Autonomous machine learning prediction module configuration update in wireless communication systems

Abstract

This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for updating a ML prediction module configuration in wireless communication systems. A UE (102) receives (204), from a network entity (104), a first indication of a first configuration for CSI prediction. The first configuration is associated with a first set of reference signals. The UE (102) receives (206), from the network entity (104), a downlink reference signal. The UE (102) transmits (208), to the network entity (104), a report for channel feedback based on the downlink reference signal. The report is associated with updating (211) to a second configuration for CSI prediction different from the first configuration for CSI prediction. The second configuration is associated with a second set of reference signals.

Description

AUTONOMOUS MACHINE LEARNING PREDICTION MODULECONFIGURATION UPDATE IN WIRELESS COMMUNICATION SYSTEMSCROSS REFERENCE TO RELATED APPLICATION(S)

[0001] This application claims the benefit of and priority to U.S. Provisional Application SerialNo. 63 / 664,095, entitled “Autonomous Machine Learning Prediction Module Configuration Update in Wireless Communication Systems” and filed on June 25, 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 machine learning (ML) prediction module configuration updates in wireless communication systems.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 sen-ices (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) prediction module configuration depends on wireless channel properties. The ML prediction module configuration should be updated when the wireless channel property changes.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 allcontemplated aspects. This summary' neither identifies key or critical elements of all aspects nor delineates the scope of any or all aspects. Its sole purpose is 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 network entity , such as a base station or a unit of a base station, may communicate with a user equipment (UE), where the UE operates a machine learning (ML) module that predicts channel state information (CSI). The ML prediction module configuration may be based on the wireless channel property, which the UE and / or the network entity determines at least in part by UE mobility7and / or the physical environment. In instances where the wireless channel property' changes (e.g., increased / improved quality or decreased / degraded quality), the ML prediction module configuration is updated accordingly’. When the ML prediction module configuration is not updated in a timely manner, an increase in prediction errors may occur, which may further lead to performance degradation in terms of data rate and transmission latency.

[0007] In some instances, the network entity monitors the UE performance, identifies the change in the wireless channel conditions based on channel feedback from the UE, and sends a ML module configuration update to the UE. In some other instances, the UE monitors and identifies the change in the wireless channel conditions, requests the network entity to update the ML module configuration, and receives the ML configuration update from the network entity accordingly. The UE may send the request for an updated ML module configuration in an effort to reduce prediction errors.

[0008] Aspects of the present disclosure address the above-noted and other deficiencies by implementing a ML configuration / module update based on a report transmitted by the UE to the network entity. In other words, the UE provides the network entity with a report of the channel conditions which may autonomously trigger an update to the ML module configuration. As such, signaling overhead and latency7may be reduced to update the ML module configuration at the UE. In some instances, an optimal configuration of a time domain prediction ML module may be based on the rate of wireless channel changes (or how fast the wireless channel is varying), which may be due to UE mobility7and may be characterized by a time domain channel property (TDCP). The ML configuration may indicate a first ML module from a set of candidate ML modules. The ML configuration may be associated with a reference signal resource configuration that the network entity provides to the UE. The UE measures the reference signal transmitted based on thereference signal resource configuration and uses the measurements as input for the current ML module. The ML configuration may indicate the identifier, functionality, content, and other parameters (e.g., periodicity) of the prediction report, which may be based on how far into the future the UE can accurately predict CSI. The UE sends the prediction report based on the output of the current ML module. In some instances, the UE is configured to send TDCP reports to the network entity. When the UE sends a TDCP report that satisfies a predefined condition and triggers a ML configuration update, the UE changes the ML configuration autonomously based on a pre-configuration or pre-defined rule. The network entity may update the ML configuration accordingly as well based on the TDCP report. In some aspects, the TDCP report may indicate the updated ML configuration selected or utilized by the UE.

[0009] According to some aspects, the UE receives, from the network entity, a first indication of a first configuration for CSI prediction. The first configuration is associated with a first set of reference signals. The UE receives, from the network entity, a downlink reference signal. The UE transmits, to the network entity, a report for channel feedback based on the downlink reference signal. The report is associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction. The second configuration is associated with a second set of reference signals.

[0010] According to some aspects, the network entity transmits, to a UE, a first indication of a first configuration for CSI prediction. The first configuration is associated with a first set of reference signals. The netw ork entity transmits, to the UE, a downlink reference signal. The network entity receives, from the UE, a report for channel feedback based on the downlink reference signal. The report is associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction. The second configuration is associated with a second set of reference signals.BRIEF DESCRIPTION OF THE DRAWINGS

[0011] 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.

[0012] FIG. 2 is a signaling diagram illustrating a report that triggers an autonomous machine learning (ML) configuration update according to an embodiment.

[0013] FIG. 3 is a signaling diagram illustrating a time domain channel property (TDCP) report that triggers an autonomous ML configuration update at the UE according to an embodiment.

[0014] FIG. 4 is a flowchart of a method of wireless communication at a UE according to an embodiment.

[0015] FIG. 5 is a flowchart of a method of wireless communication at a UE including determining whether a report triggers an update of the configuration according to an embodiment.

[0016] FIG. 6 is a flowchart of a method of wireless communication at a network entity according to an embodiment.

[0017] FIG. 7 is a flowchart of a method of wireless communication at a network entity including determining whether a report triggers an update of the configuration according to an embodiment.

[0018] FIGs. 8A-8B are timelines of updating the configuration according to embodiments.

[0019] FIG. 9 is a diagram illustrating a hardware implementation for an example UE apparatus according to some embodiments.

[0020] FIG. 10 is a diagram illustrating a hardware implementation for one or more example network entities according to some embodiments.DETAILED DESCRIPTION

[0021] FIG. 1 illustrates a diagram 100 of a wireless communications system associated with a plurality7of 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 110 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 virtualradio 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 disaggregated units of the base station, such as the RU 106 or the DU 108), may be referred to as a transmission reception point (TRP).

[0022] 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 phy sical 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 sen e the UEs 102, such as by intra-cell and / or inter-cell access links between the UEs 102 and the RUs 106 / base stations 104.

[0023] 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) 112 of the base station 104d associated with the cell 190d. The BBU 112 includes a DU 108 and a CU 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.

[0024] 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 logical node 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.

[0025] 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 intercell 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.

[0026] 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.’’

[0027] 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 from the UE 102d based on the Uuinterface associated with the access link betw een the UE 102d and the base station 104d / RU106d.

[0028] 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 carriers. 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 earners 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 secondary cell (SCell).

[0029] 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 signal from 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.

[0030] 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 communicationbeams 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.

[0031] 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 sendee 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 generationradio 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 base station 104e can be a master node and the base station / RU 160a can be a secondary node.

[0032] Still referring to FIG. 1, any of the UEs 102 may include a reporting component 140 configured to receive, from the network entity 104, a first indication of a first configuration for channel state information (CSI) prediction, the first configuration being associated with a first set of reference signals; receive, from the network entity 104, a downlink reference signal; and transmit, to the network entity 104, a report for channel feedback based on the downlink reference signal, the report being associated with updating to a secondconfiguration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals.

[0033] The base stations 104 or anetwork entity of the base stations 104 may include a machine learning (ML) configuration component 150 configured to transmit, to a UE, a first indication of a first configuration for CSI prediction, the first configuration being associated with a first set of reference signals; transmit, to the UE 102, a downlink reference signal; and receive, from the UE 102. a report for channel feedback based on the downlink reference signal, the report being associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals.

[0034] 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.

[0035] FIG. 2 is a signaling diagram 200 illustrating communication between a UE 102 and a network entity 104 for a report that triggers an autonomous ML configuration update. 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.

[0036] The UE 102 may transmit 202 a UE capability regarding the capability of the UE 102 to support report triggered ML module configuration update. The UE capability may indicate whether the UE 102 supports autonomous updating of the ML configuration. The UE capability may indicate the types of reports the UE 102 supports that may trigger the ML module configuration update. In some instances, the UE capability may indicate the types of events that the UE 102 supports as triggering events. The UE capability may indicate which ML module configurations are supported by the UE 102. For example, the UE 102 may support a set of ML module configurations and may autonomously switch between any configurations within the set of ML module configurations based on a report for channel feedback (e.g.. TDCP report).

[0037] The network entity 104 transmits 204a, to the UE 102, a CSI report configuration for configuring the report and a reference signal resource set for channel measurement. In some aspects, the CSI report configuration may configure the UE to report one or more of the following channel metrics: channel quality indicator (CQI), rank indicator (RI),precoding matrix indicator (PMI), reference signal received power (RSRP), reference signal received quality (RSRQ), received signal strength indicator (RS SI), signal to noise ratio (SNR), and / or signal to interference and noise ratio (SINR). The report may trigger an ML configuration update. The network entity 104 may transmit the CSI report configuration via RRC signaling (e.g., RRCReconfiguration or CSI-ReportConfig). The network entity may provide some of the configurations or update some of the configurations by Medium Access Control (MAC) Control Element (CE) (e.g., MAC CE triggering the (semi-persistent) CSI report), or Downlink Control Information (DCI) (e.g., different triggering states for the DCI triggering the (aperiodic) CSI report may correspond to different configurations).

[0038] The network entity 104 transmits 204b, to the UE 102, a first indication of a first ML configuration. For example, the first ML configuration may include a first configuration for CSI prediction, where the first configuration is associated with a first set of reference signals. In some aspects, the CSI report configuration and the first ML configuration can be sent, by the network entity 104. in a same message (e.g., RRC message).

[0039] The UE 102 may execute an ML module that predicts CSI. In some instances, the UE 102 may be configured with an ML module to use current and / or past channel measurements to predict a future CSI. The UE 102 may report the predicted CSI as uplink control information to the netw ork entity 104. The predicted CSI may include future CQI, RI, PMI, RSRP, RSRQ, RSSI, SNR. and / or SINR for layer 1 (LI) or layer 3 (L3). In some instances, the prediction may be beam or cell specific, such that a ML prediction module may predict multiple RSRPs that each correspond to a beam or transmission configuration indicator (TCI) state at a future time.

[0040] The ML module configuration may include a selection of ML prediction modules, a configuration of reference signals for measurement by the UE 102, and / or a configuration of reports based on an output of the ML module. The configuration of the reference signals that the UE 102 measures may indicate that the UE measures the reference signals to derive an input to the ML module, or to derive an error statistic of the prediction module using measured values in comparison to predicted values.

[0041] The network entity 104 transmits 206, to the UE 102, reference signals associated with the CSI report configuration and / or first ML configuration. For example, the netw ork entity 104 transmits 206a, to the UE 102, a reference signal (e.g., first set of reference signals) based on the first ML configuration. The UE 102 may perform the CSI prediction usingthe first set of reference signals as input to a ML module based on the first ML configuration for CSI prediction. For example, the network entity 104 transmits 206b, to the UE 102, a downlink reference signal. The UE 102 determines channel feedback (e.g., different types of CSI) based on the downlink reference signal. In some aspects, the downlink reference signals for the CSI report and the reference signals for the first ML configuration may be associated with a same reference signal resource set or different reference signal resource sets.

[0042] In some aspects, the reference signal and / or downlink reference signal may include CSLreference signals (CSI-RSs), synchronization signal blocks (SSBs), tracking reference signals (TRSs).

[0043] The ML prediction module configuration may be based on the wireless channel property, which the UE 102 and / or the network entity 104 determines at least in part by UE mobility and / or the physical environment. In instances where the wireless channel property changes (e.g., increased / improved quality or decreased / degraded quality), the ML prediction module configuration is updated accordingly. When the ML prediction module configuration is not updated in a timely manner, an increase in prediction errors may occur, which may further lead to performance degradation in terms of data rate and transmission latency.

[0044] In some instances, the netw ork entity 104 monitors the UE performance, identifies the change in the wireless channel conditions based on channel feedback from the UE 102, and sends a ML module configuration update to the UE 102. In some other instances, the UE 102 monitors and identifies the change in the wireless channel conditions, requests the network entity 104 to update the ML module configuration, and receives the ML configuration update from the network entity 104 accordingly. The UE 102 may send the request for an updated ML module configuration in an effort to reduce prediction errors.

[0045] The UE 102 transmits 208, to the network entity 104, a report based on the downlink reference signal for updating the ML configuration. In some aspects, the report may be for channel feedback and may be based on the downlink reference signal for updating to a second CSI prediction configuration, where the second CSI prediction configuration is associated with a second set of reference signals. In some aspects, the UE transmits, to the network entity 104, an event triggered report based on a channel property exceeding a threshold. The event triggered report may include an error corresponding to a current ML module associated with the first configuration for CSI prediction, the error being based onat least one of: a confidence level of the ML module or a difference between a predicted value and a measured value of the downlink reference signal.

[0046] In some aspects, the UE 102 transmits, to the network entity 104, a report indicating that a condition has been satisfied for the updating to the second configuration. The updating to the second configuration may be based on receipt of instructions from the network entity 104 to update to the second configuration. The update to the second configuration may occur autonomously at the UE 102 after transmitting the report, where selecting the second configuration is based on the report. In some aspects, the selecting of the second configuration is based on an indication from the network entity 104 of a number of candidate configurations available for selection. In some other aspects, a second indication of the second configuration may include an updated condition to trigger a future ML configuration update.

[0047] By implementing a ML configuration / module update based on a report transmitted by the UE 102 to the network entity 104, the UE 102 may provide the network entity 104 with a report of the channel conditions which may autonomously trigger an update to the ML module configuration. As such, signaling overhead and latency may be reduced to update the ML module configuration at the UE 102. The ML configuration may be associated with a reference signal resource configuration that the netw ork entity7104 provides to the UE 102. The UE 102 measures the reference signal transmitted based on the reference signal resource configuration and uses the measurements as input for the current ML module. The ML configuration may indicate the identifier, functionality, content, and other parameters (e.g., periodicity) of the prediction report, which may be based on how far into the future the UE 102 can accurately predict CSI. The UE 102 sends the prediction report based on the output of the current ML module.

[0048] In some instances, different ML modules are used for different channel conditions. For example, an ML module is utilized for channel conditions having a high quality' or above a quality threshold, while another ML module is utilized for channel conditions having a low or reduced quality or below the quality threshold. The configuration of the reference signals that the UE 102 measures related to the ML module allows the UE 102 to derive an input for the ML module and / or to derive an error statistic of the prediction module by using, for example, measured values in comparison to predicted values. The configuration of the reference signals and report may include a periodicity of reference signals or reports and a time and frequency density of the reference signals. The update of reference signalconfiguration may include an increased periodicity in instances where the wireless channel has slow changes over time, which may allow for denser reference signals for a channel having a low SNR. In some instances, the report may indicate different performance metrics when using different ML modules. For example, the report may indicate Ll-RSRP for a channel having a low SNR, or a CQI report for a channel having a high SNR. In some instances, different ML modules are used for different SNR scenarios.

[0049] The update of configurations may be common across multiple ML module configurations. For example, different modules may share the same update for reference signal measurement configuration and reporting configuration. The configuration update may be common for ML modules for multiple component carriers on a pre-configured list. In some instances, the same report may trigger the update for multiple ML modules. For example, the TDCP report may trigger ML module configuration update for multiple ML modules.

[0050] The report sent to the network entity 104 may be an event triggered report, such that the UE 102 observes that a channel property’ has met a predefined condition (e.g., change in a channel metric exceeds or falls below a threshold). The predefined condition may be indicated in the ML module configuration. In some instances, the report may be sent using a two-step approach. For example, a UE 102 sends a scheduling request to the network entity 104 requesting resources (e.g., physical uplink shared channel (PUSCH) resource or physical uplink control channel (PUSCH) resource) to send 208 the report. The network entity7104 may send the UE 1 2 an UL grant to allow the UE 102 to send 208 the report using the resources allocated in the UL grant. In some instances, the report is sent 208 in an uplink medium access control (MAC) control element (CE) (MAC-CE). The report can be a UCI report which can be periodic, semi-persistent, or aperiodic. In some instances, the UCI report can be the TDCP report. In some instances, the report is sent via a PUSCH or PUCCH.

[0051] The ML configuration update is triggered in instances where the report indicates that certain conditions have been satisfied. For example, the change from a previous report meets the certain conditions. The conditions are indicated to the UE 102 by the network entity7104. The conditions may be included in the ML configuration. The report may include an error that corresponds to the current ML module configuration. For example, the error may correspond to a confidence level of the ML module or a difference between a predicted value and an actual measured value. In some instances, if the UE 102 reportsan increasing error in terms of the difference between the measured value and a predicted value, the UE 102 may request a reference signal configuration that allows for denser reference signal measurements, in time, so as to measure the input for the ML module. In some instances, the error report may trigger the ML configuration update. In yet some instances, the report may include an ML configuration recommendation by the UE 102.

[0052] In some instances, the number of reports or triggering of ML configuration update may be limited within a time period. For example, a timer may commence when the UE 102 sends the report or triggering of the ML configuration update, such that no further reports can be sent or no further ML module updates can be triggered prior to the expiration of the timer. A maximum number of reports or ML module updates and a timer duration is configured at the UE 102. The maximum number of reports or ML module updates and a timer duration may be preconfigured at the UE 102 or configurable by the network entity 104. In some aspects, the UE 102 may send M reports within any duration of time, where M > 0, or the UE 102 may be limited to sending 208 no more than M reports within the duration of time. In some aspects, the reporting function of the UE 102 may be disabled or limited for a duration of time, when the UE 102 has sent M reports.

[0053] The network entity 104 transmits 210 a second indication of a second ML configuration. The second indication may include a second configuration for CSI prediction. In some aspects, the UE receiving the second indication may be based on the UE transmitting the report for channel feedback. In some instances, the second indication may include instructions from the network for a second ML configuration. In some instances, the second indication may include a confirmation or acknowledgement of a second ML configuration among a candidate ML configurations.

[0054] The UE 102 and the network entity 104 update 211 to the second ML configuration based on the report.

[0055] The UE may update to the second configuration based on a change in a channel condition indicated in the report for channel feedback (e.g., CQI, RI, PMI, RSRP, RSRQ, RSSI, SNR, or SINR). The update includes different ML modules selected based on the report for channel feedback. The different ML modules are utilized for different channel conditions. The report may include performance metrics for the different ML modules. In some aspects, the update is applied across multiple ML module configurations.

[0056] After the UE 102 transmits the report to the network entity 104, the UE 102 updates the ML configuration. In some instances, the UE 102 updates the ML configurationautonomously after transmitting the report to the network entity 104. In some other instances, the UE 102 updates the ML configuration after receiving instructions from the network entity 104. The network entity 104 may provide instructions via a radio resource control (RRC) update for the ML configuration, MAC-CE, or DCI to activate / deactivate pre-configured configurations. In such instances, the UE 102 updates the ML configuration based on at least one of the report, predefined rules, or predefined conditions. In some instances, the predefined rule includes a time delay S after the UE 102 transmits the report to the network entity 104. In some instances, the predefined rule includes a time delay D after the UE 102 receives, from the network entity 104, a confirmation or acknowledgement of the updated ML configuration selected by the UE 102. In some aspects, the time delay S or D may be configured the network entity. In some other aspects, the time delay S or D may be pre-defined. The time delay S or D may be a function of tone spacing (e.g., subcarrier spacing) of at least one of the component carrier or bandwidth part that the UE 102 sends the report, performs the measurement, or at least one of the ML module output is associated with. In some aspects, the UE 102 is preconfigured with a table (Table 1 below) based on the TDCP value reported. The UE 102 may update the ML module configuration to the corresponding ML module configuration (e.g., ML Module Config ID) in the table. The ML modules (e.g., ML buildl, ML build2, etc.), reference signal configurations (e.g., RS configuration!, RS configuration, etc.), and report configurations (e.g., report configuration!, report configurations, etc.) may be preconfigured, downloaded, or stored by the UE 102.Table 1

[0057] In some aspects, the network may indicate a number of ML module candidate configurations for selection, such that the UE 102 selects one of the number of ML module candidate configurations based at least on the report and / or certain conditions. For example, the UE 102 may report the predicted metric as well as the selected ML module configuration used for the prediction. The network entity 104 may validate or monitor performance of the selected ML module configuration. The network entity’ 104 may update the ML module configuration and / or candidate module configurations even if the UE 102 does not transmit a triggering report.

[0058] In some aspects, the network entity’ 104 may be informed of the updated ML module configuration utilized at the UE 102. For example, for the ML functions configured by the network entity 104 for the UE 102, the network entity 104 may be informed of how the update is performed at the UE 102 to allow for the network entity 104 to transmit the corresponding reference signals that correspond with the ML configuration update or to monitor the performance of the ML configuration update. In some instances, the network entity 104 may utilize the same table (e.g., Table 1) to determine the updated configuration after receiving a TDCP report from the UE 102. The network entity 104, after updating the ML module configuration, may update the ML performance monitoring configuration. For example, the network entity 104 may invalidate any counter and / or timer associated with the prediction errors or failures of the previous ML configuration, and a new timer or counter is applied for the updated configuration. The timer or counter is utilized to count how often or how many times the UE 102 reports an error of the ML module configuration currently in use by the UE 102. In some instances, the network entity' 104 may update the scheduling configuration, CSI report, or configured PUSCH or physical downlink shared channel (PDSCH) transmission. In some aspects, the updated ML module configuration may be indicated to the UE 102 via RRC.

[0059] In some aspects, for some ML functions that the UE 102 runs locally (e.g., network transmit beam prediction and UE receive beam prediction) or configured by third party entities (e.g., third party servers), the network entity’ 104 may be notified of the update of the ML module configuration. For example, the update of the ML module configuration may change the CSI processing unit occupation and / or the minimum processing delay for CSI processing / prediction at the UE 102. In some instances, the UE 102 may indicate the corresponding changes to the network entity 104 in the report.

[0060] The network entity 104 transmits 212. to the UE 102, a reference signal based on the second ML configuration. The UE 102 may perform the CSI prediction using the second set of reference signals as input to a ML module based on the second configuration for CSI prediction. FIG. 2 describes a report triggered ML configuration update, whereas FIG. 3 is directed towards a TDCP report triggered ML configuration update.

[0061] FIG. 3 is a signaling diagram 300 illustrating communications between a UE 102 and a network entity 104 for a TDCP report that triggers an autonomous ML configuration update. 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.

[0062] The UE 102 may transmit 302, to the network entity 104, UE capability information indicating support for an ML configuration update in association with the first configuration for CSI prediction and the second configurations for CSI prediction. The UE capability' may indicate the types of reports the UE 102 supports that may trigger the ML module configuration update. For example, the UE transmits 302, to the network entity 104, UE capability for supporting TDCP report triggered ML configuration update. In some aspects, the UE capability indicating support for an ML configuration update in association with the first configuration for CSI prediction and the second configurations for CSI prediction. Time domain channel property (TDCP) is a type of uplink control information (UCI) where the UE 102 may report the amplitude and phase of a normalized autocorrelation function between a preconfigured duration of time interval in a wireless channel. A TDCP may indicate a rate of change of the wireless channel quality' over time.

[0063] The network entity 104 transmits 304a, to the UE 104, a CSI report configuration for configuring the report and a reference signal resource set for channel measurement. The CSI report configuration sets a reporting quantity as TDCP (referred to as a TDCP report). The TDCP report may trigger an ML configuration update. In some aspects, the CSI report configuration may configure the UE to report one or more of the following channel metrics: CQI, RI, RSRP, RSRQ, RSSI, SNR, or SINR.

[0064] The network entity 104 may transmit the CSI report configuration via RRC signaling (e.g., RRCReconflguration or CSI-ReportConflg). The network entity may provide some of the configurations or update some of the configurations by Medium Access Control (MAC) Control Element (CE) (e.g., MAC CE triggering the (semi-persistent) CSI report), or Downlink Control Information (DCI) (e.g., different triggering states for the DCI triggering the (aperiodic) CSI report may correspond to different configurations).

[0065] The network entity 104 transmits 304b, to the UE 102, a configuration of multiple candidate ML configurations for CSI prediction. For example, the configuration of the multiple candidate ML configurations includes a first ML configuration for CSI prediction, where the first configuration is associated with a first set of reference signals. The multiple candidate ML configurations include a second ML configuration for CSI prediction, where the second configuration is associated with a second set of reference signals.

[0066] The network entity 104 transmits 306, to the UE 102, reference signals associated with the CSI report configuration and a first ML configuration among candidate ML configurations. For example, the network entity 104 transmits 306a, to the UE 102, a reference signal (e.g., a first set of reference signals) associated with the first ML configuration among the candidate ML configurations. The UE 102 may perform the CSI prediction using the first set of reference signals as input to a ML module based on the first ML configuration among the candidate ML configurations for CSI prediction. For example, the network entity 104 transmits 306b, to the UE 102, a downlink reference signal associated with the CSI report configuration. The UE 102 determines the TDCP based on the downlink reference signal. In some aspects, the reference signal and / or downlink reference signal may include CSI-RS, SSBs, TRS.

[0067] The UE 102 transmits 308, to the network entity 104, a TDCP report based on the downlink reference signal for updating ML configuration. The TDCP report is associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction. The second configuration may be associated with a second set of reference signals. In some aspects, the TDCP report includes performance metrics for the different ML modules. In some other aspects, the TDCP report indicates that a condition has been satisfied for the updating to the second configuration. In some instances, an optimal configuration of a time domain prediction ML module may be based on the rate of wireless channel changes (or how fast the wireless channel is varying), which may be due to UE mobility and may be characterized by a TDCP. The ML configuration may indicate a first ML module from a set of candidate ML modules. In some instances, the UE 102 is configured to send TDCP reports to the network entity 104. When the UE 102 sends a TDCP report that satisfies a predefined condition and triggers a ML configuration update, the UE 102 changes the ML configuration autonomously based on a pre-configuration or pre-defined rule. The network entity 104 may update the MLconfiguration accordingly as well based on the TDCP report. In some aspects, the TDCP report may indicate the updated ML configuration selected or utilized by the UE 102.

[0068] In some aspects, a UE 102 is configured to run a ML module and send a report to the network entity 104 based on the results of the ML module. The report may trigger an update to the ML module configuration. The ML module configuration may include a selection of ML modules, a configuration of reference signals for measurement by the UE 102 related to the ML module, and / or a configuration of reports. The ML module configuration update may be based on the rate of change of the wireless channel (e.g., TDCP value) or the Doppler spread / shift. Different ML modules are selected based on the report and / or predefined rules. For example, the ML module configuration update can be based on channel conditions (e.g., SNR, SINR).

[0069] The network entity 104, may transmit 310, to the UE 104, confirmation of a second ML configuration among the candidate ML configurations. In some aspects, the confirmation includes a second indication of the second configuration. In some other aspects, the confirmation includes an updated condition to trigger a future ML configuration update. In some aspects, transmission of the confirmation is based on reception of the TDCP report.

[0070] The UE 102 and the network entity 104 update 31 1 to the second ML configuration based on the TDCP report. The update to the second configuration may be based on a change in a channel condition indicated in the TDCP report. The update includes different ML modules selected based on the report for channel feedback. The different ML modules are utilized for different channel conditions. In some aspects, the update is applied across multiple ML module configurations.

[0071] In some aspects, the update to the second configuration may be based on instructions from the network entity 104 to update to the second configuration. In other aspects, the update to the second configuration may occur autonomously at the UE 102 after transmitting the report, where selecting the second configuration is based on the report. The selecting of the second configuration is based on an indication from the network entity 104 of a number of candidate configurations available for selection. The UE sends, to the network entity 104, a selection indication identifying the selection of the second configuration. A second indication of the second configuration may include an updated condition to trigger a future ML configuration update.

[0072] The network entity 104 transmits 312, to the UE 102, a reference signal associated with the second ML configuration (e.g., second set of reference signals). The UE 102 mayperform the CSI prediction using the second set of reference signals as input to a ML module based on the second configuration for CSI prediction. FIGs. 2-3 illustrate signaling diagrams for of example ML configuration updates, whereas FIGs. 4-7 illustrate method diagrams for the example ML configuration updates. In particular, FIGs. 4-5 show an implementation by the UE 102, whereas FIGs. 6-7 show an implementation by the network entity 104.

[0073] FIG. 4 illustrates a flowchart 400 of a method of wireless communication at a UE. With reference to FIGs. 1-3, the method may be performed by the UE 102. In embodiments, the UE 102 transmits 402, to the network entity 104, UE capability information indicating support for an ML configuration update in association with the first configuration for CSI prediction and the second configurations for CSI prediction. For example, FIG. 2 shows that the UE 102 transmits 202 a capability (e.g., UE capability information) for supporting report triggered ML configuration update to the network entity 104. For example, FIG. 3 shows that the UE 102 transmits 302 a capability (e.g., UE capability information) for supporting TDCP report triggered ML configuration update to the network entity 104.

[0074] The UE 102 receives 404a. from the network entity 104. a CSI report configuration for the report. In some aspects, the CSI report configuration sets a reporting quantity as time domain channel property, TDCP. For example, FIG. 2 shows that the UE 102 receives 204a a CSI report configuration from the network entity 104. For example, FIG. 3 shows that the UE receives 304a a CSI report configuration from the network entity 104.

[0075] The UE 102 receives 404b, from a network entity 104, a first indication of a first configuration for CSI prediction. The first configuration is associated with a first set of reference signals. For example, FIG. 2 shows that the UE 102 receives 204b a first indication of a first ML configuration from the network entity 104. For example. FIG. 3 shows that the UE 102 receives 304b a configuration (e.g., first indication) of multiple candidate ML configuration for CSI prediction from the network entity7104. The configuration of the multiple candidate ML configurations includes a first ML configuration for CSI prediction and a second ML configuration for CSI prediction.

[0076] The UE 102 receives 406. from the network entity 104, a downlink reference signal. For example, FIG. 2 shows that the UE 102 receives 206a a reference signal based on the first ML configuration. For example, FIG. 2 shows that the UE 102 receives 206b a downlink reference signal from the network entity 104. For example, FIG. 3 shows that the UE 102 receives 306a a reference signal associated with a first ML configuration. Forexample, FIG. 3 shows that the UE 102 receives 306b a downlink reference signal from the network entity 104.

[0077] In some aspects, the UE 102 optionally determines at least one of an input for an ML module or an error statistic of a CSI prediction module based on a measurement of the downlink reference signal. In some aspects, the UE 102 receives, from the network entity 104, the first set of reference signals. The UE 102 performs the CSI prediction using the first set of reference signals as input to an ML module based on the first configuration for CSI prediction. In some aspects, the first set of reference signals may include the downlink reference signal.

[0078] The UE 102 transmits 408, to the network entity 104, a report for channel feedback based on the downlink reference signal. The report is associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction. The second configuration is associated with a second set of reference signals. For example, FIG. 2 show s that the UE 102 transmits 208 a report based on the downlink reference signal for updating the ML configuration. For example, FIG. 3 shows that the UE 102 transmits 308 a TDCP report based on the downlink reference signal for updating the ML configuration. The UE may update to the second configuration based on a change in a channel condition indicated in the report for channel feedback. The update includes different ML modules selected based on the report for channel feedback. The different ML modules are utilized for different channel conditions. The report includes performance metrics for the different ML modules. In some aspects, the update is applied across multiple ML module configurations.

[0079] In some aspects, the UE transmits, to the network entity 104, an event triggered report based on a channel property exceeding a threshold. The UE may transmit a request for resources for transmitting the event triggered report. The UE may then receive an uplink grant for the resources, where transmission of the event triggered report includes transmitting the event triggered report on the resources according to the uplink grant. The event triggered report may include an error corresponding to a current ML module associated with the first configuration for CSI prediction, the error being based on at least one of: a confidence level of the ML module or a difference between a predicted value and a measured value of the dow nlink reference signal.

[0080] In some aspects, the UE 102 transmits, to the network entity 104, a TDCP report, where the TDCP report indicates that a condition is satisfied for the updating to the secondconfiguration. The update to the second ML configuration may be based on receipt of instructions from the network entity 104 to update to the second configuration. The update to the second ML configuration may occur autonomously at the UE 102 after transmitting the report, where selecting the second ML configuration is based on the report. The selecting of the second ML configuration is based on an indication from the network entity 104 of a number of candidate ML configurations available for selection. The UE sends, to the network entity 104, a selection indication identifying the selection of the second ML configuration. A second indication of the second ML configuration may include an updated condition to trigger a future ML configuration update.

[0081] The UE 102 receives 410, from the network entity 104, a second indication of the second configuration for CSI prediction. For example, FIG. 2 shows that the UE 102 receives 210 a second indication of a second ML configuration from the network entity 104. For example, FIG. 3 shows that the UE 102 receives 310 a confirmation (e.g., second indication) of a second ML configuration among the candidate ML configurations.The UE receives 412, from the network entity 104, the second set of reference signals. For example. FIG. 2 shows that the UE 102 receives 212 a reference signal based on the second ML configuration (e.g., second set of reference signals) from the network entity 104. For example, FIG. 3 shows that the UE 102 receives 312 a reference signal based on the second ML configuration (e.g., second set of reference signals) from the network entity 104. The UE 102 may perform the CSI prediction using the second set of reference signals as input to a ML module based on the second configuration for CSI prediction. In some aspects, reception of the second indication may be based on transmission of the report for channel feedback.

[0082] FIG. 4 describes a method from a UE-side of a wireless communication link, whereas FIG. 5 further describes the method to include whether a report triggers an update of the configuration.

[0083] FIG. 5 is a flowchart 500 of a method of wireless communication at a UE including determining whether a report triggers an update of the ML configuration. With reference to FIGs. 1-3. the method may be performed by the UE 102. In embodiments, the UE 102 transmits 508 a report for channel feedback for updating to a second configuration. For example, FIG. 2 shows that the UE 102 transmits 208 a report based on the downlink reference signal for updating the ML configuration. For example, FIG. 3 shows that theUE 102 transmits 308 the TDCP report based on the downlink reference signal for updating ML configuration.

[0084] The UE determines 509 whether to update to the second configuration. The determination to update to the second configuration is based on the report for the channel feedback. In some aspects, the UE may determine to update to the second configuration based on transmission of the report indicating a change in channel conditions (e.g., channel property exceeding a threshold). In some aspects, the UE may determine to update to the second configuration when the UE sends the TDCP report that satisfied the predefined condition or a preconfigured rule, as discussed in connection to Table 1.

[0085] In instances where the UE 102 determines 509 to update to the second configuration (“Yes” branch), the UE 102 may receive 510 confirmation for the second configuration for CSI prediction. For example, FIG. 2 shows that the UE 102 receive 210 a second indication (e.g., confirmation for the second configuration) of a second ML configuration. For example, FIG. 3 shows that the UE 102 receive 310 a confirmation of a second ML configuration among the candidate ML configuration.

[0086] The UE 102 may receive 512 a new set of reference signals. F or example. FIG. 2 shows that the UE 102 receives 212 a reference signal based on the second ML configuration. For example, FIG. 3 shows that the UE 102 receives 312 a reference signal associated with the second ML configuration. The UE 102 may predict 514a CSI output based on the new set of reference signals (e.g., second set of reference signals). The new set of reference signals are used as input to the ML prediction module associated with the second ML configuration.

[0087] In instances where the UE 102 determines 509 to not update to the second configuration (“No” branch), the UE maintains 513 the current ML configuration. The UE 102 may receive 506c a first set of reference signals associated with a first ML configuration. For example, FIG. 2 shows that UE 102 receives 206a a reference signal based on the first ML configuration. For example, FIG. 3 shows that the UE 102 receives 306a a reference signal associated with the first ML configuration among the candidate ML configurations. The UE 102 may predict 514b CSI output based on the first set of reference signals. The first set of reference signals are used as input to the ML prediction module associated with the first ML configuration.

[0088] FIGs. 4-5 describe a method from a UE-side of a wireless communication link, whereas FIGs. 6-7 describe a method from a network-side of the wireless communication link.

[0089] FIG. 6 is a flowchart 600 of a method of wireless communication at a network entity. With reference to FIGs. 1-3, 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. In embodiments, the network entity 104 receives 602, from the UE 102, UE capability information indicating support for a machine learning, ML, configuration update in association with the first configuration for CSI prediction and the second configurations for CSI prediction. For example. FIG. 2 shows that the network entity’ 104 receives 202 a UE capability’ (e.g., UE capability information) for supporting report triggered ML configuration update from the UE 102. For example, FIG. 3 shows that the network entity 104 receives 302 a UE capability (e.g., UE capability information) for supporting TDCP report triggered ML configuration update from the UE 102.

[0090] The network entity 102 transmits 604a, to the UE 102, a CSI report configuration from the report, where the CSI report configuration sets a reporting quantity’ as TDCP. The report may trigger the ML configuration update. For example, FIG. 2 shows that the network entity 102 transmits 204a a CSI report configuration to the UE 102. For example, FIG. 3 shows that the network entity 102 transmits 304a a CSI report configuration to the UE 102.

[0091] The network entity transmits 604b, to the UE 102, a first indication of a first configuration for CSI prediction, the first configuration being associated with a first set of reference signals. For example, FIG. 2 shows that the network entity 104 transmits 204b a first indication of a first ML configuration. For example, FIG. 3 shows that the network entity 104 transmits 304b a configuration of multiple candidate ML configurations for CSI prediction. The configuration of the multiple candidate ML configurations includes a first ML configuration for CSI prediction and a second ML configuration for CSI prediction.

[0092] The network entity 104 transmits 606 a downlink reference signal to the UE 102. For example, FIG. 2 shows that the netw ork entity’ 104 transmits 206a a reference signal based on the first ML configuration. For example, FIG. 2 shows that the network entity' 104 transmits 206b a downlink reference signal. For example, FIG. 3 shows that the network entity 104 transmits 306a a reference signal associated with a first ML configuration among the candidate ML configurations. For example, FIG. 3 shows that the network entity 104 transmits 306b a downlink reference signal. In some aspects, the network entity 104 transmits, to the UE 102, the first set of reference signals, and the network entity 104 receives, from the UE. a CSI prediction report based on the first set of reference signalsassociated with the first configuration for CSI prediction. In some aspects, the first set of reference signals include the downlink reference signal.

[0093] The network entity 104 receives 608, from the UE 102, a report for channel feedback based on the downlink reference signal. The report is associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction. The second configuration is associated with a second set of reference signals. For example, FIG. 2 shows that the network entity 104 receives 208 a report based on the downlink reference signal for updating the ML configuration from the UE 102. For example, FIG. 3 shows that the network entity 104 receives 308 a TDCP report based on the downlink reference signal for updating the ML configuration from the UE 102. In some aspects, the UE updates to the second configuration based on a change in a channel condition indicated in the report for channel feedback. The update includes different ML modules selected based on the report for channel feedback. The different ML modules may be utilized for different channel conditions. In some aspects, the report includes performance metrics for the different ML modules. In some aspects, at least one of an input for an ML module or an error statistic of a CSI prediction module are based on the downlink reference signal. The update to the second configuration may be applied across the multiple ML module configurations.

[0094] In some aspects, the network entity receives, from the UE, an event triggered report based on a channel property exceeding a threshold. The network entity may receive a request for resources for transmitting the event triggered report. The network entity’ may transmit, to the UE, an uplink grant for the resources, where reception of the event triggered report includes receiving the event triggered report on the resources according to the uplink grant. In some aspects, the event triggered report includes an error corresponding to a current ML module associated with the first configuration for CSI prediction. The error being based on at least one of: a confidence level of the ML module or a difference betw een a predicted value and a measured value of the dow nlink reference signal.

[0095] In some aspects, the network entity receives, from the UE, a TDCP report, where the TDCP report indicates that a condition is satisfied for the updating to the second configuration. In some aspects, update to the second configuration is based on receipt of instructions from the network entity to update to the second configuration. In some aspects, update to the second configuration occurs autonomously after receiving the report. The second configuration may be based on an indication from the network entity of a numberof available candidate configurations. In some aspects, such as when the UE selects the second configuration autonomously, the network entity may receive, from the UE, a selection indication identifying the second configuration.

[0096] The network entity 104 transmits 610, to the UE 102, a second indication of the second ML configuration for CSI prediction. For example, FIG. 2 shows that the network entity 104 transmits 210 a second indication of a second ML configuration for CSI prediction. For example. FIG. 3 shows that the network entity 104 transmits 310 a confirmation of a second ML configuration among the candidate ML configurations for CSI prediction. In some aspects, the second indication of the second configuration includes an updated condition to trigger a future ML configuration update. In some aspects, transmission of the second indication is based on reception of the report for channel feedback.

[0097] The network entity 104 may transmit 612, to the UE 102, the second set of reference signals. For example, FIG. 2 shows that the network entity 104 transmits 212 a reference signal based on the second ML configuration. For example, FIG. 3 shows that the network entity 104 transmits 312 a reference signal associated with the second ML configuration. The network entity may receive, from the UE. a CSI prediction report based on the second set of reference signals associated with the second configuration for CSI prediction.

[0098] FIG. 6 describes a method from a network-side of a wireless communication link, whereas FIG. 7 describes the method to include whether a report triggers an update of the configuration.

[0099] FIG. 7 is a flowchart 700 of a method of wireless communication at a network entity including determining whether a report triggers an update of the configuration. With reference to FIGs. 1-3, 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. In embodiments, the network entity 104 receives 708 a report for channel feedback based on a first set of reference signals, the report being associated with updating to a second configuration. For example, FIG. 2 shows that the network entity 104 receives 208 a report based on the downlink reference signal for updating the ML configuration. For example, FIG. 3 shows that the network entity 104 receives 308 a TDCP report based on the downlink reference signal for updating the ML configuration.

[0100] The network entity determines 709 whether to updated to the second configuration. The determination to update to the second ML configuration is based on the report for thechannel feedback. In some aspects, the network entity may determine to update to the second configuration based on receipt of the report indicating a change in channel conditions (e.g., channel property exceeding a threshold). In some aspects, the network entity' may determine to update to the second configuration when the UE sends the TDCP report that satisfied the predefined condition or a preconfigured rule, as discussed in connection to Table 1.

[0101] In instances where the network entity 104 determines 709 to update to the second configuration (“Yes” branch), the network entity 104 may transmit 710 a confirmation for the second configuration. For example, FIG. 2 shows that the network entity 104 transmits 210 a second indication of a second ML configuration. For example, FIG. 3 shows that the network entity 104 transmits 310 a confirmation of a second ML configuration among the candidate ML configurations. The network entity 104 may transmit 712 a new set of reference signals (e.g., second set of reference signals) for the second configuration to the UE 102. For example, FIG. 2 shows that the network entity 104 transmits 212 a reference signal based on the second ML configuration. For example. FIG. 3 shows that the network entity 104 transmits 312 a reference signal associated with the second ML configuration.

[0102] In instances where the network entity 104 determines 709 to not update to the second configuration (“No” branch), the network entity 104 maintains 713 the current configuration. The network entity 104 may transmit 706c the first set of reference signals for the current configuration. For example, FIG. 2 shows that the network entity 104 transmits 206a reference signals based on the first ML configuration to the UE 102. For example, FIG. 3 shows that the network entity 104 transmits 306a a reference signal associated with the first ML configuration among the candidate ML configurations to the UE 102.

[0103] FIGs. 4-7 describe example methods of ML configuration updates, whereas FIGs. 8A- 8B describes a timeline of updating the ML configuration. FIG. 8A and 8B share similar reference elements which have been described herein, and will not be repeated herein, in an effort to reduce duplicity.

[0104] In FIGs. 8A and 8B, the UE sends 808 a report to the network entity. In some aspects, the report is for channel feedback based on the downlink reference signal. The report triggers an update to the second ML configuration for CSI prediction. The second ML configuration is associated with a second set of reference signals. For example, FIG. 2 shows that the UE transmits 208 a report based on the downlink reference signal forupdating the ML configuration. For example, FIG. 3 shows that the UE transmits 308 a TDCP report based on the downlink reference signal for updating the ML configuration.

[0105] The UE updates 811 the ML configuration to another ML configuration (e.g., second ML configuration). For example, FIG. 2 show's that the UE updates 211 to the second ML configuration based on the report. For example, FIG. 3 shows that the UE updates 311 to the second configuration based on the TDCP report. In FIG. 8A, the UE may update the configuration after a time period S 820a after sending 808 the report, where S > 0. The UE updates the configuration after the time period S 820a may allow' for synchronization between the UE and the network entity'. Accordingly, the UE receives 812 new reference signals (e.g., second set of reference signals) after updating the ML configuration. The new reference signals are associated with the updated ML configuration. For example. FIG. 2 shows that the UE receives 212 a reference signal based on the second ML configuration. For example, FIG. 3 shows that the UE receives 312 a reference signal associated with the second ML configuration.

[0106] In FIG. 8B, after sending 808 the report, the UE receives 810 a confirmation of updating the configuration. For example, FIG. 2 shows that the UE receives 210 a second indication (e.g., confirmation) of a second ML configuration. For example, FIG. 3 shows that the UE receives 310 a confirmation of the second ML configuration among the candidate ML configurations.

[0107] The UE updates 811 the configuration to another ML configuration (e.g., second ML configuration) after a time period D 820b after receipt of the confirmation, where D > 0. The UE updates the configuration after the time period D 820b may allow' for synchronization between the UE and the network entity. Accordingly, the UE receives 812 new reference signals (e.g., second set of reference signals) after updating the ML configuration. The new reference signals are associated with the updated ML configuration.

[0108] A UE apparatus 902, as described in FIG. 9, may perform the method of flowcharts 400-500 with reference to the signaling diagrams 200-300. The one or more network entities 104, as described in FIG. 10, may perform the method of flowcharts 600-700 with reference to the signaling diagrams 200-300.

[0109] FIG. 9 is a diagram 900 illustrating an example of a hardw are implementation for a UE apparatus 902. The UE apparatus 902 may be the UE 102, a component of the UE 102, or may implement UE functionality. The UE apparatus 902 may include an applicationprocessor 906, which may have on-chip memory 906’. In examples, the application processor 906 may be coupled to a secure digital (SD) card 908 and / or a display 910. The application processor 906 may also be coupled to a sensor(s) module 912, a power supply 914, an additional module of memory 916, a camera 918, and / or other related components.

[0110] The UE apparatus 902 may further include a wireless baseband processor 926, which may be referred to as a modem. The wireless baseband processor 926 may have on-chip memory 926'. Along with, and similar to, the application processor 906. the wireless baseband processor 926 may also be coupled to the sensor(s) module 912, the power supply 914, the additional module of memon 916, the camera 918, and / or other related components. The wireless baseband processor 926 may be additionally coupled to one or more subscriber identity module (SIM) card(s) 920 and / or one or more transceivers 930 (e.g., wireless RF transceivers).

[0111] Within the one or more transceivers 930, the UE apparatus 902 may include a Bluetooth module 932, a WLAN module 934, an SPS module 936 (e.g., GNSS module), and / or a cellular module 938. The Bluetooth module 932, the WLAN module 934, the SPS module 936, and the cellular module 938 may each include an on-chip transceiver (TRX). or in some cases, just a transmitter (TX) or just a receiver (RX). The Bluetooth module 932, the WLAN module 934, the SPS module 936, and the cellular module 938 may each include dedicated antennas and / or utilize antennas 940 for communication with one or more other nodes. For example, the UE apparatus 902 can communicate through the transceiver(s) 930 via the antennas 940 with another UE (e.g., sidelink communication) and / or with a network entity 104 (e.g., uplink / downlink communication), where the network entity 104 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.

[0112] The wireless baseband processor 926 and the application processor 906 may each include a computer-readable medium / memory 926', 906', respectively. The additional module of memory 916 may also be considered a computer-readable medium / memory'. Each computer-readable medium / memory 926'. 906'. 916 may be non-transitory. The wireless baseband processor 926 and the application processor 906 may each be responsible for general processing, including execution of software stored on the computer-readable medium / memory 926', 906', 916. The software, when executed by the wireless baseband processor 926 / application processor 906, causes the wireless baseband processor 926 / application processor 906 to perform the various functions described herein. Thecomputer-readable medium / memory may also be used for storing data that is manipulated by the wireless baseband processor 926 I application processor 906 when executing the software. The wireless baseband processor 926 / application processor 906 may be a component of the UE 102. The UE apparatus 902 may be a processor chip (e.g., modem and / or application) and include just the wireless baseband processor 926 and / or the application processor 906. In other examples, the UE apparatus 902 may be the entire UE 102 and include the additional modules of the apparatus 902.

[0113] As discussed in FIG. 1 and implemented with respect to FIGs. 2-5, the reporting component 140 is configured to receive, from a network entity 104, a first indication of a first configuration for CSI prediction, the first configuration being associated with a first set of reference signals; receive, from the network entity 104, a downlink reference signal; and transmit, to the network entity 104, a report for channel feedback based on the downlink reference signal, the report being associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals. The reporting component 140 may be within the application processor 906 (e.g.. at 140a). the wireless baseband processor 926 (e g., at 140b), or both the application processor 906 and the wireless baseband processor 926. The reporting component 140a-140b may be one or more hardware components specifically configured to carry out 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.

[0114] FIG. 10 is a diagram 1000 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 1046, which may have on-chip memory’ 1046'. In some aspects, the CU 110 may further include an additional module of memory 1056 and / or a communications interface 1048, both of which may be coupled to the CU processor 1046. The CU 110 can communicate with the DU 108 through a midhaul link 162, such as an Fl interface between the communications interface 1048 of the CU 110 and a communications interface 1028 of the DU 108.

[0115] The DU 108 may include a DU processor 1026, which may have on-chip memory 1026'. In some aspects, the DU 108 may further include an additional module of memory 1036 and / or the communications interface 1028, both of which may be coupled to the DU processor 1026. The DU 108 can communicate with the RU 106 through a fronthaul link 160 between the communications interface 1028 of the DU 108 and a communications interface 1008 of the RU 106.

[0116] The RU 106 may include an RU processor 1006, which may have on-chip memory 1006'. In some aspects, the RU 106 may further include an additional module of memory 1016, the communications interface 1008, and one or more transceivers 1030, all of which may be coupled to the RU processor 1006. The RU 106 may further include antennas 1040, which may be coupled to the one or more transceivers 1030, such that the RU 106 can communicate through the one or more transceivers 1030 via the antennas 1040 with the UE 102.

[0117] The on-chip memory 1006', 1026', 1046' and the additional modules of memory71016, 1036. 1056 may each be considered a computer-readable medium / memory. Each computer-readable medium / memory may be non-transitory. Each of the processors 1006. 1026, 1046 is responsible for general processing, including execution of software stored on the computer-readable medium / memory7. The software, when executed by the corresponding processor(s) 1006, 1026, 1046 causes the processor(s) 1006, 1026, 1046 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) 1006, 1026, 1046 when executing the software. In examples, the ML 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.

[0118] As discussed in FIG. 1 and implemented with respect to FIGs. 2-3 and 6-7, the ML configuration component 150 is configured to transmit, to a UE 102, a first indication of a first configuration for CSI prediction, the first configuration being associated with a first set of reference signals; transmit, to the UE 102, a downlink reference signal; and receive, from the UE 102, a report for channel feedback based on the downlink reference signal, the report being associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals. The ML configuration component 150 may be withinone or more processors of the one or more network entities 104, such as the RU processor 1006 (e.g., at 150a), the DU processor 1026 (e.g., at 150b), and / or the CU processor 1046 (e.g., at 150c). The ML configuration component 150a-150c may be one or more hardware components specifically configured to carry out the stated processes / algorithm, implemented by one or more processors 1006, 1026, 1046 configured to perform the stated processes / algorithm, stored within a computer-readable medium for implementation by the one or more processors 1006, 1026, 1046. or a combination thereof.

[0119] 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.

[0120] 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 description includes 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.

[0121] 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.

[0122] 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 hardware 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 software, 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.

[0123] If the functionality described herein is implemented in software, the functions may be stored on, or encoded as, one or more instructions or code on a computer-readable medium, such as a non-transilory computer-readable storage medium. Computer-readable media includes computer storage media and can include a random-access memory' (RAM), a readonly memory’ (ROM), an electrically erasable programmable ROM (EEPROM), optical disk storage, magnetic disk storage, other magnetic storage devices, combinations of these ty pes 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.

[0124] Aspects, implementations, and / or use cases described herein may be implemented across many differing platform types, 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-module-component 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 chip-level 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.

[0125] 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 signalsnecessarily 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.

[0126] 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.

[0127] 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” do 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.

[0128] 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 “multiplewidgets” 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”.

[0129] 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.

[0130] 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.

[0131] 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 words “module.” “mechanism.” “element,” “device.” and the like may not be a substitute for the w ord “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.

[0132] The following examples are illustrative only and may be combined with other examples or teachings described herein, without limitation.

[0133] Example 1 is a method of wireless communication as described herein at a UE comprising: receiving, from a network entity, a first indication of a first configuration for CSI prediction, the first configuration being associated with a first set of reference signals; receiving, from the network entity, a downlink reference signal; and transmitting, to the network entity, a report for channel feedback based on the downlink reference signal, the report being associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals.

[0134] Example 2 is the method of example 1, further including transmitting, to the network entity, UE capability information indicating support for a ML configuration update in association with the first configuration for CSI prediction and the second configurations for CSI prediction.

[0135] Example 3 is the method of any of examples 1-2, further including updating to the second configuration based on a change in a channel condition indicated in the report for channel feedback.

[0136] Example 4 is the method of any of examples 1-3, where the updating includes different machine learning, ML, modules selected based on the report for channel feedback.

[0137] Example 5 is the method of example 4, where the different ML modules are utilized for different channel conditions.

[0138] Example 6 is the method of any of examples 4-5, where the report includes performance metrics for the different ML modules.

[0139] Example 7 is the method of any of examples 1-6, further including determining at least one of an input for an ML module or an error statistic of a CSI prediction module based on a measurement of the downlink reference signal.

[0140] Example 8 is the method of any of examples 1-7, where the updating is applied across multiple ML module configurations.

[0141] Example 9 is the method of any of examples 1-8, further including receiving, from the network entity, a CSI report configuration for the report, where the CSI report configuration sets a reporting quantity as TDCP and where the report triggers the ML configuration update.

[0142] Example 10 is the method of any of examples 1-9, where the transmitting the report further including transmitting, to the network entity, an event triggered report based on a channel property exceeding a threshold.

[0143] Example 11 is the method of example 10, further including transmitting a request for resources for transmitting the event triggered report; and receiving an uplink grant for the resources, wherein the transmitting the event triggered report includes transmitting the event triggered report on the resources according to the uplink grant.

[0144] Example 12 is the method of any of examples 10-11, where the event triggered report includes an error corresponding to a current ML module associated with the first configuration for CSI prediction, the error being based on at least one of a confidence levelof the ML module or a difference between a predicted value and a measured value of the downlink reference signal.

[0145] Example 13 is the method of any of examples 9-12, where the transmitting the report further including transmitting, to the network entity, a TDCP report, where the TDCP report indicates that a condition is satisfied for the updating to the second configuration.

[0146] Example 14 is the method of any of examples 1-13, further including receiving, from the network entity, a second indication of the second configuration for CSI prediction; receiving, from the network entity, the second set of reference signals; and performing the CSI prediction using the second set of reference signals as input to a ML module based on the second configuration for CSI prediction.

[0147] Example 15 is the method of any of examples 1-14, where the updating to the second configuration is based on receipt of instructions from the network entity to update to the second configuration.

[0148] Example 16 is the method of any of examples 1-14, where the updating to the second configuration occurs autonomously at the UE after the transmitting the report, the method further comprising selecting the second configuration based on the report.

[0149] Example 17 is the method of example 16, where the selecting of the second configuration is based on an indication from the network entity of a number of candidate configurations available for selection.

[0150] Example 18 is the method of any of examples 16-17, further including sending, to the network entity, a selection indication identifying the selection of the second configuration.

[0151] Example 19 is the method of any of examples 1-18, where a second indication of the second configuration includes an updated condition to trigger a future ML configuration update.

[0152] Example 20 is the method of any of examples 1-19, wherein the receiving the second indication is based on the transmitting the report for channel feedback.

[0153] Example 21 is the method of any of examples 1-20, further including receiving, from the network entity, the first set of reference signals; and performing the CSI prediction using the first set of reference signals as input to a ML module based on the first configuration for CSI prediction.

[0154] Example 22 is the method of any of examples 1-21, where the first set of reference signals include the downlink reference signal.

[0155] Example 23 is a method of wireless communication at a network entity, comprising: transmitting, to a UE, a first indication of a first configuration for CSI prediction, the first configuration being associated with a first set of reference signals; transmitting, to the UE, a downlink reference signal; and receiving, from the UE, a report for channel feedback based on the downlink reference signal, the report being associated with updating to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals.

[0156] Example 24 is the method of example 23, further including receiving, from the UE, UE capability information indicating support for a ML configuration update in association with the first configuration for CSI prediction and the second configurations for CSI prediction.

[0157] Example 25 is the method of any of examples 23-24, further including updating to the second configuration based on a change in a channel condition indicated in the report for channel feedback.

[0158] Example 26 is the method of any of examples 23-25, where the updating includes different ML modules selected based on the report for channel feedback.

[0159] Example 27 is the method of example 26, where the different ML modules are utilized for different channel conditions.

[0160] Example 28 is the method of any of examples 26-27, where the report includes performance metrics for the different ML modules.

[0161] Example 29 is the method of any of examples 23-28, where at least one of: an input for an ML module or an error statistic of a CSI prediction module are based on the downlink reference signal.

[0162] Example 30 is the method of any of examples 23-29, where the updating is applied across multiple ML module configurations.

[0163] Example 31 is the method of any of examples 23-30, further including transmitting, to the UE, a CSI report configuration from the report, where the CSI report configuration sets a reporting quantity as TDCP and where the report triggers the ML configuration update.

[0164] Example 32 is the method of any of examples 23-31, where the receiving the report further includes receiving, from the UE. an event triggered report based on a channel property exceeding a threshold.

[0165] Example 33 is the method of example 32, further including receiving a request for resources for transmitting the event triggered report; and transmitting an uplink grant forthe resources, where the receiving the event triggered report includes receiving the event triggered report on the resources according to the uplink grant.

[0166] Example 34 is the method of any of examples 32-33, where the event triggered report includes an error corresponding to a current ML module associated with the first configuration for CSI prediction, the error being based on at least one of: a confidence level of the ML module or a difference between a predicted value and a measured value of the downlink reference signal.

[0167] Example 35 is the method of any of examples 31-34, where the receiving the report further comprises receiving, from the UE, a TDCP report, where the TDCP report indicates that a condition is satisfied for the updating to the second configuration.

[0168] Example 36 is the method of any of examples 23-35, further including transmitting, to the UE, a second indication of the second configuration for CSI prediction; transmitting, to the UE, the second set of reference signals; and receiving, from the UE, a CSI prediction report based on the second set of reference signals associated with the second configuration for CSI prediction.

[0169] Example 37 is the method of any of examples 23-36, where the updating to the second configuration is based on receipt of instructions from the network entity to update to the second configuration.

[0170] Example 38 is the method of any of examples 23-37, where the updating to the second configuration occurs autonomously after receiving the report.

[0171] Example 39 is the method of example 38, where the second configuration is based on an indication from the network entity of a number of available candidate configurations.

[0172] Example 40 is the method of any of examples 38-39, further including receiving, from the UE. a selection indication identifying the second configuration.

[0173] Example 41 is the method of any of examples 23-40, where a second indication of the second configuration includes an updated condition to trigger a future ML configuration update.

[0174] Example 42 is the method of any of examples 23-41, where the transmitting the second indication is based on the receiving the report for channel feedback.

[0175] Example 43 is the method of any of examples 23-42, further including transmitting, to the UE, the first set of reference signals; and receiving, from the UE, a CSI prediction report based on the first set of reference signals associated with the first configuration for CSI prediction.

[0176] Example 44 is the method of any of examples 23-43, where the first set of reference signals include the downlink reference signal.

[0177] Example 45 is an apparatus for wireless communication for implementing a method as in any of examples 1-44.

[0178] Example 46 is an apparatus for wireless communication including means for implementing a method as in any of examples 1-44.

[0179] Example 47 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 1-44.

[0180] Example 48 is a computer program product for implementing a method as in any of examples 1-44.

Claims

CLAIMSWHAT IS CLAIMED IS:

1. A method of wireless communication at a user equipment, UE, (102), comprising: receiving (204, 304), from a network entity (104), a first indication of a first configuration for channel state information, CSI, prediction, the first configuration being associated with a first set of reference signals; receiving (206, 306), from the network entity (104), a downlink reference signal; and transmitting (208, 308), to the network entity' (104), a report for channel feedback based on the downlink reference signal, the report being associated with updating (211, 311) to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals.

2. The method of claim 1, further comprising: transmitting (202, 302), to the network entity (104), UE capability information indicating support for a machine learning, ML, configuration update in association with the first configuration for CSI prediction and the second configuration for CSI prediction.

3. The method of any of claims 1-2. further comprising updating (211, 311) to the second configuration based on a change in a channel condition indicated in the report for channel feedback.

4. The method of any of claims 1-3, wherein the updating (211, 311) includes different machine learning, ML. modules selected based on the report for channel feedback.

5. The method of claim 4, wherein the different ML modules are utilized for different channel conditions, wherein the report includes performance metrics for the different ML modules.

6. The method of any of claims 1-5. further comprising: determining at least one of an input for an ML module or an error statistic of a CSI prediction module based on a measurement of the downlink reference signal.

7. The method of any of claims 3-6, wherein the updating (211, 311) is applied across multiple ML module configurations.

8. The method of any of claims 1-7, further comprising: receiving (304a), from the network entity' (104), a CSI report configuration for the report, wherein the CSI report configuration sets a reporting quantity as time domain channel property. TDCP. and wherein the report triggers (308) the ML configuration update.

9. The method of any of claims 1-8, wherein the transmitting (208,308) the report comprises: transmitting, to the network entity (104), an event triggered report based on a channel property exceeding a threshold.

10. The method of claim 9, further comprising: transmitting a request for resources for transmitting the event triggered report; and receiving an uplink grant for the resources, wherein the transmitting the event triggered report includes transmitting the event triggered report on the resources according to the uplink grant.

11. The method of any of claims 9-10, wherein the event triggered report includes an error corresponding to a current ML module associated with the first configuration for CSI prediction, the error being based on at least one of: a confidence level of the ML module or a difference between a predicted value and a measured value of the downlink reference signal.

12. The method of any of claims 8-11. wherein the transmitting (308) the report further comprises: transmitting (308), to the network entity (104), a TDCP report, wherein the TDCP report indicates that a condition is satisfied for the updating (311) to the second configuration.

13. The method of any of claims 1-12, further comprising: receiving (210, 310), from the network entity (104), a second indication of the second configuration for CSI prediction; receiving (212, 312), from the netw ork entity (104), the second set of reference signals; andperforming the CSI prediction using the second set of reference signals as input to a machine learning, ML. module based on the second configuration for CSI prediction.

14. The method of any of claims 3-13, wherein the updating (21 1, 311) to the second configuration is at least one of: based on receipt of instructions from the network entity (104) to update to the second configuration, or occurs autonomously at the UE (102) after the transmitting (208, 308) the report, the method further comprising selecting the second configuration based on the report, wherein the selecting of the second configuration is based on an indication from the network entity (104) of a number of candidate configurations available for selection.

15. The method of claim 14, further comprising: sending, to the network entity7(104), a selection indication identifying the selection of the second configuration.

16. The method of any of claims 13-15, wherein the second indication of the second configuration includes an updated condition to trigger a future ML configuration update.

17. The method of any of claims 13-16, wherein the receiving (210) the second indication is based on the transmitting (208) the report for channel feedback.

18. The method of any of claims 1-17, further comprising: receiving, from the network entity (104), the first set of reference signals; and performing the CSI prediction using the first set of reference signals as input to a machine learning, ML, module based on the first configuration for CSI prediction.

19. A method of wireless communication at a network entity (104), comprising: transmitting (204, 304), to a user equipment. UE, (102), a first indication of a first configuration for channel state information, CSI, prediction, the first configuration being associated with a first set of reference signals; transmitting (206, 306), to the UE (102). a downlink reference signal; andreceiving (208, 308). from the UE (102), a report for channel feedback based on the downlink reference signal, the report being associated with updating (211, 311) to a second configuration for CSI prediction different from the first configuration for CSI prediction, the second configuration being associated with a second set of reference signals.

20. An apparatus for wireless communication comprising a memory’, a transceiver, and a processor coupled to the memory and the transceiver, the apparatus being configured to implement a method as in any of claims 1-19.

Images