Input for machine learning prediction module in wireless communication system

Abstract

A UE (102) receives (306), from a network entity (104), a configuration configuring a first set of RSs associated with an ML prediction module. The UE receives (310) a second set of RSs different from the first set of RSs. The UE transmits (316) a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module. A UE (102) receives (806), from a network entity (104), a configuration configuring a plurality of RSs associated with a plurality of ML prediction modules. The UE receives (808) an indication indicating an RS associated with an ML prediction module. The ML prediction module is being executed at the UE. The UE transmits (816) a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

Description

INPUT FOR MACHINE LEARNING PREDICTION MODULE IN WIRELESS COMMUNICATION SYSTEMCROSS REFERENCE TO RELATED APPLICATION(S)

[0001] This application claims the benefit of and priority to U.S. Provisional ApplicationSerial No. 63 / 666,620, and filed on July 1, 2024, which is expressly incorporated by reference herein in its entirety.TECHNICAL FIELD

[0002] The present disclosure relates generally to wireless communication, and more particularly, to channel prediction using machine learning (ML) modules in wireless communication system.BACKGROUND

[0003] The Third Generation Partnership Project (3GPP) specifies a radio interface referred to as fifth generation (5G) new radio (NR) (5G NR). An architecture for a 5G NR wireless communication system includes a 5G core (5GC) network, a 5G radio access network (5G-RAN), a user equipment (5G UE), etc. The 5G NR architecture seeks to provide increased data rates, decreased latency, and / or increased capacity compared to prior generation cellular communication systems.

[0004] Wireless communication systems, in general, provide various telecommunication services (e g., telephony, video, data, messaging, etc.) based on multiple-access technologies, such as orthogonal frequency division multiple access (OFDMA) technologies, that support communication with multiple UEs. Improvements in mobile broadband continue the progression of such wireless communication technologies. For example, machine learning (ML) based algorithms is used to predict wireless channel metrics.BRIEF SUMMARY

[0005] The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects. This summary neither identifies key or critical elements of all aspects nor delineates the scope of any or all aspects. Its sole purposeis to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.

[0006] A user equipment (UE) uses an ML prediction module to perform prediction of channel metrics (e.g., channel coefficients, channel quality metrics, beam information, etc.). The ML prediction module can be provided to the UE by a network entity (NE). The input of the ML prediction module can be a measurement at the UE side, e.g., the measurement of reference signals (RSs) sent by the NE. The input may be a vector of multiple dimensions, where each dimension corresponds to a RS. The input may include one or more time instances of the measurements. The UE may report the prediction output to the NE, or the UE may use the prediction output locally, e.g., to determine the receive (Rx) beam for downlink (DL) transmission.

[0007] However, the input of the ML prediction module may not be complete, as the UE may not be able to measure an RS at a given time, for example, due to scheduling conflicts. The UE may not be able to collect all the input elements for the ML prediction module. Thus, additional processing is needed for such incomplete input patterns, as ML algorithms assume complete input patterns. Further, the accuracy of the ML prediction depends on the number of inputs. To achieve a minimum required prediction accuracy, the UE may perform a minimum number of measurements as input to the ML prediction module. Alternatively, the output obtained with too few input samples may be of low' reliability, and should be treated accordingly.

[0008] The present disclosure addresses the above-noted and other deficiencies by the UE performing additional processing for incomplete input patterns. In some aspects, the UE may interpolate or extrapolate the non-measured (or missing) input elements based on available measurements, or use the latest / past measurement of the same RS. In other aspects, the UE may use a predefined reserved index for non-measured (or missing) input elements, and include additional input elements to indicate the RS identifier (ID) / timestamp for the measured input elements. In some other aspects, the UE may perform a minimum number of measurements within a predefined time period, which are used as input to the ML prediction module. The UE may indicate the measurements associated with the reported prediction to the NE. The UE may generate a confidence level of the predicted output, at least in part based on the measured RSs used as input.

[0009] In some aspects, a UE receives, from a network entity, a configuration configuring a first set of reference signals (RSs) associated w ith a machine learning(ML) prediction module. The UE receives, from the network entity, a second set of RSs different from the first set of RSs. The UE transmits, to the network entity, a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module.

[0010] In some aspects, a NE transmits, to a UE, a configuration configuring a first set of reference signals (RSs) associated with a machine learning (ML) prediction module. The NE transmits, to the UE, the first set of RSs. The NE receives, from the UE, a prediction report based on a measurement of a second set of RSs included in an input to the ML prediction module, the second set of RSs being different from the first set of RSs.

[0011] To use the DL RSs as input to the ML prediction modules, there are certain limitations on the DL RS configurations. Because there is no dedicated signaling to inform the UE of the RS configurations associated with the ML prediction module, it is difficult for the UE to determine whether the DL RS configuration satisfies the requirements of the ML prediction module. Additionally, when the NW changes a RS configuration, it is important to notify the impacted UEs who run the relevant ML modules.

[0012] The present disclosure addresses the above-noted and other deficiencies by the NE indicating, to the UE, the relevant RS configurations for a ML prediction module that the UE is running. In some aspects, the indication signaling may be triggered by a request of ML module update or in response to a UE inquiry . The UE uses the indication as input to the ML module or uses the indication to adjust the input to the ML module. In other aspects, the indication signaling may trigger an update of the ML module configuration, including reselection of the ML module and activation / deactivation of the ML module. The UE may use the indication to select the appropriate ML module. In some other aspects, the NE may inform the UE when the NE changes the RS configuration for ML prediction.

[0013] In some aspects, a UE receives, from a network entity, a configuration configuring a plurality of reference signals (RSs) associated with a plurality' of machine learning (ML) prediction modules. The UE receives, from the network entity, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality of ML prediction modules. The ML prediction module is being executed at the UE. The UE receives, from the network entity, the RS based on the indication. The UE transmits, to the network entity, a predictionreport output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0014] In some aspects, a NE transmits, to a UE, a configuration configuring a plurality of reference signals (RSs) associated with a plurality of machine learning (ML) prediction modules. The NE transmits, to the UE, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality' of ML prediction modules. The ML prediction module is being executed at the UE. The NE transmits, to the UE, the RS based on the indication. The NE receives, from the UE, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0015] In this way, the accuracy and the reliability of the ML prediction module are improved, and the possibility' of a beam failure is reduced. The proposed solutions may be applied to beam management procedures, such as beam monitoring, beam tracking, and beam failure recovery.BRIEF DESCRIPTION OF THE DRAWINGS

[0016] FIG. 1 illustrates a diagram of a wireless communications system that includes a plurality of user equipments (UEs) and network entities in communication over one or more cells according to an embodiment.

[0017] FIG. 2 illustrates an example of an ML prediction module according to an embodiment.

[0018] FIG. 3 illustrates a signaling diagram of communications between a UE and a network entity for a prediction report based on a measurement of a second set of RSs different from a first configured set of RSs according to an embodiment.

[0019] FIG. 4 illustrates an example of an input that includes both measurements and corresponding timestamps and reference signal (RS) identifiers (IDs) according to an embodiment.

[0020] FIG. 5 illustrates an example of a positional encoding based indices indication to an embodiment.

[0021] FIG. 6 is a flowchart of a method of wireless communication at a UE for the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs according to an embodiment.

[0022] FIG. 7 is a flowchart of a method of wireless communication at a network entity for the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs according to an embodiment.

[0023] FIG. 8 illustrates a signaling diagram of communications between a UE and a network entity for an indication indicating an RS associated with an ML prediction module according to an embodiment.

[0024] FIG. 9 is a flowchart of a method of wireless communication at a UE for an indication indicating an RS associated with an ML prediction module according to an embodiment.

[0025] FIG. 10 is a flowchart of a method of wireless communication at a network entity for an indication indicating an RS associated with an ML prediction module according to an embodiment.

[0026] FIG. 11 is a diagram illustrating a hardware implementation for an example UE apparatus according to some embodiments.

[0027] FIG. 12 is a diagram illustrating a hardware implementation for one or more example network entities according to some embodiments.DETAILED DESCRIPTION

[0028] FIG. 1 illustrates a diagram 100 of a wireless communications system associated with a plurality of cells 190. The wireless communications system includes user equipments (UEs) 102 and base stations / network entities 104. Some base stations may include an aggregated base station architecture and other base stations may include a disaggregated base station architecture. The aggregated base station architecture utilizes a radio protocol stack that is physically or logically integrated within a single radio access network (RAN) node. A disaggregated base station architecture utilizes a protocol stack that is physically or logically distributed among two or more units (e.g., radio unit (RU) 106, distributed unit (DU) 108, central unit (CU) 110). For example, a CU 1 10 is implemented within a RAN node, and one or more DUs 108 may be co-located with the CU 110, or alternatively, may be geographically or virtually distributed throughout one or multiple other RAN nodes. The DUs 108 may be implemented to communicate with one or more RUs 106. Any of the RU 106. the DU 108 and the CU 110 can be implemented as virtual units, such as a virtual radio unit (VRU), a virtual distributed unit (VDU), or a virtual central unit (VCU). The base station / network entity 104 (e.g., an aggregated base station or disaggregatedunits of the base station, such as the RU 106 or the DU 108). may be referred to as a transmission reception point (TRP).

[0029] Operations of the base station 104 and / or network designs may be based on aggregation characteristics of base station functionality. For example, disaggregated base station architectures are utilized in an integrated access backhaul (IAB) network, an open-radio access network (O-RAN) network, or a virtualized radio access network (vRAN), which may also be referred to a cloud radio access network (C- RAN). Disaggregation may include distributing functionality across the two or more units at various physical locations, as well as distributing functionality for at least one unit virtually, which can enable flexibility in network designs. The various units of the disaggregated base station architecture, or the disaggregated RAN architecture, can be configured for wired or wireless communication with at least one other unit. For example, the base stations 104d, 104e and / or the RUs 106a, 106b, 106c, 106d may communicate with the UEs 102a, 102b. 102c. 102d, and / or 102s via one or more radio frequency (RF) access links based on a Uu interface. In examples, multiple RUs 106 and / or base stations 104 may simultaneously serve the UEs 102, such as by intracell and / or inter-cell access links between the UEs 102 and the RUs 106 / base stations 104.

[0030] The RU 106, the DU 108, and the CU 110 may include (or may be coupled to) one or more interfaces configured to transmit or receive information / signals via a wired or wireless transmission medium. For example, a wired interface can be configured to transmit or receive the information / signals over a wired transmission medium, such as via the fronthaul link 160 between the RU 106d and the baseband unit (BBU) 1 12 of the base station 104d associated with the cell 190d. The BBU 112 includes a DU 108 and aCU 110, which may also have a wired interface (e.g., midhaul link) configured between the DU 108 and the CU 110 to transmit or receive the information / signals between the DU 108 and the CU 110. In further examples, a wireless interface, which may include a receiver, a transmitter, or a transceiver, such as an RF transceiver, configured to transmit and / or receive the information / signals via the wireless transmission medium, such as for information communicated between the RU 106a of the cell 190a and the base station 104e of the cell 190e via cross-cell communication beams 136-138 of the RU 106a and the base station 104e.

[0031] The RUs 106 may be configured to implement lower layer functionality. For example, the RU 106 is controlled by the DU 108 and may correspond to a logicalnode that hosts RF processing functions, or lower layer PHY functionality, such as execution of fast Fourier transform (FFT), inverse FFT (iFFT), digital beamforming, physical random access channel (PRACH) extraction and filtering, etc. The functionality of the RU 106 may be based on the functional split, such as a functional split of lower layers.

[0032] The RUs 106 may transmit or receive over-the-air (OTA) communication with one or more UEs 102. For example, the RU 106b of the cell 190b communicates with the UE 102b of the cell 190b via a first set of communication beams 132 of the RU 106b and a second set of communication beams 134b of the UE 102b, which may correspond to inter-cell communication beams or, in some examples, cross-cell communication beams. For instance, the UE 102b of the cell 190b may communicate with the RU 106a of the cell 190a via a third set of communication beams 134a of the UE 102b and a fourth set of communication beams 136 of the RU 106a. DUs 108 can control both real-time and non-real-time features of control plane and user plane communications of the RUs 106.

[0033] Any combination of the RU 106, the DU 108, and the CU 110, or reference thereto individually, may correspond to a base station 104. Thus, the base station 104 may include at least one of the RU 106, the DU 108, or the CU 110. The base stations 104 provide the UEs 102 with access to a core network. The base stations 104 may relay communications between the UEs 102 and the core network (not shown). The base stations 104 may be associated with macrocells for higher-power cellular base stations and / or small cells for lower-power cellular base stations. For example, the cell 190e may correspond to a macrocell, whereas the cells 190a-190d may correspond to small cells. Small cells include femtocells, picocells, microcells, etc. A network that includes at least one macrocell and at least one small cell may be referred to as a “heterogeneous network.”

[0034] Transmissions from a UE 102 to a base station 104 / RU 106 are referred to as uplink (UL) transmissions, whereas transmissions from the base station 104 / RU 106 to the UE 102 are referred to as downlink (DL) transmissions. Uplink transmissions may also be referred to as reverse link transmissions and downlink transmissions may also be referred to as forward link transmissions. For example, the RU 106d utilizes antennas of the base station 104d of cell 190d to transmit a downlink / forward link communication to the UE 102d or receive an uplink / reverse link communication fromthe UE 102d based on the Uu interface associated with the access link between the UE 102d and the base station 104d / RU 106d.

[0035] Communication links between the UEs 102 and the base stations 104 / RUs 106 may be based on multiple-input and multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming, and / or transmit diversity. The communication links may be associated with one or more earners. The UEs 102 and the base stations 104 / RUs 106 may utilize a spectrum bandwidth of Y MHz (e.g., 5, 10, 15, 20, 100, 400, 800, 1600, 2000, etc. MHz) per carrier allocated in a carrier aggregation of up to a total of Yx MHz, where x component carriers (CCs) are used for communication in each of the uplink and downlink directions. The carriers may or may not be adjacent to each other along a frequency spectrum. In examples, uplink and downlink carriers may be allocated in an asymmetric manner, with more or fewer carriers allocated to either the uplink or the downlink. A primary component carrier and one or more secondary component carriers may be included in the component carriers. The primary component carrier may be associated with a primary cell (PCell) and a secondary component carrier may be associated with a secondary7cell (SCell).

[0036] Some UEs 102, such as the UEs 102a and 102s, may perform device-to-device (D2D) communications over sidelink. For example, a sidelink communication / D2D link utilizes a spectrum for a wireless wide area network (WWAN) associated with uplink and downlink communications. Such sidelink / D2D communication may be performed through various wireless communications systems, such as wireless fidelity (Wi-Fi) systems. Bluetooth systems, Long Term Evolution (LTE) systems, New Radio (NR) systems, etc.

[0037] The UEs 102 and the base stations 104 / RUs 106 may each include a plurality of antennas. The plurality of antennas may correspond to antenna elements, antenna panels, and / or antenna arrays that may facilitate beamforming operations. For example, the RU 106b transmits a downlink beamformed signal based on a first set of communication beams 132 to the UE 102b in one or more transmit directions of the RU 106b. The UE 102b may receive the downlink beamformed signal based on a second set of communication beams 134b from the RU 106b in one or more receive directions of the UE 102b. In a further example, the UE 102b may also transmit an uplink beamformed signal (e.g., sounding reference signal (SRS)) to the RU 106b based on the second set of communication beams 134b in one or more transmit directions of the UE 102b. The RU 106b may receive the uplink beamformed signalfrom the UE 102b in one or more receive directions of the RU 106b. The UE 102b may perform beam training to determine the best receive and transmit directions for the beamformed signals. The transmit and receive directions for the UEs 102 and the base stations 104 / RUs 106 may or may not be the same.

[0038] In further examples, beamformed signals may be communicated between a first base station / RU 106a and a second base station 104e. For instance, the base station 104e of the cell 190e may transmit a beamformed signal to the RU 106a based on the communication beams 138 in one or more transmit directions of the base station 104e. The RU 106a may receive the beamformed signal from the base station 104e of the cell 190e based on the RU communication beams 136 in one or more receive directions of the RU 106a. In further examples, the base station 104e transmits a downlink beamformed signal to the UE 102e based on the communication beams 138 in one or more transmit directions of the base station 104e. The UE 102e receives the downlink beamformed signal from the base station 104e based on UE communication beams 130 in one or more receive directions of the UE 102e. The UE 102e may also transmit an uplink beamformed signal to the base station 104e based on the UE communication beams 130 in one or more transmit directions of the UE 102e, such that the base station 104e may receive the uplink beamformed signal from the UE 102e in one or more receive directions of the base station 104e.

[0039] The base station 104 may include and / or be referred to as a network entity. That is, “network entity'’ may refer to the base station 104 or at least one unit of the base station 104, such as the RU 106, the DU 108. and / or the CU 110. The base station 104 may also include and / or be referred to as a next generation evolved Node B (ng- eNB), a next generation NB (gNB), an evolved NB (eNB), an access point, a base transceiver station, a radio base station, a radio transceiver, a transceiver function, a basic service set (BSS), an extended service set (ESS), a TRP, a network node, network equipment, or other related terminology. The base station 104 or an entity at the base station 104 can be implemented as an IAB node, a relay node, a sidelink node, an aggregated (monolithic) base station, or a disaggregated base station including one or more RUs 106, DUs 108, and / or CUs 110. A set of aggregated or disaggregated base stations may be referred to as a next generation-radio access network (NG-RAN). In some examples, the UE 102a operates in dual connectivity (DC) with the base station 104e and the base station / RU 106a. In such cases, the basestation 104e can be a master node and the base station / RU 160a can be a secondary node.

[0040] Uplink / downlink signaling may also be communicated via a satellite positioning system (SPS) 114. In an example, the SPS 114 associated with the cell 190c may be in communication with one or more UEs 102. such as the UE 102c, and one or more base stations 104 / RUs 106, such as the RU 106c. The SPS 114 may correspond to one or more of a Global Navigation Satellite System (GNSS), a global position system (GPS), a non-terrestrial network (NTN), or other satellite position / location sy stem. The SPS 114 may be associated with LTE signals, NR signals (e.g., based on round trip time (RTT) and / or multi-RTT), wireless local area network (WLAN) signals, a terrestrial beacon system (TBS), sensor-based information, NR enhanced cell ID (NR E-CID) techniques, downlink angle-of-departure (DL-AoD), downlink time difference of arrival (DL-TDOA), uplink time difference of arrival (UL-TDOA), uplink angle-of-arrival (UL-AoA). and / or other systems, signals, or sensors.

[0041] Still referring to FIG. 1, in certain aspects, any of the UEs 102 may include a report component 140 configured to receive, from a network entity, a configuration configuring a first set of RSs associated with an ML prediction module. The report component 140 is configured to receive, from the network entity, a second set of RSs different from the first set of RSs. The report component 140 is configured to transmit, to the netw ork entity, a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module.

[0042] In certain aspects, any of the base stations 104 or a network entity of the base stations 104 may include a configuration component 150 configured to transmit, to a UE, a configuration configuring a first set of RSs associated with an ML prediction module. The configuration component 150 is configured to transmit, to the UE, the first set of RSs. The configuration component 150 is configured to receive, from the UE, a prediction report based on a measurement of a second set of RSs included in an input to the ML prediction module, the second set of RSs being different from the first set of RSs.

[0043] In certain aspects, any of the UEs 102 may include a report component 140 configured to receive, from the network entity, a configuration configuring a plurality of RSs associated with a plurality of ML prediction modules. The report component 140 is configured to receive, from the network entity, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality7ofML prediction modules. The ML prediction module is being executed at the UE. The report component 140 is configured to receive, from the network entity, the RS based on the indication. The report component 140 is configured to transmit, to the network entity, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0044] In certain aspects, any of the base stations 104 or a network entity of the base stations 104 may include a configuration component 150 configured to transmit, to a UE. a configuration configuring a plurality' of RSs associated with a plurality of ML prediction modules. The configuration component 150 is configured to transmit, to the UE, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality of ML prediction modules. The ML prediction module is being executed at the UE. The configuration component 150 is configured to transmit, to the UE, the RS based on the indication, configuration component 150 is configured to receive, from the UE, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0045] Accordingly, FIG. 1 describes a wireless communication system that may be implemented in connection with aspects of one or more other figures described herein. Further, although the following description may be focused on 5G NR, the concepts described herein may be applicable to other similar areas, such as 5G- Advanced and future versions, LTE, LTE-advanced (LTE-A), and other wireless technologies, such as 6G.

[0046] ML based algorithms may be used to predict wireless channel metrics. The predicted wireless channel metrics may include channel coefficients, channel quality metrics (e.g., reference signal received power (RSSP), received signal strength indicator (RSSI). signal-to-noise and interference ratio (SINR)), and beam related metrics (e.g., beam index) in frequency range 2 (FR2). The predicted channel metrics may correspond to a channel in a future time (e.g., time domain prediction), a channel at present where UE has not been performing measurements (e.g., spatial domain prediction), or a combination of the above. One example of spatial domain prediction is that UE measures the RSRP in beamforming direction A to predict the RSRP in another beamforming direction B.

[0047] FIG. 2 illustrates a diagram 200 of an example of an ML prediction module 205 according to an embodiment. The ML prediction module 205 may be provided by thenetwork entity (NE) 104 or the UE vendor, or a third party service provider. The input of the ML prediction module 205 may be the measurements at the UE 102 side, e.g., measurements of RSs sent by the network entity. The input (e.g., 204a, 204b, or 204c) may be a vector of multiple dimensions, where each dimension corresponds to a RS. The input (e.g.. 204a, 204b, or 204c) may include one or more time instances in one or more time sequences of the measurements, where different elements in the same time sequence correspond to different measurement time instances. The UE 102 may report the prediction output (e.g., 206a, 206b, or 206c) to the NE 104, or the UE may use the prediction output locally, e.g., to determine the receive (Rx) beam for downlink (DL) transmission without notifying the NE.

[0048] As illustrated in FIG. 2, when the current time is N, the input vectors 204a X, 204b Y, and 204c Z represent the time sequences of current and past measurements associated with different RSs. In the input vectors 204a X, 204b Y, and 204c Z represent, some elements, or samples, (e.g., X(N-K+1). Y(N-K+1), Z(N)) are not available at the UE 102. The input of the ML prediction module may not be complete, as UE may not be able to measure a RS at a given time. For example, due to scheduling conflicts, the UE may skip measuring the RS to receive another data transmission. The RS transmission may occur in synchronizations signal and physical broadcast channel block (SSB) measurement timing configuration (SMTC) window, where the UE is scheduled to measure other cell RSs. In some aspects, the UE may skip certain measurement occasions to save power. The UE may use the RS measurement occasion to train / adjust a Rx beams, and the measurement of the RS need not correspond to the optimal received RSRP. The input of the ML prediction module includes multiple input elements, or items, such as the input vectors 204a X, 204b Y, and 204c Z, however, some of the input elements or items (e.g., X(N-K+1), Y(N- K+l), Z(N)) are not available due to the reasons discussed above. The UE 102 may not be able to collect all the input elements or items for the ML prediction module. Such input pattern with missing input elements or items may be referred to as an incomplete input pattern. The input pattern of the ML prediction module may include the incomplete input pattern and a complete input pattern.

[0049] In some examples, the input of the ML prediction module includes input elements associated with a first set of RSs configured by the NE 104. The NE 104 configures each RS of the first set of RSs with a time instance (or occasion), resource, frequency domain parameter, sequence, etc. For the incomplete input pattern, the input elementsinclude a plurality of measured elements associated with a second set of RSs. The second set of RSs is different from the first set of RSs. As discussed above, the configured first set of RSs may include missing elements, due to one or more missing measurements. Thus, the second set of RSs may be a subset of the first set of RSs. The first set of RSs includes the second set of RSs and the missing elements. As an example, the NE 104 configures the first set of RSs 204a X (X(N-K), X(N-K+1)... , X(N)), 204b Y (Y(N-K), Y(N-K+1)... , Y(N)), and 204c Z (Z(N-K), Z(N-K+1)... , Z(N)). Each RS of the first set of RSs includes at least a time instance. Different elements in the same time sequence correspond to different RSs in the first set of RSs. Here, the missing elements are X(N-K+1), Y(N-K+1), Z(N). The second set of RSs includes measured elements (e.g., non-missing elements) of the first set of RSs. Thus, the second set of RSs includes X(N-K), X(N), Y(N-K), Y(N), Z(N-K), Z(N-K+1), etc. The UE 120 may generate one or more non-measured elements based on the measured elements. A difference between the first and second set of RSs is the missing elements. The configured first set of RSs includes the missing elements, while the second set of RSs does not include the missing elements. The one or more non-measured elements are associated with the difference between the first and second set of RSs (e.g., the missing elements). The configured first set of RSs includes the missing elements (e.g.. one or more non-measured elements) and the second set of RSs. The one or more nonmeasured elements are generated to be included in the input to the ML prediction module 205 as substitutes or replacements to the missing elements. Accordingly, the measurement of the second set of RSs and the one or more non-measured elements form a complete input pattern used as the input to the ML prediction module 205. That is, the input to the ML prediction module 205 includes the measurement of the second set of RSs and the one or more non-measured elements. In some aspects, the second set of RSs may not be a subset of the first set of RSs. For example, the NE may configure one or more additional RSs for use when there is an incomplete input pattern.

[0050] The output 206a Y (Y(N+1), ... ), 206b T (T(N+1), ... ), and 206c B(B(N+1), ... ) of the ML prediction module correspond to the predicted channel metrics of different RSs. The ML prediction module 205 may be a channel module prediction module, which is a pre-trained ML based algorithm to generate the predicted channel metrics based on the input vectors. The output vectors 206a Y, 206b T, and 206c B represent the time sequences of predicted elements associated wi th different RSs. The output(e.g., 206a Y, 206b T, or 206c B) may include one or more time instances in one or more time sequences, where different elements in the same time sequence correspond to different time instances. The predicted channel metrics in the output and the measurements in the input may or may not correspond to the same time sequence of RS. As an example, the input and output may correspond to the same time sequence of RS Y, as illustrated in FIG. 2.

[0051] The UE may perform additional processing for the incomplete input pattern, as a regular ML algorithm may assume a complete input pattern. The processing procedure may be indicated to the UE by the NE or be predefined in the standards.

[0052] The accuracy of an ML prediction by the ML prediction module may depend on the number of measured input elements. To achieve a predefined prediction accuracy, the UE may perform a minimum set of measurements to be used in the input to the ML prediction module. Alternatively, the output obtained with too few measured input elements (e.g., input samples) may be of low reliability, and should be treated accordingly. For example, the UE may skip reporting such output or notify the NE regarding the reliability issue. When the input pattern is the incomplete input pattern, the UE may perform a minimum set of measurements and additional processing to increase the prediction accuracy and reliability.

[0053] FIG. 3 illustrates a signaling diagram 300 of communications between the UE 102 and the network entity 104 for a prediction report based on the measurement of the second set of RSs different from the first configured set of RSs according to an embodiment. The network entity 104 may correspond to a base station or a unit of a base station, such as the RU 106. the DU 108. the CU 110, etc.

[0054] Referring to FIG. 3, the UE 102 may optionally transmit 302, to the network entity104, a UE capability message indicating a UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs. The UE capability message may indicate support of ML based channel state information (CSI) prediction using an incomplete input pattern. The NE 104 may receive 302, from the UE 102, a UE capability message indicating the UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs. The UE 102 may indicate the UE capability regarding supporting the prediction report based on the measurement of the second set of RSs, a subset of the first configured set of RSs.

[0055] The UE 102 receives 306, from the NE 104. a configuration configuring the first set of RSs associated with an ML prediction module. The first set of RSs includes one or more RSs. The NE 104 transmits 306, to the UE 102, the configuration configuring the first set of RSs associated with the ML prediction module. In some examples, the configuration further configures the ML prediction module. In some other examples, the configuration further configures a plurality of ML prediction modules, and the ML prediction module is included in the plurality of ML prediction modules. The NE 104 transmits 310, to the UE 102, the first set of RSs based on the configuration. The UE 102 receives 310, from the network entity 104, the second set of RSs different from the first set of RSs. As discussed above, the UE may skip measuring some RSs due to scheduling conflicts. As such, the second set of RSs may be a subset of the first set of RSs. In some aspects, the second set of RSs may not be a subset of the first set of RSs. For example, the NE may configure one or more additional RSs for use when there is an incomplete input pattern.

[0056] The UE 102 performs 311 the measurement of the second set of RSs. The measurement of the second set of RSs is used in the input to the ML prediction module to predict the wireless channel quality metrics (e.g., RSSP, RSSI, SINR), and beam related metrics (e.g.. beam index). The measurement of the second set of RSs is included in the input to the ML prediction module. The input to the ML prediction module includes the measurement of the second set of RSs and one or more nonmeasured elements. When the input pattern is the incomplete input pattern, some of the input elements are not available through measurements. The input includes multiple input elements including a set of measured input elements based on the measurement of the second set of RSs. However, a first number of the first set of RSs is greater than a second number of the second set of RSs. The second number of the measured second set of input elements is less than the first number of the first set of RSs. The total number of the multiple input elements is the first number of the first set of RSs.

[0057] The UE 102 may generate 312 the set of non-measured input elements. The multiple input elements may further include the set of non-measured input elements. In some examples, the set of non-measured input elements are generated based on an interpolation of the set of measured input elements or an immediately prior measured input item of a same corresponding RS. The UE 102 may interpolate or extrapolate the non-measured input elements. The UE102 may estimate the non-measured input based on available measurements. The interpolation may be performed in at least one of the time domain or spatial domain. The NE 104 may indicate to the UE 102 to perform the interpolation or extrapolation. As an example, the UE 102 may use an immediately prior measured input item of the same corresponding RS, e.g., the latest measurement of the same RS in the past, for a nonmeasured input element.

[0058] In some other examples, each of the set of non-measured input elements includes a predefined reserved index. The UE 102 may use a predefined reser ed index for the non-measured input elements. The reserved index may be specific to a beam (e.g., beam index) or a time stamp. The reserved index may be associated with the beam or the time stamp. The NE may configure the reserved index to the UE. Alternatively, the reserved index may be pre-defined. As an example, the reserv ed index may correspond to an out-of-range index, or a lowest or highest value of the index.

[0059] In some other examples, each of the set of measured input elements includes at least one of: a time stamp, an identification (ID) of a corresponding RS, or a positional encoding sequence. The UE 120 may include additional input item(s) to indicate the time stamp, the ID of the corresponding RS, or the positional encoding sequence for each of the set of measured input elements. As illustrated in FIG. 4, the additional input elements, 404b, and 404c, are provided to indicate the indices corresponding to the measurement of the RS 404a (Zl, Z2, Z3, . . . ). The indices maybe the time stamp, e.g.. 404b (K1, K2, K3. ... ). The indices may be the RS ID, e.g., 404c (RS1. RS2. RS3, ... ).

[0060] The indices of the measurement input elements may be indicated by using different positional encoding sequences and added to the output of a neural network (NN) layer. In a transformer, the positional encoding sequence may be selected based on the indices of the set of measured input elements. For example, the transformer is used to convert the input data into a format that can be processed by the ML prediction model. The transformer may be a function that converts the raw input data into a more structured format, such as a matrix, that can be processed by the ML prediction model. The transformer may use a positional encoding scheme, where each position / index is mapped to a vector. The positional encoding may describe the location or position of an entity in a sequence so that each position is assigned a unique representation. As illustrated in FIG. 5, the indices of themeasurement input elements may be indicated by positional encoding SI 507a. positional encoding S2 507b, and positional encoding S3 507c, and added to the NN layer 503a to generate the NN layer 503b. The positional encoding sequences may be determined based on at least one of a predefined rule or indicated to the UE by the NE.

[0061] Referring back to FIG. 3, when the incomplete input pattern meets certain criteria, the UE 120 may send an indication to the NE to indicate the incomplete input pattern meets the certain criteria, and the UE 102 may skip sending the prediction report. As an example, in the corresponding incomplete input pattern, when the number of the non-measured input elements is above a number threshold, or the percentage of the non-measured input elements over the total number of the input elements is above a percentage threshold, the UE 102 may skip transmitting the prediction report. The UE 120 may send the indication to the NE to indicate the number of the non-measured input elements is above the number threshold, or the percentage of the non-measured input elements over the total number of the input elements is above the percentage threshold. Otherwise, the UE may perform the interpolation or extrapolation based on a predefined rule or a rule indicated by the NE. The procedures to handle the incomplete input pattern may be indicated by the NE 104 or predefined in the standards.

[0062] For the same prediction task, the NE 104 may configure the UE 102 with multiple ML prediction modules (e.g., binaries) for different input patterns. In some examples, each of the multiple ML prediction modules corresponds to a corresponding input pattern. Different ML prediction modules may correspond to different input patterns. The input pattern is determined based on the input elements. The input pattern may include a plurality of input patterns. The plurality of ML prediction modules may correspond to a plurality of input patterns. The plurality of input patterns may include a complete input pattern and one or more incomplete input patterns. The UE 102 may select 313 the ML prediction module from the multiple ML prediction modules based on an input pattern of the plurality' of input patterns. In some examples, the multiple ML prediction modules correspond to different measurement patterns, e.g.. for different beams and / or different measurement time instances of the same beams. The UE 102 may select 313 which ML module to use based on a measurement pattern of the measurement patterns.The UE 102 may select 313 which ML module to use based on the measured input elements based on the second set of RSs.

[0063] In some examples, the multiple ML prediction modules correspond to the same measurement input pattern, but with different levels of completeness. As an example, when the number of the non-measured input elements is less than a number X, or the percentage of the non-measured input elements over the total number of the input elements is less than a number X, the UE 102 may select 313 ML prediction module 1: otherwise, the UE 102 may select 313 ML prediction module 2. The number X may be predefined or indicated by the NE 104.

[0064] For the incomplete input pattern, the UE 102 may use different approaches for different ML prediction modules. The UE 102 may use different approaches to handle the incomplete input pattern based on the incomplete input pattern. As an example, different interpolation or extrapolation equations may be applied to different non-measured input elements. The UE 102 may also use different approaches to handle the incomplete input pattern based on the completeness of the incomplete input pattern. For example, the UE 102 may use different approaches (e.g., selecting different ML prediction modules) based on the number of the nonmeasured input elements or the percentage of the non-measured input elements over the total number of the input elements.

[0065] Continuing with reference to FIG. 3, in some examples, the configuration indicates a minimum number for the set of measured input elements within a predefined time period. To perform the prediction task, the UE 102 may be required to perform a minimum number of measurements within the predefined time period which are used as the input to the ML prediction module.

[0066] The UE 102 may be configured with multiple ML prediction modules for the same task, where each ML prediction module may have a different minimum requirement of the input pattern, e.g., in terms of the number of measured input elements (e.g., valid RS measurement) within the predefined time period. For example, each ML prediction module of the multiple ML prediction modules includes a corresponding minimum number of measured input elements within the predefined time period. The UE 102 may select 313 the ML prediction module from the multiple ML prediction modules based on the corresponding minimum number of measured input elements. The UE 102 may indicate which module was selected to run the prediction in the prediction report.

[0067] In some examples, the UE 102 may measure a minimum number of RSs in the set of RSs. In some examples, the UE 102 may measure an RS at least every X milliseconds (or symbols, slots, subframes, etc.), a maximum time interval, which may be predefined or indicated by the NE.

[0068] When the UE 102 cannot meet the measurement requirements, the output of the ML prediction module may not be reliable. As an example, the UE 102 may not run the ML prediction module unless the measurement requirement is met. As another example, the UE 102 may skip reporting the output when the measurement requirement is not met. In this situation, a reserved index may be used in the report. Alternatively, the UE 102 may report stale information in the report in this case, e.g., the most recent measured value or the most recent predicted value which meets the measurement requirement.

[0069] In some examples, the UE 102 may need to indicate the measurements, e.g., the set of measured input elements, associated with the reported prediction to the NE, e.g., in the prediction report. The UE 102 may indicate the number of the RSs that the UE measured for the input. The UE 102 may indicate which RSs the UE measured as the input. The UE 102 may indicate which measurement pattern of RSs the UE measured as the input, e.g.. a pattern ID from multiple preconfigured measurement patterns. The UE may indicate the set of measured input elements in the prediction report.

[0070] In some examples, the UE 102 may generate 314 a confidence level based on the measured second set of RSs. The UE 102 UE may generate 314 the confidence level of the predicted output based at least in part on the measured second set of RSs as the input, e g., the set of measured input elements.

[0071] The confidence level may be generated based on at least one: a variance or a standard deviation of a prediction result, a success probability of a prediction result, a probability distribution that a predicted beam is a qualified candidate beam, or a function of at least one of the second number of the second set of measured input elements, identifications (IDs) of measured RSs, or a periodicity of the measurement. The confidence level may be a variance or a standard deviation of the prediction, an average expected error, or a success probability of the prediction (e.g., the probability that the predicted RSRP is within 1 dB from the ground truth).

[0072] The confidence level may be a probability distribution that the predicted beam will be one of the qualified candidate beams. The qualified candidate beams may bethe top X beams or within Y dB from the ground truth best beam. The confidence level may be a function of the number of measured RSs as the input, the measured RS IDs, or a periodicity of the measurements. In some examples, the confidence level may be generated from a table (e g., Table 1 as below), which may be configured at the UE 102. The confidence level may be an output of the ML prediction module.Table 1. Illustration of an example of generating a confidence level

[0073] The UE 102 transmits 316, to the network entity 104, the prediction report based on the measurement of the second set of RSs being used in the input to the ML prediction module. The input to the ML prediction module may also include the plurality of non-measured elements generated in operation 312. The NE104 receives 316, from the UE 102, the prediction report based on the measurement of the second set of RSs being used in the input to the ML prediction module. In some aspects, the prediction report may include a beam prediction report. In some other aspects, the prediction report may include a channel state information (CSI) prediction report. In some aspects, the prediction report may include the confidence level of the predicted output. The UE 120 may report the generated confidence level along with the predicted output in the prediction report. The UE 120 may transmit 316 the prediction report when the confidence level is above or equal to a confidence threshold, e.g., a predefined threshold. The UE 102 may transmit 316, to the NE 104, an indication indicating missing input elements of the first set of RSs when the confidence level is below the confidence threshold.

[0074] As an example, the UE 120 may skip transmitting the prediction report when the confidence level is below the confidence threshold. The UE 102 may skip reporting the prediction with the confidence level below' the predefined threshold. The confidence threshold may be indicated to the UE 102 by the NE 104. The UE 102may indicate which RS it has missed or will miss measuring. The NE 104 may schedule an additional RS, e.g., an aperiodic RS, as a replacement for the UE to measure. For the occasion in which the UE 102 skips transmitting the prediction report, or the UE reports a low confidence in the prediction report, the NE 104 may transmit 318 the additional RS, e.g., a dedicated RS. for the UE to measure the channel during the time slot when the prediction is not reported or is not reliable. The UE 102 may receive 318, from the NE 104, an additional RS, e.g., aperiodic RS, to replace the RS (as a substitute for the RS). FIGs. 2-5 illustrate the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs. FIGs. 6-7 show methods for implementing one or more aspects of FIGs. 2-5. In particular, FIG. 6 shows an implementation by the UE 102 of the one or more aspects of FIGs. 2-5. FIG. 7 shows an implementation by the network entity 104 of the one or more aspects of FIGs. 2-5.

[0075] FIG. 6 is a flowchart 600 of a method of wireless communication at the UE 102 for the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs according to an embodiment. With reference to FIGs. 1-5, the method may be performed by the UE 102.

[0076] In some embodiments, the UE 102 may transmit 602, to a network entity 104. a UE capability message indicating a UE capability for supporting a prediction report based on a measurement of a second set of RSs different from a first configured set of RSs. For example, referring to FIG. 3, the UE 102 may transmit 302, to the network entity 104, a UE capability message indicating a UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs.

[0077] The UE 102 receives 606, from the network entity 104, a configuration configuring a first set of RSs associated with an ML prediction module. For example, referring to FIG. 3, the UE 102 receives 306. from the NE 104, a configuration configuring the first set of RSs associated with an ML prediction module. The first set of RSs includes one or more RSs.

[0078] The UE 102 receives 610, from the network entity 104, a second set of RSs different from the first set of RSs. For example, referring to FIG. 3, the UE 102 receives 310, from the network entity 104, the second set of RSs different from the first set of RSs. In some aspects, the second set of RSs is a subset of the first set of RSs.

[0079] The UE 102 may generate 612 a plurality of non-measured elements based on an interpolation of a plurality of measured elements or an immediately prior measured element associated with a same corresponding RS. For example, referring to FIG. 3, the UE 102 may generate 312 the set of non-measured input elements. The multiple input elements may further include the set of non-measured input elements. In some examples, the set of non-measured input elements are generated based on an interpolation of the set of measured input elements or an immediately prior measured input item of a same corresponding RS.

[0080] The UE 102 may select 613 the ML prediction module from a plurality of ML prediction modules based on an input pattern of the plurality of input patterns. For example, referring to FIG. 3, the UE 102 may select 313 the ML prediction module from the multiple ML prediction modules based on an input pattern of the plurality of input patterns.

[0081] The UE 102 may generate 614 a confidence level based on the measurement of the second set of RSs. For example, referring to FIG. 3, the UE 102 may generate 314 a confidence level based on the measured second set of RSs.

[0082] The UE 102 transmits 616, to the network entity 104, a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module. For example, referring to FIG. 3, the UE 102 transmits 316, to the network entity 104, the prediction report based on the measurement of the second set of RSs being used in the input to the ML prediction module.

[0083] The UE 102 may receive 618, from the network entity 104, an additional RS. For example, referring to FIG. 3, the UE 102 may receive 318. from the NE 104, an additional RS, e.g., aperiodic RS, to replace the missed / skipped RS (as a substitute for the missed / skipped RS). FIG. 6 describes the method from a UE-side of a wireless communication link, whereas FIG. 7 describes the method from a network-side of the wireless communication link.

[0084] FIG. 7 is a flowchart 700 of a method of wireless communication at the network entity 104 for for the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs according to an embodiment. With reference to FIGs. 1-5, the method may be performed by one or more network entities 104, which may correspond to a base station or a unit of the base station, such as the RU 106, the DU 108, and / or the CU 110.

[0085] In some embodiments, the network entity 104 may receive 702, from the UE 102. a UE capability message indicating a UE capability for supporting a prediction report based on a measurement of a second set of RSs different from a first configured set of RSs. For example, referring to FIG. 3, the NE 104 may receive 302. from the UE 102, a UE capability message indicating the UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs.

[0086] The NE 104 transmits 706, to the UE 102, a configuration configuring a first set of RSs associated with an ML prediction module. For example, referring to FIG. 3. the NE 104 transmits 306, to the UE 102, the configuration configuring the first set of RSs associated with the ML prediction module.

[0087] The NE 104 transmits 710, to the UE 102, the first set of RSs. For example, referring to FIG. 3, the NE 104 transmits 310, to the UE 102. the first set of RSs based on the configuration.

[0088] The NE 104 receives 716, from the UE 102, a prediction report based on a measurement of a second set of RSs included in an input to the ML prediction module, the second set of RSs being different from the first set of RSs. For example, referring to FIG. 3, the NE104 receives 316, from the UE 102. the prediction report based on the measurement of the second set of RSs being used in the input to the ML prediction module.

[0089] The NE 104 may transmit 718, to the UE 102, an additional RS. For example, referring to FIG. 3, the NE 104 may transmit 318, to the UE 102, an additional RS, e.g., aperiodic RS, to replace the missed / skipped RS (as a substitute for the missed / skipped RS).

[0090] FIG. 8 illustrates a signaling diagram 800 of communications between the UE 102 and the network entity 104 for an indication indicating an RS configuration associated with an ML prediction module according to an embodiment. The network entity 104 may correspond to a base station or a unit of a base station, such as the RU 106, the DU 108, the CU 110, etc.

[0091] To use a downlink reference signal RS as an input to the ML prediction module, there are certain parameters associated wi th the RS configuration of the downlink RS, e.g., the transmission (Tx) power, periodicity, and / or beamforming configuration, which may match those used to generate training data for the ML prediction modules. In that regard, the NE 104 may change the configurations of thedownlink RS from time to time. For example, based on the load of UEs, the NE may change the number / directi ons / beam width of synchronization signal / PBCH blocks (SSBs). To save power, the NE may adjust the Tx power and array size associated with the downlink RS. Thus, it is important for the NE to inform the UE of the RS configurations associated with the ML prediction module, in order for the UE to determine whether the RS configuration satisfies the requirements of the ML prediction module that will be executed at the UE. Additionally, when the NW changes an RS configuration, it is important to notify the impacted UEs which run the relevant ML modules.

[0092] Referring to FIG. 8, the UE 102 may optionally transmit 802, to the network entity (104), a UE capability message indicating a UE capability for supporting an indication indicating an RS of a plurality of RSs associated with an ML prediction module being executed. The NE 104 may receive 802, from the UE 120, the UE capability message indicating a UE capability for supporting the indication indicating the RS of the plurality of RSs associated with the ML prediction module being executed. The UE may indicate the UE capability for supporting the indication indicating the RS associated with the ML prediction module the UE is running.

[0093] The UE 102 may transmit 804. to the network entity 104, at least one of: an ML prediction module update request, an ML prediction module inquiry, or an ML prediction module performance monitoring signaling. The NE 104 may receive 804, from the UE 102, at least one of: an ML prediction module update request, an ML prediction module inquiry, or an ML prediction module performance monitoring signaling. In some aspects, the UE 102 may indicate 804 the ML prediction module being executed at the UE. The UE 102 may send the ML prediction module inquiry to the NE 104. In the inquiry7, the UE 102 may report to the NE which ML module the UE is running. In some other aspects, the UE 102 may send the ML prediction module update request to the NE 104, which includes activation or deactivation of the ML prediction module. In some other aspects, the UE 102 may send the performance monitoring signaling of the ML prediction module to the NE 104, for example, the error report of the ML prediction module. The ML prediction module update request, the ML prediction module inquiry, and / or the ML prediction module performance monitoring signaling may trigger an RS configuration and / or an indication signaling of the RS configuration from the NE 104.

[0094] The UE 102 receives 806, from the network entity 104, a configuration configuring the plurality of RSs associated with the plurality of ML prediction modules. The NE 104 transmits 806, to the UE 102, a configuration configuring the plurality of RSs associated with the plurality of ML prediction modules. The configuration may be an RS configuration, which may be UE-specific (e.g., CSI- RSs) or cell-specific (e.g., SSBs). The RS configuration may correspond to a periodic RS, semi-persistent RS, and / or an aperiodic RS.

[0095] The UE 102 receives 808, from the network entity 104, an indication indicating the RS of the plurality of RSs associated with the ML prediction module of a plurality of ML prediction modules. The NE 104 transmits 808, to the UE 102, an indication indicating the RS of the plurality of RSs associated with the ML prediction module of a plurality of ML prediction modules. The ML prediction module is the one being executed at the UE 102. The indication may be received in at least one of: a UE-specific message, a group-specific message, or a cell-specific message. The indication may be included in the configuration configuring the plurality of RSs associated with the plurality of ML prediction modules. The indication may be included in the configuration signaling and / or a reconfiguration signaling.

[0096] In some examples, the indication indicates at least one of an activation of the ML prediction module or a switching to the ML prediction module. The configuration signaling and / or the reconfiguration signaling may include the indication, which may include activation signaling or module switching signaling. In some examples, the indication may include the deactivation signaling. The indication signaling may be triggered by the request of the ML prediction module update including activation and / or deactivation. The indication signaling may be triggered by the ML prediction module performance monitoring signaling.

[0097] In some examples, the indication may be binary, which indicates whether the RS configuration, e g., the current RS configuration, matches the requirement of the ML prediction module. The UE 102 may use the indication to select the appropriate ML prediction module. As an example, the NE may configure the UE with the plurality of ML prediction modules corresponding to a plurality of RS pattens (e.g., large NE array, medium NE array, small NE array). When the NE indicates that RS pattern X is currently deployed (X=l, 2, ...), the UE 102 may use the ML prediction module X corresponding to the RS pattern X (e.g., RS pattern 1 is for the RS transmitted usingthe large NE array. RS patern 2 is for the RS transmited using the medium NE array, RS patern 3 is for the RS transmited using the small NE array). The UE 102 may use the indication in the input to the ML prediction module or use the indication to adjust the input to ML prediction module.

[0098] In some examples, the indication indicates a transmission (Tx) power of the RS, and the input to the ML prediction module is based on the transmission power of the RS. The indication may include the Tx power of the RS. The UE 102 may use the Tx power in the input to the ML prediction module. The UE 102 may compute the offset of the Tx power of the RS from reference Tx power and adjust the measured RSRP of the RS. The adjusted RSRP may be used in the input to ML prediction module.

[0099] In some examples, the indication is included in a system information block (SIB). The SIB may be predefined or may be an on-demand transmission.

[0100] In some examples, the indication may be based on the UE blindly detecting a transmission patern. The transmission patern may be a combination of an RS sequence, a frequency / time location of the RS. If the detected RS matches a predefined transmission patern, then the RS matches the requirement for the input to the ML prediction module. As an example, the transmission patern is indicated by a scrambling sequence of a predefined anchor RS. The anchor RS may be a RS that UE is required to measure regardless of the transmission patern. For example, the transmission paterns may include: Transmission pateml: RS1+{RS x, y, z}; transmission patem2: RS1+{RS a, b, c}, where RSI is the anchor RS.

[0101] The UE 102 receives 810, from the network entity 104, the RS based on the indication. The NE 104 transmits 810, to the UE 102, the RS based on the indication. The UE performs 811 the measurement of the RS based on the configuration. The UE 102 selects 813 the ML prediction module from the plurality of ML prediction modules based on the indication. For example, the ML prediction module is selected according to a transmission patern of the RS.

[0102] The UE 102 transmits 816, to the network entity 104, a prediction report output from the ML prediction module based on the measurement of the RS (operation 811) being used as an input to the ML prediction module. The NE 104 receives 816, from the UE 102, the prediction report output from the ML prediction module based on the measurement of the RS being used as an input to the ML prediction module.

[0103] The UE 102 may receive 820, from the network entity 104, a second indication indicating a second RS of the plurality of RSs associated with the ML prediction module. The NE 104 may transmit 820, to the UE 102, the second indication indicating a second RS of the plurality of RSs associated with the ML prediction module. The second indication may be included in groupcast signaling. When the NE 104 changes the configuration, e.g., the RS configuration, the NE may inform the UE 102 of the change. The NE 104 may send dedicated signaling to the relevant UEs. The dedicated signaling may be a groupcast signaling to a group of the relevant UEs.

[0104] In other aspects, the NE 104 may broadcast the change of the RS configuration. As an example, the configuration signaling to inform the UE of the RS configuration change in a cell (e.g., cell A) may be sent via another cell (e.g., cell B). The cell A may be a cell of which the RS configuration is subject to change, and the cell B may be an anchor cell of which the RS configuration is stable. The change may be an adjustment of SSB patterns, and the indication of adjustment of SSB patterns in the cell A may be sent to the UE 102 via the cell B.

[0105] In this way. the accuracy and the reliability of the ML prediction module are improved, and the possibility of a beam failure is reduced. The proposed solutions may be applied to beam management procedures, e.g., in FR2 systems, such as beam monitoring, beam tracking, and beam failure recover}' .

[0106] For example, when the UE detects a beam failure event, the UE may predict the RSRPs of the candidate beams based on the ML prediction module. When the input pattern of the ML prediction module is incomplete, the UE may generate a plurality of non-measured elements based on the measurement of the second set of RSs, to predict the RSRPs of the candidate beam and provide information regarding the recommended replacement beam in a beam failure recovery request (BFRQ) to the network. The information may include the confidence level and / or the input pattern. The network may send a beam failure recovery response (BFRR). The NE 104 may determine whether the UE should perform further communication with the recommended beam, e.g., when the confidence level of the prediction is above a threshold or schedule an additional RS for the UE to measure to determine the replacement beam, e.g., when the confidence level is below the threshold. Different RS configurations may be indicated to the UE for the prediction of candidate beams.

[0107] FIGs. 1, 2, and 8 illustrate the indication indicating the RS associated with the ML prediction module. FIGs. 9-10 show methods for implementing one or more aspects of FIGs. 1 , 2, and 8. In particular, FIG. 9 shows an implementation by the UE 102 of the one or more aspects of FIGs. 2 and 8. FIG. 10 shows an implementation by the network entity 104 of the one or more aspects of FIGs. 1, 2. and 8.

[0108] FIG. 9 is a flowchart 900 of a method of wireless communication at the UE 102 for an indication indicating an RS associated with an ML prediction module according to an embodiment. In some embodiments, the UE 102 may transmit 902, to a network entity 104, a UE capability message indicating a UE capability’ for supporting the indication indicating the RS of a plurality' of RSs associated with the ML prediction module being executed. For example, referring to FIG. 8, the UE 102 may transmit 802, to the netyvork entity (104), a UE capability message indicating a UE capability for supporting an indication indicating an RS of a plurality of RSs associated with an ML prediction module being executed.

[0109] The UE 102 may transmit 904, to the network entity, at least one of: an ML prediction module update request, an ML prediction module inquiry', or an ML prediction module performance monitoring signaling. For example, referring to FIG. 8. the UE 102 may transmit 804. to the network entity 104, at least one of: an ML prediction module update request, an ML prediction module inquiry', or an ML prediction module performance monitoring signaling.

[0110] The UE 102 receives 906, from the network entity, a configuration configuring the plurality’ of reference signals (RSs) associated with a plurality of machine learning (ML) prediction modules. For example, referring to FIG. 8, the UE 102 receives 806, from the netyvork entity 104, a configuration configuring the plurality' of RSs associated yvith the plurality' of ML prediction modules.

[0111] The UE 102 receives 908, from the network entity, the indication indicating the RS of the plurality of RSs associated with the ML prediction module of the plurality of ML prediction modules, the ML prediction module being executed at the UE. For example, referring to FIG. 8, the UE 102 receives 808, from the netyvork entity' 104, an indication indicating the RS of the plurality of RSs associated with the ML prediction module of a plurality of ML prediction modules.

[0112] The UE 102 receives 910, from the network entity, the RS based on the indication. For example, referring to FIG. 8, the UE 102 receives 810, from the netyvork entity’ 104, the RS based on the indication.

[0113] The UE 102 may select 913 the ML prediction module from the plurality of ML prediction modules based on the indication. For example, referring to FIG. 8, the UE 102 selects 813 the ML prediction module from the plurality' of ML prediction modules based on the indication.

[0114] The UE 102 transmits 916. to the network entity, a prediction report output from the M based on a measurement of the RS being used as input to the ML prediction module. For example, referring to FIG. 8, the UE 102 transmits 816, to the network entity 104, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0115] The UE 102 receives 920, from the network entity, a second indication indicating a second RS of the plurality of RSs associated with the ML prediction module. For example, referring to FIG. 8, the UE 102 may receive 820, from the network entity 104, a second indication indicating a second RS of the plurality of RSs associated with the ML prediction module.

[0116] FIG. 10 is a flowchart 1000 of a method of wireless communication at a network entity for an indication indicating an RS associated yvith an ML prediction module according to an embodiment. In some embodiments, the NE 104 may receive 1002, from a UE 102, a UE capability message indicating a UE capability for supporting the indication indicating the RS of a plurality of RSs associated with the ML prediction module being executed. For example, referring to FIG. 8, the NE 104 may receive 802, from the UE 120, the UE capability message indicating a UE capability for supporting the indication indicating the RS of the plurality of RSs associated with the ML prediction module being executed.

[0117] The NE 104 may receive 1004, from the UE, at least one of: an ML prediction module update request, an ML prediction module inquiry, or an ML prediction module performance monitoring signaling. For example, referring to FIG. 8, the NE 104 may receive 804, from the UE 102, at least one of: an ML prediction module update request, an ML prediction module inquiry, or an ML prediction module performance monitoring signaling.

[0118] The NE 104 transmits 1006, to the UE 102, a configuration configuring the plurality of reference signals (RSs) associated with a plurality of machine learning (ML) prediction modules. For example, referring to FIG. 8, the NE 104 transmits 806, to the UE 102, a configuration configuring the plurality of RSs associated with the plurality of ML prediction modules.

[0119] The NE 104 transmits 1008, to the UE 102, the indication indicating the RS of the plurality of RSs associated with the ML prediction module of the plurality of ML prediction modules, the ML prediction module being executed at the UE. For example, referring to FIG. 8, The NE 104 transmits 808, to the UE 102. an indication indicating the RS of the plurality of RSs associated with the ML prediction module of a plurality of ML prediction modules.

[0120] The NE 104 transmits 1010, to the UE 102, the RS based on the indication. For example, referring to FIG. 8, the NE 104 transmits 810, to the UE 102, the RS based on the indication.

[0121] The NE 104 receives 1016, from the UE 102, a prediction report output from the ML prediction module based on a measurement of the RS being used as input to the ML prediction module. For example, referring to FIG. 8, the NE 104 receives 816, from the UE 102, the prediction report output from the ML prediction module based on the measurement of the RS being used as an input to the ML prediction module.

[0122] The NE 104 may transmit 1020, to the UE 102, a second indication indicating a second RS of the plurality of RSs associated with the ML prediction module. For example, referring to FIG. 8, the NE 104 may transmit 820, to the UE 102, the second indication indicating a second RS of the plurality of RSs associated with the ML prediction module.

[0123] A UE apparatus 1102, as described in FIG. 11 , may perform the method of flowcharts 600 and 900. The one or more network entities 104, as described in FIG. 12, may perform the method of flowcharts 700 and 1000.

[0124] FIG. 11 is a diagram 1100 illustrating an example of a hardware implementation for a UE apparatus 1102. The UE apparatus 1 102 may be the UE 102, a component of the UE 102, or may implement UE functionality. The UE apparatus 1102 may include an application processor 1106, which may have on-chip memory 1106'. In examples, the application processor 1106 may be coupled to a secure digital (SD) card 1108 and / or a display 1110. The application processor 1106 may also be coupled to a sensor(s) module 1112, a power supply 1114, an additional module of memory 1116, a camera 1118, and / or other related components.

[0125] The UE apparatus 1102 may further include a wireless baseband processor 1126, which may be referred to as a modem. The wireless baseband processor 1126 may have on-chip memory 1126'. Along with, and similar to, the application processor 1106, the wireless baseband processor 1126 may also be coupled to the sensor(s)module 1112, the power supply 1114, the additional module of memory 1116, the camera 1118, and / or other related components. The wireless baseband processor 1126 may be additionally coupled to one or more subscriber identity7module (SIM) card(s) 1120 and / or one or more transceivers 1130 (e.g.. wireless RF transceivers).

[0126] Within the one or more transceivers 1130, the UE apparatus 1102 may include a Bluetooth module 1132, a WLAN module 1 134, an SPS module 1136 (e.g., GNSS module), and / or a cellular module 1138. The Bluetooth module 1132, the WLAN module 1134, the SPS module 1136, and the cellular module 1138 may each include an on-chip transceiver (TRX), or in some cases, just a transmitter (TX) or just a receiver (RX). The Bluetooth module 1132, the WLAN module 1134, the SPS module 1136, and the cellular module 1138 may each include dedicated antennas and / or utilize antennas 1140 for communication with one or more other nodes. For example, the UE apparatus 1102 can communicate through the transceiver(s) 1130 via the antennas 1140 with another UE (e.g.. sidelink communication) and / or with a network entity 104 (e.g., uplmk / downlmk communication), where the network entity7104 may correspond to a base station or a unit of the base station, such as the RU 106, the DU 108, or the CU 110.

[0127] The wireless baseband processor 1126 and the application processor 1106 may each include a computer-readable medium / memory 1126’, 1106’, respectively. The additional module of memory 1116 may7also be considered a computer-readable medium / memory7. Each computer-readable medium / memory 1126', 1106', 1116 may be non-transitory. The wireless baseband processor 1126 and the application processor 1106 may each be responsible for general processing, including execution of software stored on the computer-readable medium / memory 1126', 1106', 1116. The software, when executed by the wireless baseband processor 1126 / application processor 1106, causes the wireless baseband processor 1126 / application processor 1106 to perform the various functions described herein. The computer-readable medium / memory may also be used for storing data that is manipulated by the wireless baseband processor 1126 / application processor 1106 when executing the software. The wireless baseband processor 1126 1 application processor 1106 may be a component of the UE 102. The UE apparatus 1102 may be a processor chip (e.g., modem and / or application) and include just the wireless baseband processor 1126 and / or the application processor 1106. In other examples, the UE apparatus1102 may be the entire UE 102 and include the additional modules of the apparatus 1102.

[0128] As discussed in FIG. 1 and implemented with respect to FIG. 6, the report component 140 is configured to receive, from a network entity, a configuration configuring a first set of RSs associated with an ML prediction module. The report component 140 is configured to receive, from the network entity, a second set of RSs different from the first set of RSs. The report component 140 is configured to transmit, to the network entity, a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module.

[0129] As discussed in FIG. 1 and implemented with respect to FIG. 9, the report component 140 is configured to receive, from the network entity, a configuration configuring a plurality of RSs associated with a plurality of ML prediction modules. The report component 140 is configured to receive, from the network entity, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality of ML prediction modules. The ML prediction module is being executed at the UE. The report component 140 is configured to receive, from the network entity, the RS based on the indication. The report component 140 is configured to transmit, to the network entity, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0130] The report component 140 may be within the application processor 1106 (e g., at 140a), the wireless baseband processor 1126 (e.g.. at 140b), or both the application processor 1106 and the wireless baseband processor 1 126. The report component 140a-140b may be one or more hardware components specifically configured to cany7out the stated processes / algorithm, implemented by one or more processors configured to perform the stated processes / algorithm, stored within a computer- readable medium for implementation by the one or more processors, or a combination thereof.

[0131] FIG. 12 is a diagram 1200 illustrating an example of a hardware implementation for one or more network entities 104. The one or more network entities 104 may be a base station, a component of a base station, or may implement base station functionality. The one or more network entities 104 may include, or may correspond to, at least one of the RU 106, the DU, 108, or the CU 110. The CU 110 may include a CU processor 1246, which may have on-chip memory 1246'. In someaspects, the CU 110 may further include an additional module of memory 1256 and / or a communications interface 1248, both of which may be coupled to the CU processor 1246. The CU 110 can communicate with the DU 108 through a midhaul link 162, such as an Fl interface between the communications interface 1248 of the CU 110 and a communications interface 1228 of the DU 108.

[0132] The DU 108 may include a DU processor 1226, which may have on-chip memory 1226'. In some aspects, the DU 108 may further include an additional module of memory 1236 and / or the communications interface 1228, both of which may be coupled to the DU processor 1226. The DU 108 can communicate with the RU 106 through a fronthaul link 160 between the communications interface 1228 of the DU 108 and a communications interface 1208 of the RU 106.

[0133] The RU 106 may include an RU processor 1206, which may have on-chip memory 1206'. In some aspects, the RU 106 may further include an additional module of memory 1216, the communications interface 1208. and one or more transceivers 1230, all of which may be coupled to the RU processor 1206. The RU 106 may further include antennas 1240, which may be coupled to the one or more transceivers 1230, such that the RU 106 can communicate through the one or more transceivers 1230 via the antennas 1240 with the UE 102.

[0134] The on-chip memory 1206', 1226', 1246' and the additional modules of memory 1216, 1236, 1256 may each be considered a computer-readable medium / memory. Each computer-readable medium / memory7may be non-transitory. Each of the processors 1206, 1226. 1246 is responsible for general processing, including execution of software stored on the computer-readable medium / memory. The software, when executed by the corresponding processor(s) 1206, 1226, 1246 causes the processor(s) 1206, 1226, 1246 to perform the various functions described herein. The computer-readable medium / memory may also be used for storing data that is manipulated by the processor(s) 1206. 1226. 1246 when executing the software. In examples, the configuration component 150 may sit at any of the one or more network entities 104, such as at the CU 110; both the CU 110 and the DU 108; each of the CU 110, the DU 108, and the RU 106; the DU 108; both the DU 108 and the RU 106; or the RU 106.

[0135] As discussed in FIG. 1 and implemented with respect to FIG. 7, the configuration component 150 is configured to transmit, to a UE, a configuration configuring a first set of RSs associated with an ML prediction module. Theconfiguration component 150 is configured to transmit, to the UE, the first set of RSs. The configuration component 150 is configured to receive, from the UE, a prediction report based on a measurement of a second set of RSs included in an input to the ML prediction module, the second set of RSs being different from the first set of RSs.

[0136] As discussed in FIG. 1 and implemented with respect to FIG. 10, the configuration component 150 is configured to transmit, to a UE, a configuration configuring a plurality of RSs associated with a plurality' of ML prediction modules. The configuration component 150 is configured to transmit, to the UE, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality of ML prediction modules. The ML prediction module is being executed at the UE. The configuration component 150 is configured to transmit, to the UE. the RS based on the indication. The configuration component 150 is configured to receive, from the UE, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0137] The configuration component 150 may be within one or more processors of the one or more network entities 104, such as the RU processor 1206 (e.g., at 150a), the DU processor 1226 (e.g., at 150b), and / or the CU processor 1246 (e.g., at 150c). The configuration component 150a- 150c may be one or more hardware components specifically configured to cany' out the stated processes / algorithm, implemented by one or more processors 1206, 1226, 1246 configured to perform the stated processes / algorithm. stored within a computer-readable medium for implementation by the one or more processors 1206, 1226, 1246, or a combination thereof.

[0138] The specific order or hierarchy of blocks in the processes and flowcharts disclosed herein is an illustration of example approaches. Hence, the specific order or hierarchy of blocks in the processes and flowcharts may be rearranged. Some blocks may also be combined or deleted. Dashed lines may indicate optional elements of the diagrams. The accompanying method claims present elements of the various blocks in an example order and are not limited to the specific order or hierarchy presented in the claims, processes, and flowcharts.

[0139] The detailed description set forth herein describes various configurations in connection with the drawings and does not represent the only configurations in which the concepts described herein may be practiced. The detailed descriptionincludes specific details for the purpose of providing a thorough explanation of various concepts. However, these concepts may be practiced without these specific details. In some instances, well known structures and components are shown in block diagram form in order to avoid obscuring such concepts.

[0140] Aspects of wireless communication systems, such as telecommunication systems, are presented with reference to various apparatuses and methods. These apparatuses and methods are described in the following detailed description and are illustrated in the accompanying drawings by various blocks, components, circuits, processes, call flows, systems, algorithms, etc. (collectively referred to as ■‘elements”). These elements may be implemented using electronic hardware, computer software, or combinations thereof. Whether such elements are implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system.

[0141] An element, or any portion of an element, or any combination of elements may be implemented as a '‘processing system” that includes one or more processors. Examples of processors include microprocessors, microcontrollers, graphics processing units (GPUs), central processing units (CPUs), application processors, digital signal processors (DSPs), reduced instruction set computing (RISC) processors, systems-on-chip (SoC), baseband processors, field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other similar hardw are configured to perform the various functionality described throughout this disclosure. One or more processors in the processing system may execute software, which may be referred to as softw are, firmware, middleware, microcode, hardware description language, or otherwise. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software components, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, or any combination thereof.

[0142] If the functionality described herein is implemented in softw are, the functions may be stored on, or encoded as. one or more instructions or code on a computer- readable medium, such as a non-transitory computer-readable storage medium. Computer-readable media includes computer storage media and can include a random-access memory' (RAM), a read-only memory (ROM), an electricallyerasable programmable ROM (EEPROM), optical disk storage, magnetic disk storage, other magnetic storage devices, combinations of these types of computer- readable media, or any other medium that can be used to store computer executable code in the form of instructions or data structures that can be accessed by a computer. Storage media may be any available media that can be accessed by a computer.

[0143] Aspects, implementations, and / or use cases described herein may be implemented across many differing platform ty pes, devices, systems, shapes, sizes, and packaging arrangements. For example, the aspects, implementations, and / or use cases may come about via integrated chip implementations and other non-modulecomponent based devices, such as end-user devices, vehicles, communication devices, computing devices, industrial equipment, retail / purchasing devices, medical devices, artificial intelligence (Al)-enabled devices, machine learning (ML)-enabled devices, etc. The aspects, implementations, and / or use cases may range from chiplevel or modular components to non-modular or non-chip-level implementations, and further to aggregate, distributed, or original equipment manufacturer (OEM) devices or systems incorporating one or more techniques described herein.

[0144] Devices incorporating the aspects and features described herein may also include additional components and features for the implementation and practice of the claimed and described aspects and features. For example, transmission and reception of wireless signals necessarily includes a number of components for analog and digital purposes, such as hardware components, antennas, RF-chains, power amplifiers, modulators, buffers, processor(s). interleavers, adders / summers, etc. Techniques described herein may be practiced in a wide variety of devices, chip-level components, systems, distributed arrangements, aggregated or disaggregated components, end-user devices, etc., of varying configurations.

[0145] The description herein is provided to enable a person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not limited to the aspects described herein but are to be interpreted in view of the full scope of the present disclosure consistent with the language of the claims.

[0146] Reference to an element in the singular does not mean “one and only one” unless specifically stated, but rather “one or more.” Terms such as “if,” “when,” and“while’7do not imply an immediate temporal relationship or reaction. That is, these phrases, e.g., “when,” do not imply an immediate action in response to or during the occurrence of an action, but simply imply that if a condition is met then an action will occur, but without requiring a specific or immediate time constraint for the action to occur. The terms “may”, “might”, and “can”, as used in this disclosure, often carry certain connotations. For example, “may” refers to a permissible feature that may or may not occur, “might” refers to a feature that probably occurs, and “can” refers to a capability (e.g., capable of). The phrase “For example” often carries a similar connotation to “may” and, therefore, “may” is sometimes excluded from sentences that include “for example” or other similar phrases.

[0147] Unless specifically stated otherwise, the term “some” refers to one or more. Combinations such as “at least one of A, B, or C” or “one or more of A, B, or C” include any combination of A, B. and / or C, such as A and B, A and C, B and C. or A and B and C, and may include multiples of A, multiples of B, and / or multiples of C, or may include A only, B only, or C only. Sets should be interpreted as a set of elements where the elements number one or more. Terms or articles such as “a”, “an”, and / or “the” may refer to one of an item, feature, element, etc., that the term or article precedes, or may refer to more than one of said item, feature, element, etc. that the term or article precedes. For example, the recitation “a widget” does not preclude reference to multiples of said widget, as “multiple widgets” necessarily includes “a widget”. Hence, the recitation “a widget” may be interpreted as “at least one widget” or. similarly, interpreted as “one or more widgets”.

[0148] Unless otherwise specifically indicated, ordinal terms such as “first” and “second” do not necessarily imply an order in time, sequence, numerical value, etc., but are used to distinguish between different instances of a term or phrase that follows each ordinal term.

[0149] Reference numbers, as used in the specification and figures, are sometimes cross-referenced among drawings to denote same or similar features. A feature that is exactly the same in multiple drawings may be labeled with the same reference number in the multiple drawings. A feature that is similar among the multiple drawings, but not exactly the same, may be labeled with reference numbers that have different leading numbers but have one or more of the same trailing numbers (e.g., 206, 306, 406, etc., may refer to similar features in the drawings). Hence, like numbers may refer to like actions.

[0150] Structural and functional equivalents to elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are encompassed by the claims. The w ords ‘‘module,'’ “mechanism,’" “element,” “device,” and the like may not be a substitute for the word “means.” As such, no claim element is to be construed as a means plus function unless the element is expressly recited using the phrase “means for.” As used herein, the phrase “based on” shall not be construed as a reference to a closed set of information, one or more conditions, one or more factors, or the like. In other words, the phrase “based on A”, where “A” may be information, a condition, a factor, or the like, shall be construed as “based at least on A” unless specifically recited differently.

[0151] The following examples are illustrative only and may be combined with other examples or teachings described herein, without limitation.

[0152] Example 1 is a method of wireless communication as described herein at a UE. including: receiving, from a network entity, a configuration configuring a first set of reference signals (RSs) associated with a machine learning (ML) prediction module; receiving, from the network entity, a second set of RSs different from the first set of RSs; and transmitting, to the network entity, a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module.

[0153] Example 2 may be combined with Example 1 and includes that transmitting, to the network entity, a UE capability message indicating a UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs.

[0154] Example 3 may be combined with any of Examples 1-2 and includes that a first number of the first set of RSs is greater than a second number of the second set of RSs, and wherein the input includes input elements associated with the first set of RSs, wherein the input elements include a plurality of measured elements associated with the second set of RSs and a plurality of non-measured elements associated with a difference between the first and second set of RSs.

[0155] Example 4 may be combined with Example 3 and includes generating the plurality of non-measured elements based on at least one of: an interpolation of the plurality of measured elements or an immediately prior measured element associated with a same corresponding RS.

[0156] Example 5 may be combined with Example 3 and includes each of the plurality of non-measured elements includes a predefined reserved index.

[0157] Example 6 may be combined with Example 3 and includes each of the plurality of measured elements includes at least one of: a time stamp, an identification (ID) of a corresponding RS, or a positional encoding sequence.

[0158] Example 7 may be combined with any of Examples 1-6 and includes that the configuration indicates a minimum number for the second set of RSs within a predefined time period.

[0159] Example 8 may be combined with any of Examples 1-7 and includes that the configuration further configures a plurality of ML prediction modules, the ML prediction module being included in the plurality of ML prediction modules.

[0160] Example 9 may be combined with Example 8 and includes the plurality of ML prediction modules corresponds to a plurality of input patterns, selecting the ML prediction module from the plurality of ML prediction modules based on an input pattern of the plurality of input patterns.

[0161] Example 10 may be combined with any of Examples 1-9 and includes that generating a confidence level based on the measurement of the second set of RSs; where the transmitting the prediction report including: transmitting, to the network entity, the prediction report including the confidence level.

[0162] Example 11 may be combined with Example 10 and includes that the confidence level is generated based on at least one: a variance of a prediction result, a success probability of a prediction result, a probability distribution that a predicted beam is a qualified candidate beam, or a function of at least one of: the second number of the second set of RSs, identifiers (IDs) of the second set of RSs, or a periodicity of the measurement of the second set of RSs.

[0163] Example 12 may be combined with Example 10 and includes that the transmitting the prediction report includes: transmitting the prediction report when the confidence level is above or equal to a confidence threshold; and skipping transmitting the prediction report when the confidence level is below the confidence threshold.

[0164] Example 13 may be combined with Example 12 and includes that transmitting, to the network entity, an indication that the confidence level is below the confidence threshold; and receiving, from the network entity, an additional RS to be included in the second set of RSs.

[0165] Example 14 is a method of wireless communication as described herein at a network entity, including: transmitting, to a UE, a configuration configuring a first set of reference signals (RSs) associated with a machine learning (ML) prediction module; transmitting, to the UE, the first set of RSs; and receiving, from the UE, a prediction report based on a measurement of a second set of RSs included in an input to the ML prediction module, the second set of RSs being different from the first set of RSs.

[0166] Example 15 may be combined with Example 14 and includes that receiving, from the UE, a UE capability’ message indicating a UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs.

[0167] Example 16 may be combined with any of Examples 14-15 and includes a first number of the first set of RSs is greater than a second number of the second set of RSs, and wherein the input includes input elements associated with the first set of RSs, wherein the input elements include a plurality of measured elements associated with the second set of RSs and a plurality of non-measured elements associated with a difference between the first and second set of RSs.

[0168] Example 17 may be combined with Example 16 and includes that the plurality of non-measured elements are generated based on at least one of: an interpolation of the plurality of measured elements or an immediately prior measured element associated with a same corresponding RS.

[0169] Example 18 may be combined with Example 16 and includes that each of the plurality of non-measured elements includes a predefined reserved index.

[0170] Example 19 may be combined with Example 16 and includes that each of the plurality of measured elements includes at least one of: a time stamp, an identification (ID) of a corresponding RS, or a positional encoding sequence.

[0171] Example 20 may be combined with any of Examples 14-19 and includes the configuration indicates a minimum number for the second set of RSs within a predefined time period.

[0172] Example 21 may be combined with any of Examples 14-20 and includes the configuration further configures a plurality of ML prediction modules, the ML prediction module being included in the plurality of ML prediction modules.

[0173] Example 22 may be combined with Example 21 and includes that each the plurality of ML prediction modules correspond to a plurality of input patterns,wherein the ML prediction module is selected from the plurality of ML prediction modules based on an input pattern of the plurality of input patterns.

[0174] Example 23 may be combined with any of Examples 14-22 and includes the receiving the prediction report includes: receiving, from the UE the prediction report including a confidence level associated with an output of the ML prediction module.

[0175] Example 24 may be combined with Example 23 and includes that the confidence level is generated based on at least one: a variance of a prediction result, a success probability of a prediction result, a probability distribution that a predicted beam is a qualified candidate beam, or a function of at least one of the second number of the second set of RSs, identifiers (IDs) of the second set of RSs, or a periodicity of the measurement of the second set of RSs.

[0176] Example 25 may be combined with Example 23 and includes that the receiving the prediction report includes: receiving the prediction report when the confidence level is above or equal to a confidence threshold.

[0177] Example 26 may be combined with Example 25 and includes that receiving, from the UE, an indication that the confidence level is below the confidence threshold; and transmitting, to the UE, an additional RS to be included in the second set of RSs.

[0178] Example 27 is a method of wireless communication as described herein at UE, including: receiving, from a network entity, a configuration configuring a plurality of reference signals (RSs) associated with a plurality of machine learning (ML) prediction modules; receiving, from the network entity, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality of ML prediction modules, the ML prediction module being executed at the UE; receiving, from the network entity, the RS based on the indication; and transmitting, to the network entity, a prediction report output from the ML prediction module based on a measurement of the RS being used as an input to the ML prediction module.

[0179] Example 28 may be combined with Example 27 and includes that transmitting, to the network entity, a UE capability message indicating a UE capability for supporting the indication indicating the RS of the plurality of RSs associated with the ML prediction module being executed.

[0180] Example 29 may be combined with any of Examples 27-28 and includes the indication is received in at least one of: a UE-specific message or a cell-specific message.

[0181] Example 30 may be combined with any of Examples 27-29 and includes the indication is included in the configuration configuring the plurality of RSs associated with the plurality of ML prediction modules.

[0182] Example 31 may be combined with any of Examples 27-30 and includes the indication indicates at least one of: an activation of the ML prediction module or a switching to the ML prediction module.

[0183] Example 32 may be combined with any of Examples 27-31 and includes transmitting, to the network entity, at least one of: an ML prediction module update request, an ML prediction module inquiry, or an ML prediction module performance monitoring signaling.

[0184] Example 33 may be combined with Example 32 and includes that the at least one of: the ML prediction module update request, the ML prediction module inquiry, or the ML prediction module performance monitoring signaling indicates the ML prediction module being executed at the UE (102).

[0185] Example 34 may be combined with any of Examples 27-33 and includes selecting the ML prediction module from the plurality of ML prediction modules based on the indication.

[0186] Example 35 may be combined with Example 34 and includes that the ML prediction module is selected according to a transmission pattern of the RS.

[0187] Example 36 may be combined with any of Examples 27-35 and includes the indication indicates a transmission power of the RS, and wherein the input to the ML prediction module is based on the transmission power of the RS.

[0188] Example 37 may be combined with any of Examples 27-36 and includes the indication is included in a system information block (SIB).

[0189] Example 38 may be combined with any of Examples 27-37 and includes receiving, from the network entity, a second indication indicating a second RS of the plurality of RSs associated with the ML prediction module, wherein the second indication is included in groupcast signaling.

[0190] Example 39 is a method of wireless communication as described herein at a network entity, including: transmitting, to a user equipment (UE), a configuration configuring a plurality7of reference signals (RSs) associated with a plurality7ofmachine learning (ML) prediction modules; transmitting, to the UE, an indication indicating an RS of the plurality of RSs associated with an ML prediction module of the plurality' of ML prediction modules, the ML prediction module being executed at the UE; transmitting, to the UE, the RS based on the indication; and receiving, from the UE. a prediction report output from the ML prediction module based on a measurement of the RS being used as input to the ML prediction module.

[0191] Example 40 may be combined with Example 39 and includes that receiving, from the UE, a UE capability message indicating a UE capability for supporting the indication indicating the RS of the plurality of RSs associated with the ML prediction module being executed.

[0192] Example 41 may be combined with any of Examples 39-40 and includes the indication is transmitted in at least one of: a UE-specific message or a cell-specific message.

[0193] Example 42 may be combined with any of Examples 39-41 and includes the indication is included in the configuration configuring the plurality of RSs associated with the plurality of ML prediction modules.

[0194] Example 43 may be combined with any of Examples 39-42 and includes 39-42, wherein the indication indicates at least one of an activation of the ML prediction module or a switching to the ML prediction module.

[0195] Example 44 may be combined with any of Examples 39-43 and includes: receiving, from the UE, at least one of: an ML prediction module update request, an ML prediction module inquiry, or an ML prediction module performance monitoring signaling.

[0196] Example 45 may be combined with Example 44 and includes that the at least one of: the ML prediction module update request, the ML prediction module inquiry, or the ML prediction module performance monitoring signaling indicates the ML prediction module being executed at the UE.

[0197] Example 46 may be combined with any of Examples 39-45 and includes: the indication indicates a transmission power of the RS, and wherein the input to the ML prediction module is based on the transmission power of the RS.

[0198] Example 47 may be combined with any of Examples 39-46 and includes: the indication is included in a system information block (SIB).

[0199] Example 48 may be combined with any of Examples 39-47 and includes: transmitting, to the UE, a second indication indicating a second RS of the pluralityof RSs associated with the ML prediction module, wherein the second indication is included in groupcast signaling.

[0200] Example 49 is an apparatus for wireless communication including a transceiver, a memory and a processor coupled to the memory and the transceiver, the apparatus being configured to implement a method as in any of examples 1-48.

[0201] Example 50 is an apparatus for wireless communication including means for implementing a method as in any of examples 1-48.

[0202] Example 51 is a non-transitory computer-readable medium storing computer executable code, the code when executed by a processor causes the processor to implement a method as in any of examples 1-48.

Claims

CLAIMSWHAT IS CLAIMED IS:

1. A method of wireless communication at a user equipment, UE, (102), comprising: receiving (306), from a network entity (104), a configuration configuring a first set of reference signals, RSs, associated with a machine learning, ML, prediction module; receiving (310), from the network entity (104), a second set of RSs different from the first set of RSs; and transmitting (316), to the network entity (104), a prediction report based on a measurement of the second set of RSs being used in an input to the ML prediction module.

2. The method of claim 1, further comprising: transmitting (302), to the network entity (104), a UE capability message indicating a UE capability for supporting the prediction report based on the measurement of the second set of RSs different from the first configured set of RSs.

3. The method of any of claims 1-2, wherein a first number of the first set of RSs is greater than a second number of the second set of RSs, and wherein the input includes input elements associated with the first set of RSs, wherein the input elements include a plurality of measured elements associated with the second set of RSs and a plurality of non-measured elements associated with a difference between the first and second set of RSs.

4. The method of claim 3, further comprising: generating (312) the plurality’ of non-measured elements based on at least one of: an interpolation of the plurality of measured elements or an immediately prior measured element associated with a same corresponding RS.

5. The method of claim 3, wherein each of the plurality of non-measured elements includes a predefined reserved index.

6. The method of claim 3. wherein each of the plurality of measured elements includes at least one of: a time stamp, an identification, ID, of a corresponding RS, or a positional encoding sequence.

7. The method of any of claims 1-6, wherein the configuration indicates a minimum number for the second set of RSs within a predefined time period.

8. The method of any of claims 1-7, wherein the configuration further configures a plurality of ML prediction modules, the ML prediction module being included in the plurality of ML prediction modules.

9. The method of claim 8, wherein the plurality of ML prediction modules corresponds to a plurality of input patterns, the method further comprising: selecting (313) the ML prediction module from the plurality of ML prediction modules based on an input pattern of the plurality of input patterns.

10. The method of any of claims 1-9, further comprising: generating (314) a confidence level based on the measurement of the second set of RSs; wherein the transmitting (316) the prediction report comprises: transmitting (316), to the network entity (104), the prediction report including the confidence level.

11. The method of claim 10, wherein the confidence level is generated based on at least one: a variance of a prediction result, a success probability of a prediction result. a probability distribution that a predicted beam is a qualified candidate beam, or a function of at least one of: the second number of the second set of RSs, identifiers, IDs. of the second set of RSs, or a periodicity of the measurement of the second set of RSs.

12. The method of claim 10. wherein the transmitting (316) the prediction report comprises: transmitting (316) the prediction report when the confidence level is above or equal to a confidence threshold; and skipping transmitting the prediction report when the confidence level is below the confidence threshold.

13. The method of claim 12, further comprising: transmitting (316), to the network entity (104), an indication that the confidence level is below the confidence threshold; and receiving (318), from the network entity (104), an additional RS to be included in the second set of RSs.

14. A method of wireless communication at a network entity (104). comprising: transmitting (306), to a user equipment, UE, (102), a configuration configuring a first set of reference signals, RSs, associated with a machine learning, ML, prediction module; transmitting (310), to the UE (102), the first set of RSs; and receiving (316), from the UE (102), a prediction report based on a measurement of a second set of RSs included in an input to the ML prediction module, the second set of RSs being different from the first set of RSs.

15. An apparatus for wireless communication comprising a transceiver, a memory, and a processor coupled to the memory and the transceiver, the apparatus being configured to implement a method as in any of claims 1-14.

Images