Systems and methods of artificial intelligence / machine learning models for beam management spatial prediction

Abstract

A user equipment (UE) uses an artificial intelligence (AI) model to generate predicted received signal strengths of transmit (Tx)-receive (Rx) beam pairs between base station Tx beams and UE Rx beams; identifies predicted best Tx-Rx beam pairs having highest predicted received signal strengths from the Tx-Rx beam pairs; indicates, to the base station, predicted best Tx beams of the predicted best Tx-Rx beam pairs; performs reference signal measurements on the predicted best Tx beams as transmitted by the base station using correspondingly paired predicted best Rx beams of the predicted best Tx-Rx beam pairs to generate measured received signal strengths for the predicted best Tx-Rx beam pairs; and computes a model validity metric for the AI model based on a comparison of the highest predicted received signal strengths and the measured received signal strengths. Other embodiments for an AI model instead sited at the base station are also discussed.

Description

SYSTEMS AND METHODS OF ARTIFICIAL INTELLIGENCE / MACHINE LEARNING MODELS FOR BEAM MANAGEMENT SPATIAL PREDICTIONTECHNICAL FIELD

[0001] This application relates generally to wireless communication systems, including wireless communication systems implementing artificial intelligence (AI) / machine learning (ML) models.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 3 GPP 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.

[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).

[0007] Frequency bands for 5G NR may be separated into two or more different frequency ranges. For example, Frequency Range 1 (FR1) may include frequency bands operating in sub-6 gigahertz (GHz) frequencies, some of which are bands that may be used by previous standards, and may potentially be extended to cover new spectrum offerings from 410 megahertz (MHz) to 7125 MHz. Frequency Range 2 (FR2) may include frequency bands from 24.25 GHz to 52.6 GHz. Note that in some systems, FR2 may also include frequency bands from 52.6 GHz to 71 GHz (or beyond). Bands in the millimeter wave (mmWave) range of FR2 may have smaller coverage but potentially higher available bandwidth than bands in FR1. Skilled persons will recognize these frequency ranges, which are provided by way of example, may change from time to time or from region to region.BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0008] 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.

[0009] FIG. 1 illustrates an operation diagram corresponding to the use of an AI / ML model for RRM performance enhancement, according to embodiments provided herein.

[0010] FIG. 2 illustrates a flow diagram corresponding to the use of an AI / ML model for RRM performance enhancement, according to embodiments provided herein.

[0011] FIG. 3 illustrates a flow diagram for the use of an AI / ML model at a UE to identify a best Tx-Rx beam pair between a base station and the UE, along with various corresponding diagrammatic illustrations, according to embodiments provided herein.

[0012] FIG. 4 illustrates a flow diagram for the use of an AI / ML model at a base station to identify a best Tx-Rx beam pair between the base station and a UE, along with various corresponding diagrammatic illustrations, according to embodiments provided herein.

[0013] FIG. 5 illustrates a flow diagram for the use of an AI / ML model at a UE to identify a best Tx-Rx beam pair between a base station and the UE and that implements monitoring of the AI / ML model, along with various corresponding diagrammatic illustrations, according to embodiments provided herein.

[0014] FIG. 6 illustrates a flow diagram for the use of an AI / ML model at a base station to identify a best Tx-Rx beam pair between the base station and a UE and that implements monitoring of the AI / ML model, along with various corresponding diagrammatic illustrations, according to embodiments provided herein.

[0015] FIG. 7 illustrates a method of a UE, according to embodiments provided herein.

[0016] FIG. 8 illustrated a method of a base station, according to embodiments provided herein.

[0017] FIG. 9 illustrates a method of a base station, according to embodiments provided herein.

[0018] FIG. 10 illustrates an example architecture of a wireless communication system, according to embodiments disclosed herein.

[0019] FIG. 11 illustrates a system for performing signaling between a wireless device and a network device, according to embodiments disclosed herein.DETAILED DESCRIPTION

[0020] 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.

[0021] In various wireless communication systems, overhead penalties corresponding to the transmission of reference signals for radio resource management (RRM) (e g., synchronization signal blocks (SSBs) and / or channel state information-reference signals (CSI-RSs)) are not insignificant. For example, in cases where an SSB-based RRM measurement timing configuration (SMTC) periodicity of 20 milliseconds (ms) and anSSB burst of 5 ms are used in FR2 in an NR wireless communication system, a corresponding SMTC-associated overhead is understood as 25%.

[0022] Accordingly, in some cases, it may be beneficial to reduce a number of beams used at the transmit (Tx) side and / or the receive (Rx) side in Layer 3 (L3)-related measurements (e.g.. beam measurements for RRM purposes).

[0023] Embodiments disclosed herein relate enhancements for RRM functionality with artificial intelligence (Al)Zmachine learning (ML) models. In various embodiments, an L3 measurement delay and / or overhead reduction may be achieved by minimizing a Tx / Rx beam sweeping set through the use of spatial beam prediction by an AI / ML model. For example, an SMTC window duration can be reduced by reducing the beam sweeping factor used by / in the SMTC window (e g., by reducing the number of Tx and / or Rx beams used by the beam sweeping mechanism of the SMTC window).

[0024] In various embodiments, L3 measurement reduction can be achieved by periodically skipping one or more Tx / Rx beam sweeping instances and using a spatial beam prediction by an AI / ML model in place of actual measurements. As one conceptual example, an SMTC periodicity may be increased, representing a conceptual skipping of various instances of SMTC-based measurements that would have otherwise occurred at the lower periodicity. The skipped measurements may be replaced by predictions for those measurements generated at an AI / ML model.

[0025] As a result of such reductions of L3 measurements, throughput can be increased. For example, scheduling restrictions at orthogonal frequency division multiplexing (OFDM) symbols that would have otherwise been used for measurements (e.g., symbols that would have otherwise been used for SSBs to be measured) can be removed.Accordingly, as a result of using an AI / ML model and / or algorithm for RRM functionalities, RRM performance (and therefore overall system performance) can be enhanced.

[0026] FIG. 1 illustrates an operation diagram 100 corresponding to the use of an AI / ML model for RRM performance enhancement, according to embodiments provided herein. The operation diagram 100 may be understood in terms of repeated outer loops 102 over time. As shown, an outer loop 102 may begin with operation according to an AI / ML retraining and full beam measurement mode 104 and eventually proceed into an AI / ML beam measurement mode 106. In some embodiments, the operation diagram 100corresponds to operations at a UE. In some embodiments, the operation diagram 100 corresponds to operations at a base station.

[0027] A transition 108 into an outer loop 102 may occur according to a trigger. The trigger may be periodic or aperiodic in nature.

[0028] In a first example, the operation diagram 100 corresponds to operations at a UE, and the trigger for the transition 108 may be based on a network request 110 received at the UE.

[0029] In a second example, the operation diagram 100 corresponds to operations at either a base station or a UE. In such a case, the trigger for the transition 108 may be based on the expiration of a timer 112 for an AI / ML beam measurement mode 106.

[0030] In a third example, the operation diagram 100 corresponds to operations at either a base station or a UE. In such a case, the trigger for the transition 108 may be a model-monitoring-based deactivation 114 of an AI / ML model that is used for predicting beam measurements (e.g., as described herein).

[0031] At the beginning of an outer loop 102 (after a transition 108), the device (UE, base station) operates in an AI / ML retraining and full beam measurement mode 104. In the AI / ML retraining and full beam measurement mode 104. the device uses (e.g., legacy) beam sweeping without any AI / ML model involvement in order to perform (actual) beam measurements to generate received signal strengths corresponding to one or more Tx-Rx beam pairs. These received signal strengths are then used for RRM. During this period, an AI / ML model at the device for generating predicted beam measurements may be trained and / or retrained on these actual received signal strength measurements.

[0032] Within the outer loop 102, the device may transition 116 from the AI / ML retraining and full beam measurement mode 104 to the AI / ML beam measurement mode 106. In some cases, the AI / ML beam measurement mode 106 may be considered an “inner loop” mode, as illustrated. In the AI / ML beam measurement mode 106, the device uses the AI / ML model to predict received signal strengths corresponding to one or more Tx-Rx beam pairs. These predicted receive signal strengths are then used for RRM.

[0033] As illustrated, the transition 116 into the AI / ML beam measurement mode 106 may occur according to a trigger. The trigger may be periodic or aperiodic in nature.

[0034] In a first example, the operation diagram 100 corresponds to operations at either a base station or a UE. In such a case, the trigger for the transition 116 may be based on the expiration of a timer 118 for an AI / ML retraining and full beam measurement mode 104.

[0035] In a second example, the operation diagram 100 corresponds to operations at a UE, and the trigger for the transition 116 may be based on a network request 120 received at the UE.

[0036] During the AI / ML beam measurement mode 106, the AI / ML model that is being used at the device (UE, base station) may be monitored such that a transition 108 into a new outer loop 102 (and thus into a new AI / ML retraining and full beam measurement mode 104) is triggered in the event that the AI / ML model is not providing sufficiently accurate results (as is discussed herein). Alternatively, as previously described, a network request 110 or a timer 112 may cause the transition 108 to occur.

[0037] Various embodiments disclosed herein relate to the use of an AI / ML model while the device (UE, base station) is in the AI / ML beam measurement mode 106.

[0038] FIG. 2 illustrates a flow diagram 200 corresponding to the use of an AI / ML model for RRM performance enhancement, according to embodiments provided herein. In some cases, the flow diagram 200 corresponds to behavior of a base station that is using an AI / ML model at the base station to perform beam measurement tasks with respect to its communications with a UE. In some cases, the flow diagram 200 corresponds to behavior of a UE that is using an AI / ML model at the UE to perform beam measurement tasks with respect to its communications with a base station.

[0039] The flow diagram 200 operates according to an outer loop 202 that contains functionalities for AI / ML model updating / training 204 an AI / ML for beam measurement, a legacy RRM / beam measurement mode 206 in which legacy beam measurement may be performed in the event that the AI / ML model is being (re)trained, a trigger check 208 against a timer or (in the case of a UE) a network request, an inner loop 210 during which the AI / ML model is used for beam measurement / RRM, and a model monitoring check 212 during which the device evaluates the AI / ML model for accuracy.

[0040] The portions of the outer loop 202 other than the inner loop 210 may be understood to be generally directed toward the (e.g., eventual) use of the inner loop 210, where the AI / ML model is used for beam measurement (and thus beam measurement is relatively more efficient). For example, during the AI / ML model updating / training 204,the AI / ML model to be used during the inner loop 210 is (re)trained (and a legacy RRM / beam measurement mode 206 is used in the meantime, as illustrated).

[0041] Once the model has been successfully retrained, a trigger check 208 is performed. This trigger check 208 determines whether a trigger for entering the inner loop 210, such as a timer expiration and / or a network request have occurred. If not, the system returns to the legacy RRM / beam measurement mode 206. If so, the system proceeds to transition to the inner loop 210. Note that in the event that the device operating the AI / ML model is a UE, it may be that the UE is configured to send a trigger to enter the inner loop 210 to the network and accordingly enter the inner loop 210.

[0042] As part of the inner loop 210, the AI / ML model is used to predict beam measurements. For example, a subset of possible beam measurements (e.g., a subset of beam measurements that are used when in the legacy RRM / beam measurement mode 206) may be (actually) performed. The resulting data from this subset of measurements is then fed to the AI / ML model, which uses the data to generate predicted receive signal strengths for Tx-Rx beam pairs. These predicted receive signal strengths are used to identify predicted best Tx-Rx beam pairs (Tx-Rx beam pairs having the highest predicted receive signal strengths). As part of the inner loop 210, a beam sweep using these predicted best Tx-Rx beam pairs may then be performed in order to identify one of these for use for communications.

[0043] Corresponding to the operation in the inner loop 210, model monitoring checks 212 may occur. A model monitoring check 212 compares one or more predicted receive signal strengths for the Tx-Rx beam pairs to measured receive signal strengths for those Tx-Rx beam pairs. A similarity metric corresponding to the accuracy of the AI / ML model is then computed using the predicted receive signal strengths and the measured received signal strengths.

[0044] In some embodiments, a model monitoring check 212 occurs periodically (e.g., with respect to only some results of the inner loop 210). In other embodiments, a model monitoring check 212 may occur with respect to every set of predicted receive signal strengths generated from the inner loop 210.

[0045] If a model monitoring check 212 passes (e g., if the similarity metric indicates that the AI / ML model of the inner loop 210 is sufficiently accurate, '‘model valid” in FIG. 2), the procedure returns to the trigger check 208. Here, if a trigger (e.g., a timer expiration or, in the case of a UE, a network request) indicates that the device shouldtransition out of the inner loop 210 and back to the legacy RRM / beam measurement mode 206 (such that AI / ML model updating / training 204 may occur), the device does so. Otherwise (if no such trigger is present), the inner loop 210 may continue to be used for subsequent beam measurement tasks.

[0046] In the event that a model monitoring check 212 fails (e.g., if the similarity metric indicates that the AI / ML model of the inner loop 210 is not sufficiently accurate, “model invalid” in FIG. 2), the flow diagram 200 immediately falls back to the legacy RRM / beam measurement mode 206 (such that AI / ML model updating / training 204 may occur). The AI / ML model may become inaccurate due to, for example, mobility of a UE and / or a change in characteristics of the channel between the UE and the base station. In the case that the device operating the AI / ML model is a base station, the base station may signal to a UE that it is not using the AI / ML model / that it has returned to the legacy RRM / beam measurement mode 206. In the event that the device operating the AI / ML model is a UE, the UE may signal to a base station that it is not using the AI / ML model / that it has returned to the legacy RRM / beam measurement mode 206.

[0047] FIG. 3 illustrates a flow diagram 300 for the use of an AI / ML model 306 at a UE 304 to identify a best Tx-Rx beam pair between a base station 302 and the UE 304, along with various corresponding diagrammatic illustrations, according to embodiments provided herein. As illustrated, the base station 302 may transmit 310 reference signals on one or more probing beams 308 to the UE 304. The one or more probing beams 308 used may be a subset of all Tx beams available / useable at the base station 302 per a selected Tx codebook (may be fewer than all the Tx beams of the selected Tx codebook). The particular subset of the one or more probing beams 308 used from the Tx codebook may be specific to the active serving cell. The transmission of the reference signals on the one or more probing beams 308 may occur one Tx beam at a time, according to a beam sweeping fashion.

[0048] Note that in some embodiments, the base station 302 may also transmit 310 information corresponding to the one or more probing beams 308. For example, the base station 302 may transmit 310 beam directions of the one or more probing beams 308 (e g., in terms of angle domain information and / or SSB information for each of the one or more probing beams 308). Further, the base station 302 may also transmit 310 an applicable Tx codebook size for the one or more probing beams 308 (e.g., a size of the Tx codebook from which the one or more probing beams 308 are taken / sampled). Beamdirection and / or codebook information as provided to the UE may enable the UE to properly measure / interpret the reference signals on the one or more probing beams 308 as received.

[0049] The UE 304 scans 312 through one or more of its own Rx beams with respect to the one or more probing beams 308. In other words, the UE 304 takes reference signal measurements of the reference signals on the one or more probing beams 308 using one or more of its own Rx beams, and stores a measured reference signal strength (e.g., reference signal received power (RSRP)) corresponding to each such measurement.

[0050] The set of one or more Rx beams to be used for these measurements may be selected to correspond to the beam direction and / or codebook information for the one or more probing beams 308 as received from the base station 302. For example, the set of one or more Rx beams may be selected from an Rx codebook that is understood to correspond to a Tx codebook indicated by the base station 302, and / or the selection may be based on indicated beam directions from the base station 302.

[0051] In some cases, the UE 304 uses all available Rx beams at the UE 304 (e.g., all Rx beams of a selected Rx codebook) for this procedure. In other cases, the UE 304 may be configured to use only a subset of all the available Rx beams at the UE 304.

[0052] As a result of these measurements, the UE 304 obtains measured reference signal strengths 314 for various Tx-Rx beam pairs (with each such measurement corresponding to a unique combination of one of the one or more probing beams 308 and one of the Rx beams used to measure the reference signals on those one or more probing beams 308, as described).

[0053] These measured reference signal strengths 314 for these Tx-Rx beam pairs are then provided to the AI / ML model 306, which is trained to use the measured reference signal strengths 314 to generate predicted receive signal strengths 316 of a larger overall set (e.g., all) of Tx-Rx beam pairs between the Tx beams of the base station and the Rx beams of the UE.

[0054] The predicted receive signal strengths 316 may be understood in terms of a reference signal strength map 318 for the relationship between the base station 302 and the UE 304. FIG. 3 illustrates the reference signal strength map 318 in terms of three dimensions. The X dimension 320 and the Y dimension 322 respectively correspond to horizontal and vertical indexes that together identify applicable Tx beams of the base station 302 to which a predicted reference signal strength applies. The Z dimension 324then contains multiple X-Y planes of such reference signal strengths, where each individual plane represents reference signal strengths for a different one of the Rx beams of the UE 304 with respect to the Tx beams of the base station 302 (as illustrated).Accordingly, the reference signal strength map 318 is understood to contain predicted reference signal strengths (as predicted by the AI / ML model 306) for the larger overall set (e.g., all) of Tx-Rx beam pairs between the base station 302 and the UE 304.

[0055] The UE 304 then identifies a number K of predicted best Tx-Rx beam pairs that correspond to the highest reference signal strengths among the Tx-Rx beam pairs represented within the reference signal strength map 318. The value of K may be (e.g., previously) configured to the UE 304 by the base station 302, or may be pre-configured per a specification for the type of wireless communication system of the base station 302 and the UE 304. Example values for K include 2, 4, 8, etc.

[0056] With respect to disclosure herein, such a number K of predicted best Tx-Rx beam pairs may sometimes be more simply referred to as the “top-X beam pairs.” Further, when, as in the case of FIG. 3, the UE 304 uses an AI / ML model 306 to generate a reference signal strength map 318, and where the applicable type of reference signal strength is an RSRP. these predicted reference signal strengths associated with these top-X beam pairs may be denoted RSRPAI&uErop-l<-

[0057] FIG. 3 illustrates (by way of example) a case where K = 4. Accordingly, as illustrated, four highest reference signal strengths are identified from the reference signal strength map 318, which ultimately makes the UE 304 aware of the Tx beam and the Rx beam for each of the set of lop-X (K = 4) beam pairs (e.g., according to the correspondence of the X dimension 320. the Y dimension 322, and the Z dimension 324 of the reference signal strength map 318 to these Tx-Rx beams, as has been described).

[0058] After identifying the top-X beam pairs, the UE 304 signals 326 the K predicted best Tx beams of the predicted best Tx-Rx beam pairs (the Tx beams represented in the predicted best Tx-Rx beam pairs) to the base station 302. With respect to disclosure herein, such a number K of predicted best Tx beams may sometimes be more simply referred to as the “top-X Tx beams.” FIG. 3 accordingly illustrates one example of top-X Tx beams 328 as may be understood at the base station 302 according to the signaling 326 from the UE 304.

[0059] The base station 302 proceeds to transmit 330 reference signals on the top-X Tx beams 328. During these transmissions, the UE 304 performs reference signalmeasurements of the reference signals using 332 the K predicted best Rx beams of the predicted best Tx-Rx beam pairs (the Rx beams represented in the predicted best Tx-Rx beam pairs). With respect to disclosure herein, such a number of predicted best Rx beams may sometimes be more simply referred to as the “top- T Rx beams.” At this part of the procedure, the UE 304 measures a reference signal on a Tx beam using its correspondingly paired Rx beam (as this pairing is understood per the set of the predicted best Tx-Rx beam pairs).

[0060] The UE 304 then identifies a highest received signal strength from among the received signal strengths from these (actual) reference signal measurements. The corresponding one of the predicted best Tx-Rx beam pairs (the predicted best Tx-Rx beam pair associated with the highest measured received signal strength) is identified by the UE 304 as the Tx-Rx beam pair to use for communications between the base station 302 and the UE 304 going forward. Accordingly, the UE 304 reports 334 the Tx beam of this Tx-Rx beam pair to the base station 302. The base station 302 then indicates 336 back to the UE 304 that it has determined to use this Tx beam for subsequent communication with the UE 304. The UE 304 then correspondingly determines to use 338 the Rx beam of this Tx-Rx beam pair for subsequent communication with the base station 302.

[0061] FIG. 4 illustrates a flow diagram 400 for the use of an AI / ML model 406 at a base station 402 to identify a best Tx-Rx beam pair between the base station 402 and a UE 404, along with various corresponding diagrammatic illustrations, according to embodiments provided herein. As illustrated, the base station 402 transmits 408 reference signals on one or more probing beams to the UE 404. The one or more probing beams used may be a subset of all Tx beams available / useable at the base station per a selected Tx codebook (may be fewer than all the Tx beams of the selected Tx codebook). The particular subset of the one or more probing beams used from the Tx codebook may be specific to the active serving cell. The transmission of the reference signals on the one or more probing beams may occur one Tx beam at a time, according to a beam sweeping fashion.

[0062] Note that in some embodiments, the base station 402 may also transmit 408 information corresponding to the one or more probing beams. For example, the base station 402 may transmit 408 beam directions of the one or more probing beams (e.g., in terms of angle domain information and / or SSB information for each of the one or moreprobing beams). Further, the base station 402 may also transmit 408 an applicable Tx codebook size for the one or more probing beams (e.g., a size of the Tx codebook from which the one or more probing beams are taken / sampled). Beam direction and / or codebook information as provided to the UE may enable the UE to properly measure / interpret the reference signals on the one or more probing beams as received.

[0063] The UE 404 scans 410 through one or more of its own Rx beams with respect to the one or more probing beams. In other words, the UE 404 takes reference signal measurements of the reference signals on the one or more probing beams using one or more of its own Rx beams, and stores a correspondingly measured reference signal strength (e.g., RSRP) corresponding to each such measurement. The set of one or more Rx beams to be used may be selected to correspond to the beam direction and / or codebook information for the one or more probing beams as received from the base station 402. For example, the set of one or more Rx beams may be selected from an Rx codebook that is understood to correspond to a Tx codebook indicated by the base station 402, and / or the selection may be based on indicated beam directions from the base station 402.

[0064] In some cases, the UE 404 uses all available Rx beams at the UE 404 (e.g., all Rx beams of a selected Rx codebook) for this procedure. In other cases, the UE 404 may be configured to use only a subset of all the available Rx beams at the UE 404.

[0065] As a result of these measurements, the UE 404 obtains measured reference signal strengths for various Tx-Rx beam pairs (with each such measurement corresponding to a unique combination of one of the one or more probing beams and one of the Rx beams used to measure the reference signals on those one or more probing beams, as described).

[0066] The UE 404 then transmits 414 the measured reference signal strengths for these Tx-Rx beam pairs to the base station 402. The UE may indicate the applicable Rx beam for a reported measured reference signal strength to the base station in terms of an identifier (ID) encoding in an angle domain.

[0067] As illustrated, these measured reference signal strengths may be received / understood at the base station in terms of a down-sampled reference signal strength image 412 for Tx-Rx beam pairs between the UE 404 and the base station 402. Within the down-sampled reference signal strength image 412, the reported reference signal strength values corresponding to beam pairs on which reference signalmeasurements were taken are organized according to the Tx beam 416 of the Tx codebook used by the base station 402 corresponding to each measurement on the X-axis and the Rx beam 418 used at the UE 404 corresponding to each measurement on the Y- axis.

[0068] The measured reference signal strengths in the down-sampled reference signal strength image 412 are then provided to an AI / ML model 406 at the base station 402 that is trained to use the down-sampled reference signal strength image 412 to generate a full predicted reference signal strength image 420 having predicted receive signal strengths of a larger overall set (e.g., all) Tx-Rx beam pairs between the Tx beams of the base station and the Rx beams of the UE. Note that the full predicted reference signal strength image 420 is arranged according to Tx beam 416 and Rx beam 418, analogously to the down-sampled reference signal strength image 412.

[0069] The base station 402 then identifies a number K of predicted best Tx-Rx beam pairs that correspond to the highest reference signal strengths among the Tx-Rx beam pairs represented within the full predicted reference signal strength image 420. The value of K may be, for example, dynamically selected by the base station 402, or it may be preconfigured per a specification for the type of wireless communication system of the base station 402 and the UE 404. Example values for A" include 2, 4, 8, etc.

[0070] With respect to disclosure herein, such a number K of predicted best Tx-Rx beam pairs may sometimes be more simply referred to as the "top-X beam pairs.” Further, when, as in the case of FIG. 4, it is the base station 402 that uses an AI / ML model 406 to generate a full predicted reference signal strength image 420, and where the applicable ty pe of reference signal strength is an RSRP, reference signal strengths for these top-X beam pairs may be denoted RSRPA1@NWtop-K-

[0071] FIG. 4 illustrates (by way of example) a case where K = 5. Accordingly, as illustrated, five highest reference signal strengths are identified from the full predicted reference signal strength image 420, which ultimately makes the base station 402 aware of the Tx beam and the Rx beam for each of the set of lop-X (K = 5) beam pairs (e.g., according to the arrangement in the full predicted reference signal strength image 420 along Tx beam 416 and Rx beam 418 dimensions, as has been described).

[0072] By identifying the top-X beam pairs, the base station 402 is made aware of the K predicted best Tx beams of the predicted best Tx-Rx beam pairs (the Tx beams represented in the predicted best Tx-Rx beam pairs). With respect to disclosure herein,such a number K of predicted best Tx beams may sometimes be more simply referred to as theL‘top-A? Tx beams.”

[0073] The base station 402 then transmits 422 reference signals on the top-A? Tx beams. As part of this procedure, as illustrated, the base station 402 may first identify the K predicted best Rx beams of the predicted best Tx-Rx beam pairs (the Rx beams represented in the predicted best Tx-Rx beam pairs) to the UE 404. With respect to disclosure herein, such a number K of predicted best Rx beams may sometimes be more simply referred to as the “top-AT Rx beams.” The reporting of the top-AT Rx beams to the UE enables the UE to use these top-AT Rx beams for the measurements of the reference signals as transmitted 422 on the top-A? Tx beams by the base station 402. The UE 404 accordingly performs the reference signal measurements using 424 the top-A' Rx beams. At this part of the procedure, the UE 404 measures a reference signal on a Tx beam using its correspondingly paired Rx beam (as this pairing is understood per the set of the predicted best Tx-Rx beam pairs).

[0074] The UE 404 then identifies a highest received signal strength from among the received signal strengths from these (actual) reference signal measurements. The corresponding one of the predicted best Tx-Rx beam pairs (the predicted best Tx-Rx beam pair associated with the highest measured received signal strength) is identified by the UE 404 as the Tx-Rx beam pair to use for communications between the base station 402 and the UE 404 going forward. Accordingly, the UE 404 reports 426 the Tx beam of this Tx-Rx beam pair to the base station 402. The base station 402 then indicates 428 back to the UE 404 that it has determined to use this Tx beam for subsequent communication with the UE 404. The UE 404 then correspondingly determines to use 430 the Rx beam of this Tx-Rx beam pair for subsequent communication with the base station 402.

[0075] An AI / ML model as discussed herein may be trained on data (e.g., reference signal strengths generated by reference signal strength measurements) that corresponds to a situation where the UE is in a particular environment / that corresponds to particular characteristics of channel between the base station and the UE. It may be, however, that a UE finds itself in a new channel environment in an unpredictable fashion (e.g., due to mobility and / or to dynamic environmental interference factors). Under such circumstances, an AI / ML beam spatial prediction model may become outdated. The use of an outdated AI / ML model may result in inaccurate received signal strengthpredictions, leading to a selection of a sub-optimal beam pair and thus a degradation in performance.

[0076] Embodiments disclosed herein describe procedures under which such AI / ML model deterioration can be detected, and, in response, deactivation and / or (re)training of the AI / ML model can occur (a process which may include any appropriate signaling to other devices in the system). An AI / ML model monitoring and deactivation can be triggered by either the UE or the network. Additionally, during AI / ML model retraining, the RRM or beam management procedure used between the UE and the base station can fall back to legacy RRM / beam measurement modes.

[0077] FIG. 5 illustrates a flow diagram 502 for the use of an AI / ML model 306 at a UE 304 to identify a best Tx-Rx beam pair between a base station 302 and the UE 304 and that implements monitoring of the AI / ML model 306, along with various corresponding diagrammatic illustrations, according to embodiments provided herein.

[0078] As illustrated by the use of repeated series reference numbers in FIG. 5, the procedure for the use of an AI / ML model 306 sited at the UE 304 proceeds in similar fashion as was discussed previously in relation to the flow diagram 300 of FIG. 3.

[0079] However, in the embodiment illustrated in FIG. 5, as the UE 304 takes the reference signal measurements of the reference signals transmitted by the base station 302 on the top-A Tx beams using 332 the lop- A Rx beams, the UE may understand that it is to use the attendant reference signal received strengths so measured for more than the determination of the best Tx beam that it reports 334 back to the base station 302. As illustrated, these measured reference signal strengths associated with these top-A beam pairs may be denoted RSRPMeasUEtop K.

[0080] For example, the embodiment illustrated in FIG. 5 corresponds to a case where the predicted reference signal strengths associated with these top-A? beam pairs RSRPAi@uEtop-K) and the measured reference signal strengths associated with the top- ? beam pairs (RSRPMeasUEtop K) are together to be used to calculate a model validity metric 504 (denoted “AT’) that indicates / corresponds to the present accuracy of the AI / ML model 306 (e.g., how well the AI / ML model 306 accounts for the presently applicable channel circumstances between the UE 304 and the base station 302). This is denoted in FIG. 5 as M = f(RSRPMeasUEtop-K, RSRPAI@UEtop-K). In some cases, the model validity metric 504 is evaluated based on a comparison of one or more predicted received signalstrengths RSRPAl@UE_Kof the top-A- beam pairs to one or more measured received signal strengths RSRPMeasUEtop Kof the top- ? beam pairs.

[0081] In some cases, to evaluate the model validity metric 504, a difference between each predicted received signal strength in RSRPAI@UEtop Kand its corresponding measured received signal strength in RSRPMeasUEKis evaluated. The resulting values are collected into a vector. Then, this vector is normalized to generate a value that represents the model validity metric 504. This value (the model validity metric 504) may then be compared to a threshold. In cases where the threshold is not exceeded, it is determined that the AI / ML model 306 is sufficiently accurate. In cases where the threshold is exceeded, it is determined that the AI / ML model 306 is not sufficiently accurate. Note that this procedure for calculating and / or evaluating a model validity metric 504 is given by way of example and not by way of limitation. Other mechanisms for calculating and / or evaluating a model validity metric 504 are contemplated.

[0082] With respect to cases where, as in FIG. 5, the AI / ML model 306 is sited at the UE 304, the model validity metric 504 can be ultimately calculated according to one of various different options.

[0083] In a first such option ("Option A" in FIG. 5), after the UE 304 determines boththe UE 304 reports 506 these values to the base station 302. The base station 302 then uses these values to calculate the model validity metric 504.

[0084] In a second such option ("Option B" in FIG. 5), after the UE 304 determines both RSRPAI@UEKand RSRPMeasUEtop K, the UE 304 itself calculates the model validity metric 504. In the event that the model validity metric 504 indicates that the AI / ML model 306 is not sufficiently accurate for the presently applicable channel conditions, the UE 304 may then report 508 the value Mto the base station 302 (and / or additionally or alternatively report 508 more directly an indication to the base station 302 that the AI / ML model 306 is to be deactivated).

[0085] Once the model validity metric 504 is calculated (by whatever option), the procedure represented in the flow diagram 502 may be considered to have a pair of alternative possible outcomes. In a first alternative, corresponding to cases where the model validity metric 504 indicates that the AI / ML model 306 is sufficiently accurate for the presently applicable channel conditions, the base station 302 proceeds to indicate 336its determination to use the best Tx beam going forward for communications with the UE and accordingly use that best Tx beam for those communications while the UE uses the corresponding best Rx beam (e.g., as was described in relation to the flow diagram 300 of FIG. 3).

[0086] The second alternative instead applies in cases where the model validity metric 504 indicates that the AI / ML model 306 is not sufficiently accurate for the presently applicable channel conditions. In such cases, the base station 302 and the UE 304 proceed to model deactivation 510. For example, when the model validity metric 504 corresponding to inaccuracy is calculated and / or used at the base station 302 (corresponding to the case where the UE 304 reports 506 RSRPAI@UEtop-Kand RSRPMeasUEtop-Kto the base station and / or reports the UE-based calculation of Mto the base station 302), then the base station 302 may instruct the UE 304 to deactivate its use of the AI / ML model 306. In another example, when the model validity metric 504 corresponding to inaccuracy instead triggers a UE-driven reaction procedure, the UE 304 may provide signaling to the base station 302 indicating that it has deactivated its use of the AI / ML model 306.

[0087] Once deactivated, the system may fall back or transition to legacy beam measurement modes (where the AI / ML model is not used) and / or participate in retraining the AI / ML model based on corresponding measurement results. Then, some later time, once the AI / ML model 406 is sufficiently retrained, the system may transition back to the use of the predictive beam measurement mechanism using the AI / ML model 306 (e.g., as these functions are described herein).

[0088] FIG. 6 illustrates a flow diagram 600 for the use of an AI / ML model 406 at a base station 402 to identify a best Tx-Rx beam pair between the base station 402 and a UE 404 and that implements monitoring of the AI / ML model 406. along with various corresponding diagrammatic illustrations, according to embodiments provided herein.

[0089] As illustrated by the use of repeated series reference numbers in FIG. 6, the procedure for the use of an AI / ML model 406 sited at the UE 404 proceeds in similar fashion as was discussed previously in relation to the flow diagram 400 of FIG. 4.

[0090] However, in the embodiment illustrated in FIG. 6, as the UE 404 takes the reference signal measurements of the reference signals transmitted by the base station 402 on the top- / / Tx beams using 424 the top-Al Rx beams, the UE may report 602 thereference signal received strengths so measured back to the base station 402 (instead of, for example, merely reporting the determination of the best Tx beam, as in the flow diagram 400 of FIG. 4). As illustrated, these measured reference signal strengths associated with these top- A1beam pairs may be denoted RSRPMeasUEtop K.

[0091] The embodiment illustrated in FIG. 6 corresponds to a case where the predicted reference signal strengths associated with these top- A" beam pairs RSRPAi@Nwtop-K) and the measured reference signal strengths associated with the top -A" beam pairs RSRPMeasUE K) are together to be used to calculate a model validity metric 604 (denoted “AT’) that indicates / corresponds to the present accuracy of the AI / ML model 406 (e.g., how well the AI / ML model 406 accounts for the presently applicable channel circumstances between the UE 404 and the base station 402). This is denoted in FIG. 6 as M = f RSRPMeasUEtop-K, RSRPAI@NWtop KIn some cases, the model validity metric 504 is evaluated based on a comparison of one or more predicted received signal strengths RSRPAI&NWKof the top-A' beam pairs to one or more measured received signal strengths RSRPMeasUEKof the top-A’ beam pairs.

[0092] In some cases, to evaluate the model validity metric 604, a difference between each predicted received signal strength inand its corresponding measured received signal strength in RSRPMeasUEtop-Kis evaluated. The resulting values are collected into a vector. Then, this vector is normalized to generate a value that represents the model validity7metric 604. This value (the model validity metric 604) may then be compared to a threshold. In cases where the threshold is not exceeded, it is determined that the AI / ML model 406 is sufficiently accurate. In cases where the threshold is exceeded, it is determined that the AI / ML model 406 is not sufficiently accurate. Note that this procedure for calculating and / or evaluating a model validity metric 604 is given by way of example and not by way of limitation. Other mechanisms for calculating and / or evaluating a model validity metric 604 are contemplated.

[0093] With respect to cases where, as in FIG. 6, the AI / ML model 406 is sited at the base station 402, the model validity metric 604 is calculated at the base station 402 after the base station receives RSRPMeasUEti:p l<.

[0094] Once the model validity metric 604 is calculated, the procedure represented in the flow diagram 600 may be considered to have a pair of alternative possible outcomes.In a first alternative, corresponding to cases where the model validity metric 604 indicates that the AI / ML model 406 is sufficiently accurate for the presently applicable channel conditions, the base station 402 proceeds to indicate 428 its determination to use the best Tx beam going forward for communications with the UE and accordingly use that best Tx beam for those communications while the UE uses the corresponding best Rx beam (e.g.. as was described in relation to the flow diagram 400 of FIG. 4).

[0095] The second alternative instead applies in cases where the model validity metric 604 indicates that the AI / ML model 406 is not sufficiently accurate for the presently applicable channel conditions. In such cases, the base station 402 and the UE 404 proceed to model deactivation 606. For example, once the model validity metric 604 corresponding to inaccuracy is calculated and / or used at the base station 402, the base station 302 may instruct the UE 404 that the base station is deactivating the AI / ML model 406.

[0096] Once deactivated, the system may fall back or transition to legacy beam measurement modes (where the AI / ML model is not used) and / or participate in retraining the AI / ML model based on corresponding measurement results. Then, some later time, once the AI / ML model 406 is sufficiently retrained, the system may transition back to the use of the predictive beam measurement mechanism using the AI / ML model 406 (e.g., as these functions are described herein).

[0097] FIG. 7 illustrates a method 700 of a UE, according to embodiments provided herein. The illustrated method 700 includes using 702 an Al model at the UE to generate predicted received signal strengths of Tx-Rx beam pairs between Tx beams of a base station and Rx beams of the UE. The method 700 further includes identifying 704 one or more predicted best Tx-Rx beam pairs having one or more highest predicted received signal strengths from the Tx-Rx beam pairs. The method 700 further includes indicating 706, to the base station, one or more predicted best Tx beams of the one or more predicted best Tx-Rx beam pairs. The method 700 further includes performing 708 one or more first reference signal measurements on the one or more predicted best Tx beams as transmitted by the base station using correspondingly paired one or more predicted best Rx beams of the one or more predicted best Tx-Rx beam pairs to generate one or more first measured received signal strengths for the one or more predicted best Tx-Rx beam pairs. The method 700 further includes computing 710 a model validity metric for the Al model based on a comparison of the one or more highest predicted received signalstrengths of the one or more predicted best Tx-Rx beam pairs and the one or more first measured received signal strengths of the one or more predicted best Tx-Rx beam pairs.

[0098] In some embodiments of the method 700, the indicating, to the base station, the one or more predicted best Tx beams comprises indicating angle domain information for the one or more predicted best Tx beams.

[0099] In some embodiments of the method 700, the indicating, to the base station, the one or more predicted best Tx beams comprises indicating SSB information for the one or more predicted best Tx beams.

[0100] In some embodiments, the method 700 further comprises determining, based on the model validity metric, to transition the UE from a first mode that uses the Al model to perform RRM to a second mode that does not use the Al model for the RRM, and transitioning the UE from the first mode to the second mode. Some such embodiments further comprise performing one or more second reference signal measurements on one or more of the Tx beams of the base station using one or more of the Rx beams of the UE to generate one or more second measured received signal strengths while the UE is in the second mode, and training the Al model using the one or more second measured received signal strengths. Some other such embodiments further comprise starting a timer upon the transitioning the UE to the second mode, and transitioning the UE from the second mode to the first mode upon expiration of the timer. Some further such embodiments further comprise receiving, from the base station, while in the second mode, an indication from the base station that the UE is to transition to the first mode, and transitioning the UE from the second mode to the first mode in response to the indication from the base station.

[0101] In some embodiments, the method 700 further comprises sending, to the base station, the model validity metric, receiving, from the base station, in response to the model validity metric, an instruction to transition the UE from a first mode that uses the Al model to perform RRM to a second mode that does not use the Al model for the RRM, and transitioning the UE from the first mode to the second mode. Some such embodiments further comprise performing one or more second reference signal measurements on one or more of the Tx beams of the base station using one or more of the Rx beams of the UE to generate one or more second measured received signal strengths while the UE is in the second mode, and training the Al model using the one or more second measured received signal strengths. Some other such embodiments furthercomprise starting a timer upon the transitioning the UE to the second mode, and transitioning the UE from the second mode to the first mode upon expiration of the timer. Some further such embodiments further comprise receiving, from the base station, while in the second mode, an indication from the base station that the UE is to transition to the first mode, and transitioning the UE from the second mode to the first mode in response to the indication from the base station.

[0102] In some embodiments, the method 700 further comprises determining, based on the model validity metric, to perform a retraining of the Al model, performing second reference signal measurements on the Tx beams of the base station using the Rx beams of the UE. and retraining the Al model based on the second reference signal measurements.

[0103] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 700. This apparatus may be. for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein).

[0104] 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 700. 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).

[0105] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 700. 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 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 700. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein).

[0107] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 700.

[0108] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by aprocessor is to cause the processor to carry out one or more elements of the method 700. 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).

[0109] FIG. 8 illustrates a method 800 of a base station, according to embodiments provided herein. The illustrated method 800 includes using 802 an Al model at the base station to generate predicted received signal strengths of Tx-Rx beam pairs between Tx beams of the base station and Rx beams of a UE. The method 800 further includes identifying 804 one or more predicted best Tx-Rx beam pairs having one or more highest predicted received signal strengths from the Tx-Rx beam pairs. The method 800 further includes indicating 806, to the UE, one or more predicted best Rx beams of the one or more predicted best Tx-Rx beam pairs. The method 800 further includes transmitting 808, to the UE, one or more reference signals on one or more predicted best Tx beams of the one or more predicted best Tx-Rx beam pairs. The method 800 further includes receiving 810, from the UE, one or more first measured received signal strengths for the one or more predicted best Tx-Rx beam pairs. The method 800 further includes computing 812 a model validity metric for the Al model based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx-Rx beam pairs and the one or more first measured received signal strengths of the one or more predicted best Tx-Rx beam pairs.

[0110] In some embodiments of the method 800, the indicating, to the UE, the one or more predicted best Rx beams comprises indicating angle domain information for the one or more predicted best Rx beams.

[0111] In some embodiments, the method 800 further comprises determining, based on the model validity metric, to transition the base station from a first mode that uses the Al model to perform RRM to a second mode that does not use the Al model for the RRM, and transitioning the base station from the first mode to the second mode. Some such embodiments further comprise receiving one or more second measured received signal strengths corresponding to one or more of the Tx-Rx beam pairs between the Tx beams of a base station and the Rx beams of the UE while the base station is in the second mode, and training the Al model based on the one or more second measured received signal strengths. Some other such embodiments further comprise starting a timer uponthe transitioning the base station to the second mode, and transitioning the base station from the second mode to the first mode upon expiration of the timer.

[0112] 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 base station (such as a network device 1118 that is a base station, as described herein).

[0113] 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 800. This non-transitory computer-readable media may be, for example, a memory of a base station (such as a memory71122 of a network device 1118 that is a base station, as described herein).

[0114] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry7to perform one or more elements of the method 800. 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 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 base station (such as a network device 1118 that is a base station, as described herein).

[0116] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 800.

[0117] 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 carry7out one or more elements of the method 800. 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 memory71122 of a network device 1118 that is a base station, as described herein).

[0118] FIG. 9 illustrates a method 900 of a base station, according to embodiments provided herein. The illustrated method 900 includes receiving 902. from a UE, one or more highest predicted received signal strengths of one or more predicted best Tx-Rx beam pairs between the base station and the UE. The method 900 further includes receiving 904, from the UE, one or more measured received signal strengths of the one or more predicted best Tx-Rx beam pairs. The method 900 further includes computing 906 a model validity metric for an Al model at the UE based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx- Rx beam pairs and the one or more measured received signal strengths of the one or more predicted best Tx-Rx beam pairs. The method 900 further includes sending 908, based on the model validity metric, an indication to the UE to transition from a first mode that uses the Al model to perform RRM to a second mode that does not use the Al model for the RRM.

[0119] 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).

[0120] 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).

[0121] 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).

[0122] 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).

[0123] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 900.

[0124] 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).

[0125] 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.

[0126] 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.

[0127] 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.

[0128] 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 UTE and / or NR.

[0129] 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 wayof 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.

[0130] 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.

[0131] 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).

[0132] 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).

[0133] 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 interface 1028 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).

[0134] 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).

[0135] 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.

[0136] 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.

[0137] 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 hardwaredevice, a firmware device, or any combination thereof configured to perform the operations described herein.

[0138] The wireless device 1102 may include a memory 1106. The memory' 1106 may be a non-transitory' computer-readable storage medium that stores instructions 1108 (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.

[0139] 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.

[0140] 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).

[0141] 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).

[0142] 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 for input 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).

[0143] The wireless device 1102 may include a beam measurement module 1116. The beam measurement module 1116 may be implemented via hardware, software, or combinations thereof. For example, the beam measurement 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 beam measurement module 1116 may be integrated within the processor(s) 1104 and / or the transceiver(s) 1110. For example, the beam measurement 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.

[0144] The beam measurement module 1116 may be used for various aspects of the present disclosure, for example, aspects of FIG. 7. The beam measurement module 1116 may configure the wireless device 1102 to: use an Al model at the wireless device 1102 to generate predicted received signal strengths of Tx-Rx beam pairs between Tx beams of a network device 1118 and Rx beams of the wireless device 1102; identify one or more predicted best Tx-Rx beam pairs having one or more highest predicted received signal strengths from the Tx-Rx beam pairs; indicate, to the network device 1118, one or more predicted best Tx beams of the one or more predicted best Tx-Rx beam pairs; perform one or more first reference signal measurements on the one or more predicted best Tx beams as transmitted by the network device 1118 using correspondingly paired one or more predicted best Rx beams of the one or more predicted best Tx-Rx beam pairs to generate one or more first measured received signal strengths for the one or more predicted best Tx-Rx beam pairs; and compute a model validity metric for the Al model based on a comparison of the one or more highest predicted received signal strengths ofthe one or more predicted best Tx-Rx beam pairs and the one or more first measured received signal strengths of the one or more predicted best Tx-Rx beam pairs.

[0145] 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.

[0146] 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.

[0147] 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 from the network device 1118 with other devices (e.g., the wireless device 1102) according to corresponding RATs.

[0148] 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.

[0149] 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.

[0150] The network device 1118 may include a beam measurement module 1132. The beam measurement module 1132 may be implemented via hardware, software, or combinations thereof. For example, the beam measurement 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 beam measurement module 1132 may be integrated within the processor(s) 1120 and / or the transceiver(s) 1126. For example, the beam measurement 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.

[0151] The beam measurement module 1132 may be used for various aspects of the present disclosure, for example, aspects of FIG. 8 and / or FIG. 9. The beam measurement module 1132 may configure the network device 1118 to: use an Al model at the network device 1118 to generate predicted received signal strengths of Tx-Rx beam pairs between Tx beams of the network device 1118 and Rx beams of wireless device 1102; identify one or more predicted best Tx-Rx beam pairs having one or more highest predicted received signal strengths from the Tx-Rx beam pairs; indicate, to the wireless device 1102, one or more predicted best Rx beams of the one or more predicted best Tx-Rx beam pairs; transmit, to the wireless device 1102. one or more reference signals on one or more predicted best Tx beams of the one or more predicted best Tx-Rx beam pairs; receive, from the wireless device 1102, one or more first measured received signal strengths for the one or more predicted best Tx-Rx beam pairs; and compute a model validity metric for the Al model based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx-Rx beam pairs and the one or more first measured received signal strengths of the one or more predicted best Tx-Rx beam pairs. The beam measurement module 1132 may configure the network device 1118 to: receiving, from a wireless device 1102, one or more highest predicted received signal strengths of one or more predicted best Tx-Rx beam pairs between the network device 1118 and the wireless device 1102; receiving, from the wireless device 1102, one or more measured received signal strengths of the one or more predicted best Tx-Rx beam pairs; computing a model validity metric for an Al model at the wireless device 1102 based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx-Rx beam pairs and the one or moremeasured received signal strengths of the one or more predicted best Tx-Rx beam pairs; and sending, based on the model validity metric, an indication to the wireless device 1102 to transition from a first mode that uses the Al model to perform RRM to a second mode that does not use the Al model for the RRM.

[0152] 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.

[0153] 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 to the precise form disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from practice of various embodiments.

[0154] 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.

[0155] 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 forparameters, attributes, aspects, etc. of another embodiment unless specifically disclaimed herein.

[0156] 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.

[0157] 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.

Claims

CLAIMS1. A method of a user equipment (UE), comprising: using an artificial intelligence (Al) model at the UE to generate predicted received signal strengths of transmit (Tx)-receive (Rx) beam pairs between Tx beams of a base station and Rx beams of the UE; identifying one or more predicted best Tx-Rx beam pairs having one or more highest predicted received signal strengths from the Tx-Rx beam pairs; indicating, to the base station, one or more predicted best Tx beams of the one or more predicted best Tx-Rx beam pairs; performing one or more first reference signal measurements on the one or more predicted best Tx beams as transmitted by the base station using correspondingly paired one or more predicted best Rx beams of the one or more predicted best Tx-Rx beam pairs to generate one or more first measured received signal strengths for the one or more predicted best Tx-Rx beam pairs; and computing a model validity metric for the Al model based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx-Rx beam pairs and the one or more first measured received signal strengths of the one or more predicted best Tx-Rx beam pairs.

2. The method of claim 1, wherein the indicating, to the base station, the one or more predicted best Tx beams comprises indicating angle domain information for the one or more predicted best Tx beams.

3. The method of claim 1, wherein the indicating, to the base station, the one or more predicted best Tx beams comprises indicating synchronization signal block (SSB) information for the one or more predicted best Tx beams.

4. The method of claim 1, further comprising: determining, based on the model validity metric, to transition the UE from a first mode that uses the Al model to perform radio resource management (RRM) to a second mode that does not use the Al model for the RRM; and transitioning the UE from the first mode to the second mode.

5. The method of claim 4, further comprising: performing one or more second reference signal measurements on one or more of the Tx beams of the base station using one or more of the Rx beams of the UE to generate one or more second measured received signal strengths while the UE is in the second mode; and training the Al model using the one or more second measured received signal strengths.

6. The method of claim 4, further comprising: starting a timer upon the transitioning the UE to the second mode; and transitioning the UE from the second mode to the first mode upon expiration of the timer.

7. The method of claim 4, further comprising: receiving, from the base station, while in the second mode, an indication from the base station that the UE is to transition to the first mode; and transitioning the UE from the second mode to the first mode in response to the indication from the base station.

8. The method of claim 1, further comprising: sending, to the base station, the model validity metric; receiving, from the base station, in response to the model validity metric, an instruction to transition the UE from a first mode that uses the Al model to perform radio resource management (RRM) to a second mode that does not use the Al model for the RRM; and transitioning the UE from the first mode to the second mode.

9. The method of claim 8, further comprising: performing one or more second reference signal measurements on one or more of the Tx beams of the base station using one or more of the Rx beams of the UE to generate one or more second measured received signal strengths while the UE is in the second mode; and training the Al model using the one or more second measured received signal strengths.

10. The method of claim 8, further comprising: starting a timer upon the transitioning the UE to the second mode; and transitioning the UE from the second mode to the first mode upon expiration of the timer.

11. The method of claim 8, further comprising: receiving, from the base station, while in the second mode, an indication from the base station that the UE is to transition to the first mode; and transitioning the UE from the second mode to the first mode in response to the indication from the base station.

12. The method of claim 1, further comprising: determining, based on the model validity metric, to perform a retraining of the Al model; performing second reference signal measurements on the Tx beams of the base station using the Rx beams of the UE; and retraining the Al model based on the second reference signal measurements.

13. A method of a base station, comprising: using an artificial intelligence (Al) model at the base station to generate predicted received signal strengths of Tx-Rx beam pairs between transmit (Tx) beams of the base station and Rx beams of a user equipment (UE); identifying one or more predicted best Tx-Rx beam pairs having one or more highest predicted received signal strengths from the Tx-Rx beam pairs; indicating, to the UE, one or more predicted best Rx beams of the one or more predicted best Tx-Rx beam pairs; transmitting, to the UE, one or more reference signals on one or more predicted best Tx beams of the one or more predicted best Tx-Rx beam pairs; receiving, from the UE, one or more first measured received signal strengths for the one or more predicted best Tx-Rx beam pairs; and computing a model validity metric for the Al model based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx-Rx beam pairs and the one or more first measured received signal strengths of the one or more predicted best Tx-Rx beam pairs.

14. The method of claim 13, wherein the indicating, to the UE, the one or more predicted best Rx beams comprises indicating angle domain information for the one or more predicted best Rx beams.

15. The method of claim 13, further comprising: determining, based on the model validity’ metric, to transition the base station from a first mode that uses the Al model to perform radio resource management (RRM) to a second mode that does not use the Al model for the RRM; and transitioning the base station from the first mode to the second mode.

16. The method of claim 15, further comprising: receiving one or more second measured received signal strengths corresponding to one or more of the Tx-Rx beam pairs between the Tx beams of a base station and the Rx beams of the UE while the base station is in the second mode; and training the Al model based on the one or more second measured received signal strengths.

17. The method of claim 15, further comprising: starting a timer upon the transitioning the base station to the second mode; and transitioning the base station from the second mode to the first mode upon expiration of the timer.

18. A method of a base station, comprising: receiving, from a user equipment (UE), one or more highest predicted received signal strengths of one or more predicted best transmit (Tx)-receive (Rx) beam pairs between the base station and the UE; receiving, from the UE, one or more measured received signal strengths of the one or more predicted best Tx-Rx beam pairs; computing a model validity metric for an Al model at the UE based on a comparison of the one or more highest predicted received signal strengths of the one or more predicted best Tx-Rx beam pairs and the one or more measured received signal strengths of the one or more predicted best Tx-Rx beam pairs; and sending, based on the model validity metric, an indication to the UE to transition from a first mode that uses the Al model to perform radio resource management (RRM) to a second mode that does not use the Al model for the RRM.

19. An apparatus comprising means to perform the method of any of claim 1 to claim 18.

20. 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 18.

21. An apparatus comprising logic, modules, or circuitry to perform the method of any of claim 1 to claim 18.

22. A baseband processor of a user equipment (UE) that is configured to perform any of claim 1 to claim 12.

Images