs and the like) that connect to an EPC, and / or between two eNBs connecting to the EPC. In embodiments where the wireless communication system 1000 is an NR system (e.g., when CN 1024 is a 5GC), the interface 1022 may be an Xn interface. The Xn interface is defined between two or more base stations (e.g., two or more gNBs and the like) that connect to 5GC, between a base station 1012 (e.g., a gNB) connecting to 5GC and an eNB, and / or between two eNBs connecting to 5GC (e.g., CN 1024).

[0085] The RAN 1006 is shown to be communicatively coupled to the CN 1024. The CN 1024 may comprise one or more network elements 1026, which are configured to offer various data and telecommunications services to customers / subscribers (e.g., users of UE 1002 and UE 1004) who are connected to the CN 1024 via the RAN 1006. The components of the CN 1024 may be implemented in one physical device or separate physical devices including components to read and execute instructions from a machine- readable or computer-readable medium (e.g., a non-transitory machine-readable storage medium).

[0086] In embodiments, the CN 1024 may be an EPC, and the RAN 1006 may be connected with the CN 1024 via an SI interface 1028. In embodiments, the SI interface17P68965WO1 4924-2349-0130U1028 may be split into two parts, an SI user plane (Sl-U) interface, which carries traffic data between the base station 1012 or base station 1014 and a serving gateway (S-GW), and the SI -MME interface, which is a signaling interface between the base station 1012 or base station 1014 and mobility management entities (MMEs).

[0087] In embodiments, the CN 1024 may be a 5GC, and the RAN 1006 may be connected with the CN 1024 via an NG interface 1028. In embodiments, the NG interface 1028 may be split into two parts, an NG user plane (NG-U) interface, which carries traffic data between the base station 1012 or base station 1014 and a user plane function (UPF), and the SI control plane (NG-C) interface, which is a signaling interface between the base station 1012 or base station 1014 and access and mobility management functions (AMFs).

[0088] Generally, an application server 1030 may be an element offering applications that use internet protocol (IP) bearer resources with the CN 1024 (e.g., packet switched data services). The application server 1030 can also be configured to support one or more communication services (e.g., VoIP sessions, group communication sessions, etc.) for the UE 1002 and UE 1004 via the CN 1024. The application server 1030 may communicate with the CN 1024 through an IP communications interface 1032.

[0089] FIG. 11 illustrates a system 1100 for performing signaling 1134 between a wireless device 1102 and a network device 1118, according to embodiments disclosed herein. The system 1100 may be a portion of a wireless communications system as herein described. The wireless device 1102 may be, for example, a UE of a wireless communication system. The network device 1118 may be, for example, a base station (e.g., an eNB or a gNB) of a wireless communication system.

[0090] The wireless device 1102 may include one or more processor(s) 1104. The processor(s) 1104 may execute instructions such that various operations of the wireless device 1102 are performed, as described herein. The processor(s) 1104 may include one or more baseband processors implemented using, for example, a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.

[0091] The wireless device 1102 may include a memory 1106. The memory 1106 may be a non-transitory computer-readable storage medium that stores instructions 110818P68965WO1 4924-2349-0130U(which may include, for example, the instructions being executed by the processor(s) 1104). The instructions 1108 may also be referred to as program code or a computer program. The memory 1106 may also store data used by, and results computed by, the processor(s) 1104.

[0092] The wireless device 1102 may include one or more transceiver(s) 1110 that may include radio frequency (RF) transmitter circuitry and / or receiver circuitry that use the antenna(s) 1112 of the wireless device 1102 to facilitate signaling (e.g., the signaling 1134) to and / or from the wireless device 1102 with other devices (e.g., the network device 1118) according to corresponding RATs.

[0093] The wireless device 1102 may include one or more antenna(s) 1112 (e.g., one, two, four, or more). For embodiments with multiple antenna(s) 1112, the wireless device 1102 may leverage the spatial diversity of such multiple antenna(s) 1112 to send and / or receive multiple different data streams on the same time and frequency resources. This behavior may be referred to as, for example, multiple input multiple output (MIMO) behavior (referring to the multiple antennas used at each of a transmitting device and a receiving device that enable this aspect). MIMO transmissions by the wireless device 1102 may be accomplished according to precoding (or digital beamforming) that is applied at the wireless device 1102 that multiplexes the data streams across the antenna(s) 1112 according to known or assumed channel characteristics such that each data stream is received with an appropriate signal strength relative to other streams and at a desired location in the spatial domain (e.g., the location of a receiver associated with that data stream). Certain embodiments may use single user MIMO (SU-MIMO) methods (where the data streams are all directed to a single receiver) and / or multi user MIMO (MU-MIMO) methods (where individual data streams may be directed to individual (different) receivers in different locations in the spatial domain).

[0094] In certain embodiments having multiple antennas, the wireless device 1102 may implement analog beamforming techniques, whereby phases of the signals sent by the antenna(s) 1112 are relatively adjusted such that the (joint) transmission of the antenna(s) 1112 can be directed (this is sometimes referred to as beam steering).

[0095] The wireless device 1102 may include one or more interface(s) 1114. The interface(s) 1114 may be used to provide input to or output from the wireless device 1102. For example, a wireless device 1102 that is a UE may include interface(s) 1114 such as microphones, speakers, a touchscreen, buttons, and the like in order to allow for19P68965WO1 4924-2349-0130Uinput and / or output to the UE by a user of the UE. Other interfaces of such a UE may be made up of transmitters, receivers, and other circuitry (e.g., other than the transceiver(s) 1110 / antenna(s) 1112 already described) that allow for communication between the UE and other devices and may operate according to known protocols (e.g., Wi-Fi®, Bluetooth®, and the like).

[0096] The wireless device 1102 may include an AI / ML based beam management module 1116. The AI / ML based beam management module 1116 may be implemented via hardware, software, or combinations thereof. For example, the AI / ML based beam management module 1116 may be implemented as a processor, circuit, and / or instructions 1108 stored in the memory 1106 and executed by the processor(s) 1104. In some examples, the AI / ML based beam management module 1116 may be integrated within the processor(s) 1104 and / or the transceiver(s) 1110. For example, the AI / ML based beam management module 1116 may be implemented by a combination of software components (e.g., executed by a DSP or a general processor) and hardware components (e.g., logic gates and circuitry) within the processor(s) 1104 or the transceiver(s) 1110.

[0097] The AI / ML based beam management module 1116 may be used for various aspects of the present disclosure, for example, aspects of FIGS. 1-8.

[0098] The network device 1118 may include one or more processor(s) 1120. The processor(s) 1120 may execute instructions such that various operations of the network device 1118 are performed, as described herein. The processor(s) 1120 may include one or more baseband processors implemented using, for example, a CPU, a DSP, an ASIC, a controller, an FPGA device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.

[0099] The network device 1118 may include a memory 1122. The memory 1122 may be a non-transitory computer-readable storage medium that stores instructions 1124 (which may include, for example, the instructions being executed by the processor(s) 1120). The instructions 1124 may also be referred to as program code or a computer program. The memory 1122 may also store data used by, and results computed by, the processor(s) 1120.

[0100] The network device 1118 may include one or more transceiver(s) 1126 that may include RF transmitter circuitry and / or receiver circuitry that use the antenna(s) 1128 of the network device 1118 to facilitate signaling (e.g., the signaling 1134) to and / or from20P68965WO1 4924-2349-0130Uthe network device 1118 with other devices (e.g., the wireless device 1102) according to corresponding RATs.

[0101] The network device 1118 may include one or more antenna(s) 1128 (e.g., one, two, four, or more). In embodiments having multiple antenna(s) 1128, the network device 1118 may perform MIMO, digital beamforming, analog beamforming, beam steering, etc., as has been described.

[0102] The network device 1118 may include one or more interface(s) 1130. The interface(s) 1130 may be used to provide input to or output from the network device 1118. For example, a network device 1118 that is a base station may include interface(s) 1130 made up of transmitters, receivers, and other circuitry (e.g., other than the transceiver(s) 1126 / antenna(s) 1128 already described) that enables the base station to communicate with other equipment in a core network, and / or that enables the base station to communicate with external networks, computers, databases, and the like for purposes of operations, administration, and maintenance of the base station or other equipment operably connected thereto.

[0103] The network device 1118 may include an AI / ML based beam management module 1132. The AI / ML based beam management module 1132 may be implemented via hardware, software, or combinations thereof. For example, the AI / ML based beam management module 1132 may be implemented as a processor, circuit, and / or instructions 1124 stored in the memory 1122 and executed by the processor(s) 1120. In some examples, the AI / ML based beam management module 1132 may be integrated within the processor(s) 1120 and / or the transceiver(s) 1126. For example, the AI / ML based beam management module 1132 may be implemented by a combination of software components (e.g., executed by a DSP or a general processor) and hardware components (e.g., logic gates and circuitry) within the processor(s) 1120 or the transceiver(s) 1126.

[0104] The AI / ML based beam management module 1132 may be used for various aspects of the present disclosure, for example, aspects of FIGS.1-8.

[0105] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein).

[0106] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon21P68965WO1 4924-2349-0130Uexecution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 800. This non-transitory computer-readable media may be, for example, a memory of a UE (such as a memory 1106 of a wireless device 1102 that is a UE, as described herein).

[0107] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein).

[0108] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein).

[0109] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 800.

[0110] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processor is to cause the processor to carry out one or more elements of the method 800. The processor may be a processor of a UE (such as a processor(s) 1104 of a wireless device 1102 that is a UE, as described herein). These instructions may be, for example, located in the processor and / or on a memory of the UE (such as a memory 1106 of a wireless device 1102 that is a UE, as described herein).[OHl] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 900. This apparatus may be, for example, an apparatus of a base station (such as a network device 1118 that is a base station, as described herein).

[0112] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 900. This non-transitory computer-readable media may be, for example, a memory of a base station (such as a memory 1122 of a network device 1118 that is a base station, as described herein).22P68965WO1 4924-2349-0130U

[0113] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 900. This apparatus may be, for example, an apparatus of a base station (such as a network device 1118 that is a base station, as described herein).

[0114] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 900. This apparatus may be, for example, an apparatus of a base station (such as a network device 1118 that is a base station, as described herein).

[0115] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 900.

[0116] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processing element is to cause the processing element to carry out one or more elements of the method 900. The processor may be a processor of a base station (such as a processor(s) 1120 of a network device 1118 that is a base station, as described herein). These instructions may be, for example, located in the processor and / or on a memory of the base station (such as a memory 1122 of a network device 1118 that is a base station, as described herein).

[0117] For one or more embodiments, at least one of the components set forth in one or more of the preceding figures may be configured to perform one or more operations, techniques, processes, and / or methods as set forth herein. For example, a baseband processor as described herein in connection with one or more of the preceding figures may be configured to operate in accordance with one or more of the examples set forth herein. For another example, circuitry associated with a UE, base station, network element, etc. as described above in connection with one or more of the preceding figures may be configured to operate in accordance with one or more of the examples set forth herein.

[0118] Any of the above described embodiments may be combined with any other embodiment (or combination of embodiments), unless explicitly stated otherwise. The foregoing description of one or more implementations provides illustration and description, but is not intended to be exhaustive or to limit the scope of embodiments to23P68965WO1 4924-2349-0130Uthe precise form disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from practice of various embodiments.

[0119] Embodiments and implementations of the systems and methods described herein may include various operations, which may be embodied in machine-executable instructions to be executed by a computer system. A computer system may include one or more general-purpose or special-purpose computers (or other electronic devices). The computer system may include hardware components that include specific logic for performing the operations or may include a combination of hardware, software, and / or firmware.

[0120] It should be recognized that the systems described herein include descriptions of specific embodiments. These embodiments can be combined into single systems, partially combined into other systems, split into multiple systems or divided or combined in other ways. In addition, it is contemplated that parameters, attributes, aspects, etc. of one embodiment can be used in another embodiment. The parameters, attributes, aspects, etc. are merely described in one or more embodiments for clarity, and it is recognized that the parameters, attributes, aspects, etc. can be combined with or substituted for parameters, attributes, aspects, etc. of another embodiment unless specifically disclaimed herein.

[0121] It is well understood that the use of personally identifiable information should follow privacy policies and practices that are generally recognized as meeting or exceeding industry or governmental requirements for maintaining the privacy of users. In particular, personally identifiable information data should be managed and handled so as to minimize risks of unintentional or unauthorized access or use, and the nature of authorized use should be clearly indicated to users.

[0122] Although the foregoing has been described in some detail for purposes of clarity, it will be apparent that certain changes and modifications may be made without departing from the principles thereof. It should be noted that there are many alternative ways of implementing both the processes and apparatuses described herein. Accordingly, the present embodiments are to be considered illustrative and not restrictive, and the description is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.24P68965WO1 4924-2349-0130U

Claims

CLAIMS1. A method performed by a User Equipment (UE), the method comprising: receiving a first resource configuration including a configuration for a first set of beams and an association identifier (ID); performing training or performance monitoring for a model based on measurements of a first set of reference signals corresponding to the first set of beams; receiving a second resource configuration including a configuration for a second set of beams and the association ID; identifying that the first set of beams are linked to the second set of beams based on the association ID being included in both the first resource configuration and the second resource configuration; and inferencing predicted measurements for the first set of beams using the model based on measurements of a second set of reference signals corresponding to the second set of beams.

2. The method of claim 1, further comprising generating a channel state information (CSI) report based on the predicted measurements for the first set of beams, wherein index values of beams in the CSI report correlate to resource index values configured during data collection for the first set of beams with the association ID.

3. The method of claim 2, wherein the CSI includes a resource indicator ordered based on inference results starting from best candidate beam.

4. The method of claim 1, wherein the UE performs training for the model, wherein the first resource configuration is a CSI-MeasConfig information element and further comprises the second set of beams, wherein the first set of beams is configured by NZP-CSI-RS-ResourceSet, and wherein the second set of beams is configured by CSI-SSB-ResourceSet or NZP-CSI-RS-ResourceSet.

5. The method of claim 1, wherein the second resource configuration is a CSI- ReportConfig information element and further includes CSI-SSB-ResourceSet or NZP- CSI-RS-ResourceSet to configure the second set of beams.

6. The method of claim 1, further comprising:25P68965WO1 4924-2349-0130Uin response to the second set of beams being used for inference and the first set of beams being configured for performance monitoring, determining that transmissions of the first set of beams and the second set of beams that are the closest in time are related.

7. The method of claim 1, further comprising generating a channel state information (CSI) report based on the predicted measurements for the first set of beams, wherein index values of beams in the CSI report correlate to resource index values configured for performance monitoring of the first set of beams with the same association ID.

8. The method of claim 1, wherein there is a quasi co-location (QCL) type D relationship for the second resource configuration during training procedure, inference and performance monitoring.

9. The method of claim 1, wherein there is a quasi co-location (QCL) type D relationship for the first resource configuration during training procedure and performance monitoring.

10. A method performed by a network node, the method comprising: receiving, from a User Equipment (UE), a first resource configuration including a configuration for a first set of beams and an association identifier (ID); sending, to the UE, the first set of beams for training or performance monitoring for a model; sending, to the UE, a second resource configuration including a configuration for a second set of beams and the association ID, wherein the first set of beams are linked to the second set of beams based on the association ID being included in both the first resource configuration and the second resource configuration; sending a second set of reference signals corresponding to the second set of beams; and receiving a channel state information (CSI) based on inferences using the model and measurements of the second set of reference signals.

11. The method of claim 10, wherein index values of beams in the CSI report correlate to resource index values configured during data collection for the first set of beams with the association ID.26P68965WO1 4924-2349-0130U12. The method of claim 11, wherein the CSI includes a resource indicator ordered based on inference results starting from best candidate beam.

13. The method of claim 10, wherein the first resource configuration is a CSI- MeasConfig information element and further comprises the second set of beams, wherein the first set of beams is configured by NZP-CSI-RS-ResourceSet, and wherein the second set of beams is configured by CSI-SSB-ResourceSet or NZP- CSI-RS-ResourceSet.

14. The method of claim 10, wherein the second resource configuration is a CSI- ReportConfig information element and further includes CSI-SSB-ResourceSet or NZP- CSI-RS-ResourceSet to configure the second set of beams.

15. The method of claim 10, wherein index values of beams in the CSI report correlate to resource index values configured for performance monitoring of the first set of beams with the same association ID.

16. The method of claim 10, wherein there is a quasi co-location (QCL) type D relationship for the second resource configuration during training procedure, inference and performance monitoring.

17. The method of claim 10, wherein there is a quasi co-location (QCL) type D relationship for the first resource configuration during training procedure and performance monitoring.

18. A User Equipment (UE) apparatus comprising: a processor; and a memory storing instructions that, when executed by the processor, configure the apparatus to: receive a first resource configuration including a configuration for a first set of beams and an association identifier (ID); perform training or performance monitoring for a model based on measurements of a first set of reference signals corresponding to the first set of beams; receive a second resource configuration including a configuration for a second set of beams and the association ID;27P68965WO1 4924-2349-0130Uidentify that the first set of beams are linked to the second set of beams based on the association ID being included in both the first resource configuration and the second resource configuration; and inferencing predicted measurements for the first set of beams using the model based on measurements of a second set of reference signals corresponding to the second set of beams.

19. The UE apparatus of claim 18, wherein the instructions further configure the apparatus to generate a channel state information (CSI) report based on the predicted measurements for the first set of beams, wherein index values of beams in the CSI report correlate to resource index values configured during data collection for the first set of beams with the association ID.

20. The UE apparatus of claim 19, wherein the CSI includes a resource indicator ordered based on inference results start from best candidate beam.

21. An apparatus comprising means to perform the method of any of claim 1 to claim 17.

22. A computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform the method of any of claim 1 to claim 17.

23. A computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform the method of any of claim 1 to claim 17.

24. An apparatus comprising logic, modules, or circuitry to perform the method of any of claim 1 to claim 17.

25. A system for providing wireless communication comprising means to perform the method of any of claim 1 to claim 17.

26. A baseband processor for a user equipment (UE) that is configured to cause the UE to perform one or more elements of the method of any of claims 1-9.

27. A baseband processor for a base station that is configured to cause the base station to perform one or more elements of the method of any of claims 10-17.28P68965WO1 4924-2349-0130U

Images