WO2025072002A1

WO2025072002A1 - Methods and apparatus for two-sided model pairing

Info

Publication number: WO2025072002A1
Application number: PCT/US2024/047218
Authority: WO
Inventors: Huaning Niu; Weidong Yang; Chunxuan Ye; Wei Zeng; Oghenekome Oteri; Dawei Zhang
Original assignee: Apple Inc
Current assignee: Apple Inc
Priority date: 2023-09-27
Filing date: 2024-09-18
Publication date: 2025-04-03
Anticipated expiration: 2026-03-27
Also published as: CN121794930A

Abstract

Systems and methods of two-sided model alignment and pairing are disclosed herein. A user equipment (UE) reports, to a base station, a UE capability comprising UE-supported model pairing information and an indication that the UE supports artificial intelligence (AI) based channel state information (CSI) compression. The UE receives, from the base station, pairing configuration information corresponding to at least part of the UE-supported model pairing information. The UE selects, based on the pairing configuration information, a selected model pair for an encoder at the UE corresponding to a decoder at the base station and generates a compressed CSI by providing downlink (DL) channel data or a DL precoder to the encoder. The UE transmits, from the UE to the base station, an uplink control information (UCI) report including the compressed CSI.

Description

METHODS AND APPARATUS FOR TWO-SIDED MODEL PAIRING

TECHNICAL FIELD

[0001] This application relates generally to wireless communication systems, including model alignment and two-sided model pairing.

BACKGROUND

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

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

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

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

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

.BRIEF..PESCMPTIQN.QF

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

[0008] FIG. 1 illustrates an encoder and a decoder in a CSI feedback operation according to certain embodiments.

[0009] FIG. 2 illustrates an example signal diagram of a UE capability report indicating support of Al based CSI compression according to one embodiment.

[0010] FIG. 3 illustrates an example signal diagram of a UE capability report indicating support of Al based CSI compression according to another embodiment.

[0011] FIG. 4 illustrates an example signal diagram of model alignment by sending pairing configuration information in a CSI-reportConfig signal according to one embodiment.

[0012] FIG. 5 illustrates an example of a CSI-reportConfig message with an additional entry to provide Al based CSI compression related signaling according to one embodiment.

[0013] FIG. 6 illustrates an example signal diagram of model alignment by indicating a selected model pair in a UCI report according to one embodiment.

[0014] FIG. 7 illustrates a flowchart of a method for a UE according to embodiments herein.

[0015] FIG. 8 illustrates a flowchart of a method for a base station according to embodiments herein.

[0016] FIG. 9 illustrates an example architecture of a wireless communication system according to embodiments disclosed herein. [0017] FIG. 10 illustrates a system for performing signaling between a wireless device and a network device according to embodiments disclosed herein.

DETAILED DESCRIPTION

[0018] Various embodiments are described with regard to a UE. However, reference to a UE is merely provided for illustrative purposes. The example embodiments may be utilized with any electronic component that may establish a connection to a network (NW) and is configured with the hardware, software, and/or firmware to exchange information and data with the network. Therefore, the UE as described herein is used to represent any appropriate electronic component.

[0019] Downlink channel state information (CSI) (e.g., for frequency division duplex (FDD) operation) may be sent from a UE to a base station through feedback channels. The base station may use the CSI feedback, for example, to reduce interference and increase throughput for massive multiple-input multiple-output (MIMO) communication. However, such feedback uses excessive overhead. Vector quantization or codebookbased feedback may be used to reduce feedback overhead. The feedback quantities resulting from these approaches, however, are scaled linearly with the number of transmit antennas, which may be difficult when hundreds or thousands of centralized or distributed transmit antennas are used.

[0020] Artificial intelligence (Al) and/or machine learning (ML) may be used for CSI feedback enhancement to reduce overhead, improve accuracy, and/or generate predictions. Al and/or ML may also be used, for example, for beam management (e.g., beam prediction in time/spatial domain for overhead and latency reduction and beam selection accuracy improvement) and/or positioning accuracy enhancements.

[0021] CSI feedback using Al and/or ML may be formulated as a joint optimization of an encoder and a decoder. At a high level, FIG. 1 illustrates an encoder 102 of a UE and a decoder 104 of a base station (e.g.. gNB) in an Al based CSI feedback operation according to certain embodiments. As shown, the encoder 102 receives a downlink (DL) channel H or a DL precoder and outputs Al based CSI feedback. The encoder 102 learns a transformation from original transformation matrices to compressed representations (codewords) through training data. The decoder 104 learns an inverse transformation from the codewords to the original channels. Thus, the decoder 104 can receive the Al based CSI feedback (codewords) from the encoder 102 and output a reconstructed channel H or DL precoder (which can be also denoted by H). End-to-end learning (e g., with an unsupervised learning algorithm) may be used to train the encoder 102 and the decoder 104. Typically, normalized mean square error (NMSE) or cosine similarity is the optimization metric. In some designs, the DL channel H can be replaced with DL precoder. Hence, the encoder 102 takes the DL precoder as input and generates Al based CSI feedback and the decoder 104 takes the Al based CSI feedback and reconstructs the DL precoder.

[0022] Various types of neural network (NN) encoders/decoders can be trained for different purposes, with different tradeoffs of complexity, overhead, and performance. A convolutional neural network (CNN) may, for example, be used for CSI feedback for frequency and spatial domain CSI reference signal (CSI-RS) compression. Other examples include using a transformer or a generative adversarial network (GAN). Depending on the number of receive antennas and rank, either channel feedback or precoding matrix feedback can be used. For example, with four receive antennas, rank 1 and rank 2 feedback may potentially use Al NN trained with eigenvectors as input, whereas rank 3 and rank 4 can potentially use a channel matrix as input to a trained Al NN. Data preprocessing can be used on input of an Al model. Preprocessing from frequency domain to time domain may be used and some of the small paths may be removed before input to the Al NN. A maximum rank indicates a maximum number of layers per UE, which corresponds to a lack of correlation or interference between the UE's antennas. For example, rank 1 corresponds to a maximum of one spatial layer for the UE, rank 2 corresponds to a maximum of two spatial layers for the UE, rank 3 corresponds to a maximum of three layers for the UE, and rank 4 corresponds to a maximum of four layers for the UE.

[0023] CNN+RNN (recurrent NN) based NN may be used for time domain, frequency domain, and spatial domain CSI-RS compression. The input may be a time sequence with a set of CSI-RS configurations. A preprocessed time sequence such as frequency domain pre-processing (to time domain and removing small channel taps), and Doppler domain preprocessing can also be applied as Al input. Angular domain preprocessing is also possible, however, angular domain preprocessing may not be efficient in certain implementations.

[0024] An AI/ML model may be developed and managed during an entire lifecycle using either a functionality-based life cycle management (LCM) or model-identifier (ID)-based LCM. However, certain aspects of functionality -based LCM and model-ID- based LCM have not been fully developed or analyzed, such as model identification procedures, model delivery and/or transfer analysis, two-sided model training type analysis, data collection and performance monitoring for both one-sided and two-sided models (including ground-truth related and dataset delivery' related aspects), inference- related framework (e.g., CSI configuration, payload related aspects, and quantization), and developing and clarifying two-sided model pairing mechanisms.

[0025] Model Pairing Information

[0026] In wireless communication systems with CSI compression using a two-sided model, procedures are needed to align the information that enables a UE to select CSI generation model(s) compatible with the CSI reconstruction model(s) used by a gNB. For example, at least the following cases may be used to define the pairing information to enable the UE to select CSI generation model(s) compatible with the CSI reconstruction model(s) used by the gNB. In one case, the pairing configuration information is provided in the form of a CSI reconstruction model ID that the network will use. In a second case, the pairing information is provided in the form of a CSI generation model ID that the UE will use. In a third case, the pairing information is provided in the form of a paired CSI generation model and CSI reconstruction model ID. In a fourth case, the pairing information is provided by a dataset ID during ty pe 3 sequential training. In a fifth case, the pairing information is provided in the form of a training session ID to a prior training session (e.g., a training application programming interface (API)) between the network and the UE. In a sixth case, the pairing information is up to UE and network offline co-engineering alignment between a UE vendor and a network vendor, transparent to the 3GPP specification as defined in current wireless communication systems.

[0027] It may be noted that the disclosure of the vendor information during the model pairing procedure and model identification procedure should be considered. If each UE side model is compatible with all network side models, the described information is not needed for the UE. Further, it may be that the above cases do not imply that there is a need for a central entity for defining, storing, and/or maintaining the IDs.

[0028] One to Many (1 :M) Network First Training

[0029] In network first separate training with dataset sharing for CSI compression, for the pairing between one UE part model and more than one separate network part models, when taking 1-on-l joint training between the network part model and the UE part model as a benchmark, a larger performance loss is observed in general as compared to the network first separately training with one UE part model and one network part model pairing. For example, six vendors observed minor loss of 0% — 1.6% compared to the 1- on-1 joint training, three vendors observed moderate loss of -1.9% — 6.64% compared to the 1-on-l joint training, and five vendors observed significant loss of -37.9% — 87% compared to the 1-on-l joint training. It may be noted that as opposed to vendors which observed significant loss, the minor loss observed by other vendors may be attributed to special handling (e.g., further definition of an adaptation layer) being performed to pair with more than one network part models during the training at the UE side. In view of there being a larger performance loss when using 1-on-l joint training between the network part model and the UE part model, the UE may train a separate UE part model for each network dataset. In this case the UE may need to know the network part model that is deployed in the field to derive the UE part encoder, thus indicating which encoder to use. In some embodiments disclosed herein, however, the UE may not care about the vendor of various base stations, thus resulting in wide range of possible UE part encoders.

[0030] In certain systems with CSI compression using two-sided models, an uplink control information (UCI) format is based on CSI reporting principles with CSI Part 1 and CSI Part 2, where CSI Part 1 has a network configured size and CSI Part 2 size is dynamic, as determined by the information in CSI Part 1. The CSI Part 1 is multiplexed with UCI Part 1 and the CSI Part 2 is multiplexed with UCI Part 2.

[0031] Embodiments disclosed herein include procedures for model alignment that are described, at a high level, in terms of a first step, a second step, and a third step. However, each step is not used in every embodiment. Indeed, examples are provided for embodiments that use only one step, only two steps, or all three steps.

[0032] In the first step, the UE provides a UE capability report indicating that the UE supports Al based CSI compression. The UE capability report may be part of a UE capability inquiry and response, e.g., as part of a radio resource control (RRC) reconfiguration message. In certain embodiments, a vendor ID is not part of the information exchange.

[0033] In the second step, which is performed if the first step has more than one instance of model pairing information, pairing configuration information is provided from the base station to the UE, e.g.. as a part of CSI report configuration (CSI- reportConfig) message. In some such embodiments, a new field may be added to the CSI-ReportConfig indicating Al based CSI compression and pairing configuration information is included as part of the RRC configuration.

[0034] In the third step, which is performed if the second step includes pairing configuration information for more than one model pair, includes the UE indicating a selected model pair to the base station, e.g., as part of a UCI report.

[0035] UE Capability Report

[0036] For the first step, FIG. 2 illustrates an example signal diagram 200 of a UE capability report indicating support of Al based CSI compression according to one embodiment. In this example, a UE 202 transmits a UE capability report 206 to a network (NW) via a base station 204 indicating support of Al based CSI compression. The UE capability report 206 also includes supported pairing information, e.g., based on offline model identification type A.

[0037] Depending on the case used to define the pairing information, as described herein, the UE reports supported model pairing information to the network. If a case with adaptation layer is used, a sub-model ID can be included to indicate the adaptation layer when, for example, the pairing configuration information is provided in the form of a CSI reconstruction model ID that the network will use. the pairing information is provided in the form of a CSI generation model ID that the UE will use, or the pairing information is provided in the form of a paired CSI generation model and CSI reconstruction model ID.

[0038] In some situations, if the capability report that the UE 202 sends to the base station 204 is static, a large amount of overhead may be used. Thus, in some embodiments, as illustrated in FIG. 2, it may be assumed that utilized models have good generalization performance across different scenarios, cells, and/or antenna configurations. In other words, the utilized model is more general and can work for various scenarios or implementations that may not need a large number of models to be defined. Further, the UE part model can be vendor specific (i.e., the UE trains a different encoder for different network vendors), or a vendor specific adaptation layer is used. However, in both cases, the UE part model may be reported back to the base station 204.

[0039] Returning to signal diagram 200 shown in FIG. 2, the NW may check 208 which of the model pairing information that the UE 202 supports, as provided in the UE capability report 206, is also supported by the current cell of the NW. This may, for example, determine which model pairing information both the UE and the network support. Further, the network may send RRC signaling, to the UE, including the model pairing information that is supported by both the UE and the network. The RRC signaling transmitted to the UE may take various forms. For example, the RRC signaling can be a CSI-reportConfig 210 (as shown in FIG. 2) or another RRC DL message. As an example, there may be multiple instances of model pairing information indicated in the UE capability report that the UE sends to the base station 204, and the base station 204 may choose all or multiple of the instances of model pairing information and send the supported pairing information through DL RRC messages. One example of DL RRC message is the CSI-reportConfig 210. In another example, the base station 204 may choose only one instance of model pairing information to include in the DL RRC message, such as CSI-reportConfig 210 to send the UE 202, and the UE 202 and base station 204 may utilize the one instance of model pairing information, which may end any further model alignment procedures as the UE and network models have been aligned.

[0040] For the first step, FIG. 3 illustrates an example signal diagram 300 of a UE capability report indicating support of Al based CSI compression according to another embodiment. In some situations, the UE 202 may implement a large number of models for different cells and for different configurations (i.e., a model for each various scenario). In such situations, the example shown in FIG. 2 may utilize a large amount of overhead as the UE 202 may need to send to the network the supported UE part model for each model. In view of this, in the example shown in FIG. 3. it may be assumed that a large number of models are trained for targeted scenarios, cells, and/or antenna configurations. A UE part model is either vendor specific (i.e., the UE trains a different encoder for different network vendors), or a vendor specific adaptation layer is used. However, in both cases, the UE part model may be reported back to the base station 204. [0041] In the example illustrated in the signal diagram 300 of FIG. 3, in the first step for model alignment, the UE 202 transmits a UE capability report 302 (e.g., as part of a capability inquiry/response) indicating support of an Al based CSI compression functionality to the base station 204. To avoid the large overhead, the UE capability report 302 does not include information for every UE-supported model pair. In response, the base station 204 sends an RRC configuration message 304 indicating a list of related model pairing information that the current base station 204 supports. As an example, if the base station 204 supports multiple instances of model pairing information, it may send the multiple supported instances of model pairing information in one RRC configuration list. In some embodiments, if an adaptation layer is used at the network side model to support different UE vendors, a sub-model ID can be included in the RRC configuration message 304 to indicate an adaptation layer when, for example, the pairing configuration information is provided in the form of a CSI reconstruction model ID that the network will use, the pairing information is provided in the form of a CSI generation model ID that the UE will use, or the pairing information is provided in the form of a paired CSI generation model and CSI reconstruction model ID.

[0042] After receiving the RRC configuration message from the network, the UE 202 checks 306 which model pairing information in the RRC configuration message 304 it supports. In response, the UE 202 sends an RRC reconfiguration complete message 308 back to the base station 204 indicating the UE-supported model pairing information based on the case used to define the pairing information, as described herein. In some embodiments, if an adaptation layer is used, a sub model ID may also be indicated in the RRC reconfiguration complete message 308 transmitted back to the base station 204. If, for example, there is only one instance of configuration information in the RRC reconfiguration complete message 308, the UE 202 and base station 204 may utilize the one instance of supported model pairing information, which may stop any further model alignment procedures as the UE and network models are aligned.

[0043] CSI-ReportConfig

[0044] For the second step, FIG. 4 illustrates an example signal diagram 400 of model alignment by sending pairing configuration information in a CSI-reportConfig signal according to one embodiment. If the first step 402 in model alignment, as described herein (see, e.g., FIG. 2 or FIG. 3), indicates only one instance of supported model pairing information, the model alignment procedure stops as the UE model and network model use the one indicated supported model pair. For example, the network may have one configured decoder for the UEs within the cell.

[0045] In some embodiments, as illustrated in signal diagram 400 of FIG. 4, if the first step 402 in model alignment indicates more than one instance of supported model pairing information, the base station 204 sends a CSI-reportConfig message 404 to the UE 202 including model pairing configuration information. The model pairing configuration information may be determined from a list of supported models in the first step 402. [0046] In certain embodiments, if an adaptation layer is used, or a different vector quantization (VQ) codebook is used to accommodate different CSI pay load sizes, the payload size can be signaled directly to the UE. This can, in some cases, implicitly indicate which adaptation layer to use. In some other cases, if rank specific models (i.e. , a different model for each rank) and/or layer specific models (i.e., a different model for each layer) are used, the different model pairing configuration information and corresponding rank indicator (RI) specific and/or layer specific information is included as part of the information in the CSI-reportConfig message 404 that the base station 204 sends to the UE 202 in the second step of model alignment.

[0047] FIG. 5 illustrates an example of a CSI-reportConfig message 500 with an additional entry 502 to provide Al based CSI compression related signaling according to one embodiment. For example, the entry 502 of the CSI-reportConfig message 500 may include an AiCSI-ReportConfig information element 504. The AiCSI-ReportConfig information element 504 may include, in some cases, a model pairing information list 506 that indicates a sequence of the model pairing information based on a maximum number of model pairs. The AiCSI-ReportConfig information element 504 may further include, in some cases, a RI restriction parameter 508 that includes information to assist in selecting the model pairing information when, for example, different models or different layers are used to generate different UCI bit sizes. The AiCSI-ReportConfig information element 504 may further include, in some cases, a quantization method parameter 510 that indicates which quantization method is used as a bit string such as a vector quantization method or a scalar quantization method. The AiCSI-ReportConfig information element 504 may further include, in some cases, a max payload size per UCI report parameter 512 that indicates a parameter for the size of the container that the UE can use based on rank per UCI report.

[0048] Skilled persons will recognize from the disclosure herein that the illustrated entry 502 of the CSI-reportConfig message 500 is not limited to the parameters described herein and may provide further information useful in model alignment and/or Al based CSI compression related signaling.

[0049] UCI. Report [0050] For the third step, FIG. 6 illustrates an example signal diagram 600 of model alignment by indicating a selected model pair in a UCI report 606 according to one embodiment. If the first step 602 in model alignment (see, e.g., FIG. 2 or FIG. 3) indicates only one instance of supported model pairing information, or if the CSI- reportConfig 604 of the second step (see, e.g., FIG. 4) indicates only one instance of model pairing configuration information, the model alignment procedure stops as the UE model and network model use the one indicated model pair. As an example, the procedure of model alignment may stop if the network specifies one payload per layer and the network and UE models may utilize that one payload per layer. As an example, in some use cases, the maximum rank is 2, and per layer payload is X bits, for a layer common model.

[0051] If, however, the first step 602 indicates more than one instance of supported model pairing information and the CSI-reportConfig 604 of the second step indicates more than one instance of model pairing configuration information, the UE 202 chooses a RI and a layer model and includes the information in a UCI report transmitted to the NW via the base station 204. In some embodiments, the RI and payload size per layer indication are included in a Part 1 of the UCI report.

[0052] In certain embodiments, the UE 202 indicates the RI and the payload size per layer separately from each other, which implicitly indicates the model pairing information. For example, indicating the RI and payload size per layer separately may be done where models can be uniquely indexed by a UCI size and a RI.

[0053] In other embodiments, the UE 202 indicates the model pairing information back to the base station 204 explicitly based on the UE’s choice of model. For example, if the base station 204 indicates two pairing models, one for indoor use and one for outdoor use, when the UE 202 chooses the indoor model, the UE 202 indicates the chosen indoor model in the UCI report. The RI and UCI bits can be further included in the UCI report for the indoor model (i.e., the RI and UCI bits may be further included in the UCI for the chosen model). Further, the model pairing information can be a short index based on the configuration of the second step of model alignment, as described herein. For example, if in the second step, two instances of model pairing information are configured, one bit is enough for the indication. In another example, if in the second step, four instances of model pairing information are configured, two bits are enough for the indication. [0054] In some embodiments, the explicit indication of the model pairing information and the RI and UCI bits are included of a Part 1 of the UCI report. A size of the Part 1 may be based on a UE capability report including the UE-supported model pairing information, and the size of the Part 1 may be indicated to the UE in a C SI report configuration. For example, when the UE-supported model pairing information includes a single model pair, the size of the Part 1 is reduced and the Part 1 includes the RI but not the indication of the selected model pair.

[0055] FIG. 7 illustrates a flowchart of a method 700 for a UE according to embodiments herein. The illustrated method 700 includes reporting 702, from the UE to a base station of a wireless network, a UE capability comprising an indication that the UE supports Al based CSI compression. The UE capability also comprises UE-supported model pairing information. The method 700 further includes receiving 704, at the UE from the base station, pairing configuration information corresponding to at least part of the UE-supported model pairing information. The method 700 further includes selecting 706, at the UE, based on the pairing configuration information, a selected model pair for an encoder at the UE corresponding to a decoder at the base station. The method 700 further includes generating 708, compressed CSI by providing DL channel data or a DL precoder to the encoder at the UE. The method 700 further includes transmitting 710, from the UE to the base station, an UCI report including the compressed CSI.

[0056] In some embodiments of the method 700, reporting the UE capability comprises: transmitting, in response to a UE capability inquiry from the base station, a UE capability report comprising the indication that the UE supports Al based CSI compression and the UE-supported model pairing information; and receiving, from the base station in response to the UE capability report, an RRC signal to confirm support by the wireless network for at least part of the UE-supported model pairing information. In some such embodiments, the RRC signal comprises a CSI-reportConfig message. In addition, or in other embodiments, the UE-supported model pairing information includes a respective sub-model ID to indicate an adaptation layer for each UE-supported model pair.

[0057] In some embodiments of the method 700, reporting the UE capability comprises: transmitting, in response to a UE capability inquiry from the base station, a UE capability' report comprising the indication that the UE supports Al based CSI compression; receiving, from the base station in response to the UE capability report, an RRC reconfiguration message indicating cell-supported model pairing information; determining, at the UE, the UE-supported model pairing information from UE-supported model pairs based on the cell-supported model pairing information; and transmitting, from the UE to the base station, an RRC reconfiguration complete message comprising the UE-supported model pairing information. In some such embodiments, the cell- supported model pairing information includes a respective sub-model ID to indicate an adaptation layer for each cell-supported model pair. The UE-supported model pairing information may also include the respective sub-model ID of the adaptation layer.

[0058] In some embodiments, the method 700 further comprises receiving the pairing configuration information in a CSI-reportConfig signal. In some such embodiments, the pairing configuration information includes a payload size to implicitly indicate an adaptation layer to use. In other embodiments, the pairing configuration information includes respective RI and layer specific information used to derive the UE-supported model pairing information.

[0059] In some embodiments of the method 700, the UCI report indicates the selected model pair to the base station. In some such embodiments, the UCI report indicates a RI and a payload size per layer separately to implicitly indicate the selected model pair. In other embodiments, the UCI report explicitly indicates the selected model pair using one or more bits. For example, the UCI report may include a RI and UCI bits for the selected model pair. In certain such embodiments, the UCI report is a UCI Part 1. In certain embodiments, one or more of the RI, the payload size per layer, and an indication of the selected model pair is in a Part 1 of the UCI report, and the compressed CSI is in a Part 2 of the UCI report. The size of the Part 1 may be based on a UE capability⁷ report comprising the UE-supported model pairing information, and the size of the Part 1 may be indicated to the UE in a CSI report configuration. For example, when the UE- supported model pairing information comprises a single model pair, the size of the Part 1 is reduced and the Part 1 includes the RI but not the indication of the selected model pair.

[0060] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 700. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1002 that is a UE, as described herein). [0061] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 700. This non-transitory computer-readable media may be, for example, a memory of a UE (such as a memory 1006 of a wireless device 1002 that is a UE, as described herein).

[0062] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 700. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1002 that is a UE, as described herein).

[0063] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 700. This apparatus may be. for example, an apparatus of a UE (such as a wireless device 1002 that is a UE. as described herein). [0064] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 700.

[0065] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processor is to cause the processor to carry out one or more elements of the method 700. The processor may be a processor of a UE (such as a processor(s) 1004 of a wireless device 1002 that is a UE, as described herein). These instructions may be, for example, located in the processor and/or on a memory of the UE (such as a memory 1006 of a wireless device 1002 that is a UE, as described herein).

[0066] FIG. 8 illustrates a flowchart of a method 800 for a base station according to embodiments herein. The illustrated method 800 includes receiving 802, from a UE at the base station, a UE capability comprising a first indication that the UE supports Al based CSI compression. The UE capability also comprises UE-supported model pairing information. The method 800 further includes transmitting 804, to the UE from the base station, pairing configuration information corresponding to at least part of the UE- supported model pairing information. The method 800 further includes receiving 806, from the UE at the base station, an UCI report including compressed CSI and a second indication of a selected model pair for an encoder at the UE corresponding to a decoder at the base station. [0067] In some embodiments of the method 800, receiving the UE capability comprises: receiving, in response to a UE capability inquiry from the base station, a UE capability report comprising the first indication that the UE supports Al based CSI compression and the UE-supported model pairing information; determining, at the base station, that the at least part of the UE-supported model pairing information is supported by the wireless network; and transmitting, from the base station in response to the UE capability report, an RRC signal to confirm network support by the wireless network for the at least part of the UE-supported model pairing information. In some such embodiments, the RRC signal comprises a CSI-reportConfig message. In addition, or in other embodiments, the UE-supported model pairing information includes a respective sub-model ID to indicate an adaptation layer for each UE-supported model pair.

[0068] In some embodiments of the method 800, reporting the UE capability comprises: receiving, in response to a UE capability inquiry from the base station, a UE capability report comprising the first indication that the UE supports Al based CSI compression; transmitting, from the base station in response to the UE capability report, an RRC reconfiguration message indicating cell-supported model pairing information; and receiving, from the UE at the base station, an RRC reconfiguration complete message comprising the UE-supported model pairing information. In some such embodiments, the cell-supported model pairing information includes a respective sub-model ID to indicate an adaptation layer for each cell-supported model pair. The UE-supported model pairing information may also include the respective sub-model ID of the adaptation layer.

[0069] In some embodiments, the method 800 further comprises transmitting the pairing configuration information in a CSI-reportConfig signal. In some such embodiments, the pairing configuration information includes a payload size to implicitly indicate an adaptation layer to use. In other embodiments, the pairing configuration information includes respective RI and layer specific information used to derive the UE- supported model pairing information.

[0070] In some embodiments of the method 800, the UCI report indicates a RI and a payload size per layer separately to implicitly indicate the selected model pair.

[0071] In some embodiments of the method 800, the UCI report explicitly indicates the selected model pair using one or more bits. In some such embodiments, the UCI report includes a RI and UCI bits for the selected model pair. [0072] In some embodiments of the method 800, one or more of the RI, the payload size per layer, and an indication of the selected model pair is in a Part 1 of the UCI report, and the compressed CSI is in a Part 2 of the UCI report. A size of the Part 1 may be based on a UE capability report including the UE-supported model pairing information, and the size of the Part 1 may be indicated to the UE in a CSI report configuration. For example, when the UE-supported model pairing information comprises a single model pair, the size of the Part 1 is reduced and the Part 1 includes the RI but not the indication of the selected model pair.

[0073] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 800. This apparatus may be. for example, an apparatus of a base station (such as a network device 1018 that is a base station, as described herein).

[0074] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 800. This non-transitory computer-readable media may be. for example, a memory of a base station (such as a memory 1022 of a network device 1018 that is a base station, as described herein).

[0075] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a base station (such as a network device 1018 that is a base station, as described herein).

[0076] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a base station (such as a network device 1018 that is a base station, as described herein).

[0077] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 800.

[0078] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processing element is to cause the processing element to carry out one or more elements of the method 800. The processor may be a processor of a base station (such as a processor(s) 1020 of a network device 1018 that is a base station, as described herein). These instructions may be, for example, located in the processor and/or on a memory of the base station (such as a memory 1022 of a network device 1018 that is a base station, as described herein).

[0079] FIG. 9 illustrates an example architecture of a wireless communication system 900, according to embodiments disclosed herein. The following description is provided for an example wireless communication system 900 that operates in conjunction with the LTE system standards and/or 5G or NR system standards as provided by 3GPP technical specifications.

[0080] As shown by FIG. 9, the wireless communication sy stem 900 includes UE 902 and UE 904 (although any number of UEs may be used). In this example, the UE 902 and the UE 904 are illustrated as smartphones (e.g., handheld touchscreen mobile computing devices connectable to one or more cellular networks), but may also comprise any mobile or non-mobile computing device configured for wireless communication.

[0081] The UE 902 and UE 904 may be configured to communicatively couple with a RAN 906. In embodiments, the RAN 906 may be NG-RAN, E-UTRAN, etc. The UE 902 and UE 904 utilize connections (or channels) (shown as connection 908 and connection 910, respectively) with the RAN 906, each of which comprises a physical communications interface. The RAN 906 can include one or more base stations (such as base station 912 and base station 914) that enable the connection 908 and connection 910.

[0082] In this example, the connection 908 and connection 910 are air interfaces to enable such communicative coupling, and may be consistent with RAT(s) used by the RAN 906, such as, for example, an LTE and/or NR.

[0083] In some embodiments, the UE 902 and UE 904 may also directly exchange communication data via a sidelink interface 916. The UE 904 is shown to be configured to access an access point (shown as AP 918) via connection 920. By way of example, the connection 920 can comprise a local wireless connection, such as a connection consistent with any IEEE 802.11 protocol, wherein the AP 918 may comprise a Wi-Fi® router. In this example, the AP 918 may be connected to another network (for example, the Internet) without going through a CN 924. [0084] In embodiments, the UE 902 and UE 904 can be configured to communicate using orthogonal frequency division multiplexing (OFDM) communication signals with each other or with the base station 912 and/or the base station 914 over a multicarrier communication channel in accordance with various communication techniques, such as, but not limited to, an orthogonal frequency division multiple access (OFDMA) communication technique (e.g., for downlink communications) or a single carrier frequency division multiple access (SC-FDMA) communication technique (e.g., for uplink and ProSe or sidelink communications), although the scope of the embodiments is not limited in this respect. The OFDM signals can comprise a plurality of orthogonal subcarriers.

[0085] In some embodiments, all or parts of the base station 912 or base station 914 may be implemented as one or more software entities running on server computers as part of a virtual network. In addition, or in other embodiments, the base station 912 or base station 914 may be configured to communicate with one another via interface 922. In embodiments where the wireless communication system 900 is an LTE system (e.g., when the CN 924 is an EPC), the interface 922 may be an X2 interface. The X2 interface may be defined between two or more base stations (e.g., two or more eNBs and the like) that connect to an EPC, and/or between two eNBs connecting to the EPC. In embodiments where the wireless communication system 900 is an NR system (e.g., when CN 924 is a 5GC), the interface 922 may be an Xn interface. The Xn interface is defined between two or more base stations (e.g., two or more gNBs and the like) that connect to 5GC. between a base station 912 (e.g.. a gNB) connecting to 5GC and an eNB, and/or between two eNBs connecting to 5GC (e.g.. CN 924).

[0086] The RAN 906 is shown to be communicatively coupled to the CN 924. The CN 924 may comprise one or more network elements 926, which are configured to offer various data and telecommunications services to customers/subscribers (e.g.. users of UE 902 and UE 904) who are connected to the CN 924 via the RAN 906. The components of the CN 924 may be implemented in one physical device or separate physical devices including components to read and execute instructions from a machine-readable or computer-readable medium (e.g., a non-transitory machine-readable storage medium).

[0087] In embodiments, the CN 924 may be an EPC, and the RAN 906 may be connected with the CN 924 via an S I interface 928. In embodiments, the S I interface 928 may be split into two parts, an SI user plane (S l-U) interface, which carries traffic data between the base station 912 or base station 914 and a serving gateway (S-GW), and the SI -MME interface, which is a signaling interface between the base station 912 or base station 914 and mobility management entities (MMEs).

[0088] In embodiments, the CN 924 may be a 5GC, and the RAN 906 may be connected with the CN 924 via an NG interface 928. In embodiments, the NG interface 928 may be split into two parts, an NG user plane (NG-U) interface, which carries traffic data between the base station 912 or base station 914 and a user plane function (UPF), and the SI control plane (NG-C) interface, which is a signaling interface between the base station 912 or base station 914 and access and mobility management functions (AMFs).

[0089] Generally , an application server 930 may be an element offering applications that use internet protocol (IP) bearer resources with the CN 924 (e.g., packet switched data services). The application server 930 can also be configured to support one or more communication services (e.g., VoIP sessions, group communication sessions, etc.) for the UE 902 and UE 904 via the CN 924. The application server 930 may communicate with the CN 924 through an IP communications interface 932.

[0090] FIG. 10 illustrates a system 1000 for performing signaling 1034 between a wireless device 1002 and a network device 1018, according to embodiments disclosed herein. The system 1000 may be a portion of a wireless communications system, as herein described. The wireless device 1002 may be, for example, a UE of a wireless communication system. The network device 1018 may be, for example, a base station (e.g., an eNB or a gNB) of a wireless communication system.

[0091] The wireless device 1002 may include one or more processor(s) 1004. The processor(s) 1004 may execute instructions such that various operations of the wireless device 1002 are performed, as described herein. The processor(s) 1004 may include one or more baseband processors implemented using, for example, a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.

[0092] The wireless device 1002 may include a memory 1006. The memory 1006 may be a non -transitory computer-readable storage medium that stores instructions 1008 (which may include, for example, the instructions being executed by the processor(s) 1004). The instructions 1008 may also be referred to as program code or a computer program. The memory’ 1006 may also store data used by. and results computed by. the processor(s) 1004.

[0093] The wireless device 1002 may include one or more transceiver(s) 1010 that may include radio frequency (RF) transmitter circuitry and/or receiver circuitry that use the antenna(s) 1012 of the wireless device 1002 to facilitate signaling (e.g., the signaling 1034) to and/or from the wireless device 1002 with other devices (e.g., the network device 1018) according to corresponding RATs.

[0094] The wireless device 1002 may include one or more antenna(s) 1012 (e.g., one, two, four, or more). For embodiments with multiple antenna(s) 1012, the wireless device 1002 may leverage the spatial diversity' of such multiple antenna(s) 1012 to send and/or receive multiple different data streams on the same time and frequency resources. This behavior may be referred to as. for example, multiple input multiple output (MIMO) behavior (referring to the multiple antennas used at each of a transmitting device and a receiving device that enable this aspect). MIMO transmissions by the wireless device 1002 may be accomplished according to precoding (or digital beamforming) that is applied at the wireless device 1002 that multiplexes the data streams across the antenna(s) 1012 according to known or assumed channel characteristics such that each data stream is received with an appropriate signal strength relative to other streams and at a desired location in the spatial domain (e.g., the location of a receiver associated with that data stream). Certain embodiments may use single user MIMO (SU-MIMO) methods (where the data streams are all directed to a single receiver) and/or multi user MIMO (MU-MIMO) methods (where individual data streams may’ be directed to individual (different) receivers in different locations in the spatial domain).

[0095] In certain embodiments having multiple antennas, the wireless device 1002 may implement analog beamforming techniques, whereby’ phases of the signals sent by the antenna(s) 1012 are relatively adjusted such that the (joint) transmission of the antenna(s) 1012 can be directed (this is sometimes referred to as beam steering).

[0096] The wireless device 1002 may include one or more interface(s) 1014. The interface(s) 1014 may be used to provide input to or output from the wireless device 1002. For example, a wireless device 1002 that is a UE may include interface(s) 1014 such as microphones, speakers, a touchscreen, buttons, and the like in order to allow for input and/or output to the UE by a user of the UE. Other interfaces of such a UE may be made up of transmitters, receivers, and other circuitry' (e.g., other than the transceiver(s) 1010/antenna(s) 1012 already described) that allow for communication between the UE and other devices and may operate according to known protocols (e.g., Wi-Fi®, Bluetooth®, and the like).

[0097] The wireless device 1002 may include a model alignment module 1016. The model alignment module 1016 may be implemented via hardware, software, or combinations thereof. For example, the model alignment module 1016 may be implemented as a processor, circuit, and/or instructions 1008 stored in the memory 1006 and executed by the processor(s) 1004. In some examples, the model alignment module 1016 may be integrated within the processor(s) 1004 and/or the transceiver(s) 1010. For example, the model alignment module 1016 may be implemented by a combination of software components (e.g., executed by a DSP or a general processor) and hardware components (e.g., logic gates and circuitry ) within the processor(s) 1004 or the transceiver(s) 1010.

[0098] The model alignment module 1016 may be used for various aspects of the present disclosure, for example, aspects of FIG. 1 to FIG. 7. The model alignment module 1016 is configured to align a UE model and a network model.

[0099] The network device 1018 may include one or more processor(s) 1020. The processor(s) 1020 may execute instructions such that various operations of the network device 1018 are performed, as described herein. The processor(s) 1020 may include one or more baseband processors implemented using, for example, a CPU, a DSP, an ASIC, a controller, an FPGA device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.

[0100] The network device 1018 may include a memory 1022. The memory 1022 may be a non -transitory computer-readable storage medium that stores instructions 1024 (which may include, for example, the instructions being executed by the processor(s) 1020). The instructions 1024 may also be referred to as program code or a computer program. The memory 1022 may also store data used by, and results computed by, the processor(s) 1020.

[0101] The network device 1018 may include one or more transceiver(s) 1026 that may include RF transmitter circuitry and/or receiver circuitry that use the antenna(s) 1028 of the network device 1018 to facilitate signaling (e.g., the signaling 1034) to and/or from the network device 1018 with other devices (e.g., the wireless device 1002) according to corresponding RATs.

[0102] The network device 1018 may include one or more antenna(s) 1028 (e.g., one, two, four, or more). In embodiments having multiple antenna(s) 1028, the network device 1018 may perform MIMO, digital beamforming, analog beamforming, beam steering, etc., as has been described.

[0103] The network device 1018 may include one or more interface(s) 1030. The interface(s) 1030 may be used to provide input to or output from the network device 1018. For example, a network device 1018 that is a base station may include interface(s) 1030 made up of transmitters, receivers, and other circuitry (e.g., other than the transceiver(s) 1026/antenna(s) 1028 already described) that enables the base station to communicate with other equipment in a core network, and/or that enables the base station to communicate with external networks, computers, databases, and the like for purposes of operations, administration, and maintenance of the base station or other equipment operably connected thereto.

[0104] The network device 1018 may include a model alignment module 1032. The model alignment module 1032 may be implemented via hardware, software, or combinations thereof. For example, the model alignment module 1032 may be implemented as a processor, circuit, and/or instructions 1024 stored in the memory 1022 and executed by the processor(s) 1020. In some examples, the model alignment module 1032 may be integrated within the processor(s) 1020 and/or the transceiver(s) 1026. For example, the model alignment module 1032 may be implemented by a combination of software components (e.g., executed by a DSP or a general processor) and hardware components (e.g., logic gates and circuitry) within the processor(s) 1020 or the transceiver(s) 1026.

[0105] The model alignment module 1032 may be used for various aspects of the present disclosure, for example, aspects of FIG. 1 to FIG. 6, and FIG. 8. The model alignment module 1032 is configured to align a UE model and a network model.

[0106] For one or more embodiments, at least one of the components set forth in one or more of the preceding figures may be configured to perform one or more operations, techniques, processes, and/or methods as set forth herein. For example, a baseband processor as described herein in connection with one or more of the preceding figures may be configured to operate in accordance with one or more of the examples set forth herein. For another example, circuitry associated with a UE, base station, network element, etc. as described above in connection with one or more of the preceding figures may be configured to operate in accordance with one or more of the examples set forth herein.

[0107] Any of the above described embodiments may be combined with any other embodiment (or combination of embodiments), unless explicitly stated otherwise. The foregoing description of one or more implementations provides illustration and description, but is not intended to be exhaustive or to limit the scope of embodiments to the precise form disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from practice of various embodiments.

[0108] Embodiments and implementations of the systems and methods described herein may include various operations, which may be embodied in machine-executable instructions to be executed by a computer system. A computer system may include one or more general-purpose or special-purpose computers (or other electronic devices). The computer system may include hardware components that include specific logic for performing the operations or may include a combination of hardware, software, and/or firmware.

[0109] It should be recognized that the systems described herein include descriptions of specific embodiments. These embodiments can be combined into single systems, partially combined into other systems, split into multiple systems or divided or combined in other ways. In addition, it is contemplated that parameters, attributes, aspects, etc. of one embodiment can be used in another embodiment. The parameters, attributes, aspects, etc. are merely described in one or more embodiments for clarity, and it is recognized that the parameters, attributes, aspects, etc. can be combined with or substituted for parameters, attributes, aspects, etc. of another embodiment unless specifically disclaimed herein.

[0110] It is well understood that the use of personally identifiable information should follow privacy policies and practices that are generally recognized as meeting or exceeding industry or governmental requirements for maintaining the privacy of users. In particular, personally identifiable information data should be managed and handled so as to minimize risks of unintentional or unauthorized access or use, and the nature of authorized use should be clearly indicated to users. [0111] Although the foregoing has been described in some detail for purposes of clarity, it will be apparent that certain changes and modifications may be made without departing from the principles thereof. It should be noted that there are many alternative ways of implementing both the processes and apparatuses described herein. Accordingly, the present embodiments are to be considered illustrative and not restrictive, and the description is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.

Claims

1. A method for a user equipment (UE), the method comprising: reporting, from the UE to a base station of a wireless network, a UE capability comprising an indication that the UE supports artificial intelligence (Al) based channel state information (CSI) compression and UE-supported model pairing information; receiving, at the UE from the base station, pairing configuration information corresponding to at least part of the UE-supported model pairing information; selecting, at the UE, based on the pairing configuration information, a selected model pair for an encoder at the UE corresponding to a decoder at the base station; generating compressed CSI by providing downlink (DL) channel data or a DL precoder to the encoder at the UE; and transmitting, from the UE to the base station, an uplink control information (UCI) report including the compressed CSI.

2. The method of claim 1, wherein reporting the UE capability comprises: transmitting, in response to a UE capability inquiry from the base station, a UE capability' report comprising the indication that the UE supports the Al based CSI compression and the UE-supported model pairing information; and receiving, from the base station in response to the UE capability report, a radio resource control (RRC) signal to confirm support by the wireless network for the at least part of the UE-supported model pairing information.

3. The method of claim 2, wherein the RRC signal comprises a CSI report configuration (CSI-reportConfig) message.

4. The method of claim 2, wherein the UE-supported model pairing information includes a respective sub-model identifier (ID) to indicate an adaptation layer for each UE- supported model pair.

5. The method of claim 1, wherein reporting the UE capability comprises: transmitting, in response to a UE capability inquiry from the base station, a UE capability report comprising the indication that the UE supports Al based CSI compression; receiving, from the base station in response to the UE capability report, a radio resource configuration (RRC) reconfiguration message indicating cell-supported model pairing information; determining, at the UE, the UE-supported model pairing information from UE- supported model pairs based on the cell-supported model pairing information; and transmitting, from the UE to the base station, an RRC reconfiguration complete message comprising the UE-supported model pairing information.

6. The method of claim 5, wherein the cell-supported model pairing information includes a respective sub-model identifier (ID) to indicate an adaptation layer for each cell- supported model pair.

7. The method of claim 6, wherein the UE-supported model pairing information includes the respective sub-model ID of the adaptation layer.

8. The method of claim 1, further comprising receiving the pairing configuration information in a C SI report configuration (CSI-reportConfig) signal.

9. The method of claim 8, wherein the pairing configuration information includes a payload size to implicitly indicate an adaptation layer to use.

10. The method of claim 8, wherein the pairing configuration information includes respective rank indicator (RI) and layer specific information used to derive the UE- supported model pairing information.

11. The method of claim 1, wherein the UCI report indicates the selected model pair to the base station.

12. The method of claim 11, wherein the UCI report indicates a rank indicator (RI) and a payload size per layer separately to implicitly indicate the selected model pair.

13. The method of claim 11, wherein the UCI report explicitly indicates the selected model pair using one or more bits.

14. The method of claim 13, wherein the UCI report includes a rank indicator (RI) and UCI bits for the selected model pair.

15. The method of any of claim 12 to claim 14, wherein one or more of the RI, the payload size per layer, and an indication of the selected model pair is in a Part 1 of the UCI report, and wherein the compressed CSI is in a Part 2 of the UCI report.

16. The method of claim 15, wherein a size of the Part 1 is based on a UE capability report comprising the UE-supported model pairing information, and wherein the size of the Part 1 is indicated to the UE in a CSI report configuration.

17. The method of claim 16, wherein when the UE-supported model pairing information comprises a single model pair, the size of the Part 1 is reduced and the Part 1 includes the RI but not the indication of the selected model pair.

18. A method for a base station of a wireless network, the method comprising: receiving, from a user equipment (UE) at the base station, a UE capability’ comprising a first indication that the UE supports artificial intelligence (Al) based channel state information (CSI) compression and UE-supported model pairing information; transmitting, to the UE from the base station, pairing configuration information corresponding to at least part of the UE-supported model pairing information; and receiving, from the UE at the base station, an uplink control information (UCI) report including compressed CSI and a second indication of a selected model pair for an encoder at the UE corresponding to a decoder at the base station.

19. The method of claim 18, wherein receiving the UE capability comprises: receiving, in response to a UE capability⁷ inquiry from the base station, a UE capability⁷ report comprising the first indication that the UE supports Al based CSI compression and the UE-supported model pairing information; determining, at the base station, that the at least part of the UE-supported model pairing information is supported by the wireless network; and transmitting, from the base station in response to the UE capability⁷ report, a radio resource control (RRC) signal to confirm network support by the wireless network for the at least part of the UE-supported model pairing information.

20. The method of claim 19, wherein the RRC signal comprises a CSI report configuration (CSI-reportConfig) message.

21. The method of claim 19, wherein the UE-supported model pairing information includes a respective sub-model identifier (ID) to indicate an adaptation layer for each UE-supported model pair.

22. The method of claim 18, wherein reporting the UE capability comprises: receiving, in response to a UE capability inquiry from the base station, a UE capability report comprising the first indication that the UE supports Al based CSI compression; transmitting, from the base station in response to the UE capability' report, a radio resource configuration (RRC) reconfiguration message indicating cell-supported model pairing information; and receiving, from the UE at the base station, an RRC reconfiguration complete message comprising the UE-supported model pairing information.

23. The method of claim 22, wherein the cell-supported model pairing information includes a respective sub-model identifier (ID) to indicate an adaptation layer for each cell-supported model pair.

24. The method of claim 23, wherein the UE-supported model pairing information includes the respective sub-model ID of the adaptation layer.

25. The method of claim 18, further comprising transmitting the pairing configuration information in a CSI report configuration (CSI-reportConfig) signal.

26. The method of claim 25, wherein the pairing configuration information includes a payload size to implicitly indicate an adaptation layer to use.

27. The method of claim 25, wherein the pairing configuration information includes respective rank indicator (RI) and layer specific information.

28. The method of claim 18, wherein the UCI report indicates a rank indicator (RI) and a payload size per layer separately to implicitly indicate the selected model pair.

29. The method of claim 18, wherein the UCI report explicitly indicates the selected model pair using one or more bits.

30. The method of claim 29, wherein the UCI report includes a rank indicator (RI) and UCI bits for the selected model pair.

31. The method of any of claim 27 to claim 30, wherein one or more of the RI, the payload size per layer, and an indication of the selected model pair is in a Part 1 of the UCI report, and wherein the compressed CSI is in a Part 2 of the UCI report.

32. The method of claim 31, wherein a size of the Part 1 is based on a UE capability report comprising the UE-supported model pairing information, and wherein the size of the Part 1 is indicated to the UE in a CSI report configuration.

33. The method of claim 32, wherein when the UE-supported model pairing information comprises a single model pair, the size of the Part 1 is reduced and the Part 1 includes the RI but not the indication of the selected model pair.

34. An apparatus comprising means to perform the method of any of claim 1 to claim 33.

35. A computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform the method of any of claim 1 to claim 33.

36. An apparatus comprising logic, modules, or circuitry to perform the method of any of claim 1 to claim 33.

37. A baseband processor for a user equipment (UE) that is configured to cause the UE to perform one or more elements of any one of claim 1 to claim 17.

38. A baseband processor for a base station that is configured to cause the base station to perform one or more elements of any one of claim 18 to claim 33.