Dynamic alignment of ai / ML-based CSI compression models

Abstract

Alignment of artificial intelligence (AI) / machine learning (ML)-based channel state information (CSI) compression models is discussed. A UE receives a data collection request including an associated identifier (ID) corresponding to network-side conditions. The UE generates, based on a measurement of one or more reference signals, CSI feedback based on the network-side conditions corresponding to the associated ID. The UE tags the CSI feedback with the associated ID and transmits, to the wireless network, a CSI report comprising the CSI feedback. The UE receives network-side model information marked with the associated ID. The network-side model information includes at least one of a model ID and a reference data set. The UE trains an AI / ML model, using the network-side model information to align an AI / ML CSI encoder model of the UE with an AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

Description

DYNAMIC ALIGNMENT OF AI / ML-BASED CSI COMPRESSION MODELSTECHNICAL FIELD

[0001] This application relates generally to wireless communication systems, including systems implementing artificial intelligence (AI) / machine learning (ML) models for channel state information (CSI) compression and decompression.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, 5GNR 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.14938-8423-3285' 1 P71483WO1

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

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

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

[0008] FIG. 1 is a diagram illustrating an example of a two-sided AI / ML model for CSI compression and decompression, according to embodiments herein.

[0009] FIG. 2 is a diagram illustrating a procedure for an inter-vendor training collaboration procedure for training an AI / ML model for CSI compression and decompression that relies on a model / parameter exchange from a base station to a UE.

[0010] FIG. 3 is a diagram illustrating an inter-vendor training collaboration procedure for training an AI / ML model for CSI compression and decompression that relies on a standardized dataset format and a dataset exchange from a base station to a UE.

[0011] FIG. 4 illustrates an example of an associated ID framework, according to embodiments herein.

[0012] FIG. 5 is a flow diagram of an example process of a model identification ID framework, according to embodiments herein.

[0013] FIG. 6 is a flowchart illustrating a method for a UE. according to embodiments herein.

[0014] FIG. 7 is a flowchart illustrating a method for a base station in a wireless network, according to embodiments herein.

[0015] FIG. 8 illustrates an example architecture of a wireless communication system, according to embodiments herein.24938-8423-3285' 1 P71483WO1

[0016] FIG. 9 illustrates a system for performing signaling between a wireless device and a network device, according to embodiments herein.DETAILED DESCRIPTION

[0017] 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.

[0018] Downlink channel state information (CSI) may be sent from a UE to a base station through feedback channels. The base station may use the CSI feedback to, for example, reduce interference and increase throughput for massive multiple input multiple output (MIMO) communication.

[0019] It may be understood that such feedback represents a relatively large amount of signaling overhead. In various wireless communication systems, vector quantization or codebook-based feedback may be used with the expectation of reducing this overhead. The feedback quantities resulting from these approaches, however, scale linearly with the number of transmit antennas. Accordingly, these approaches may still, for some cases (e.g., when hundreds or thousands of centralized or distributed transmit antennas are used) represent a high level of signaling overhead.

[0020] Accordingly, in various wireless communication systems, artificial intelligence (AI) / machine learning (ML)-based CSI encoding / compression and decoding / decompression mechanisms may be used to reduce signaling overhead associated with the transmission of CSI from the UE to the base station.

[0021] FIG. 1 is a diagram illustrating an example of a two-sided AI / ML model 102 for CSI compression and decompression, according to embodiments discussed herein. The AI / ML model 102 includes the (logical) UE side 104 (a portion of the AI / ML model 102 that exists at a UE 106) and the (logical) network side 108 (a portion of the AI / ML model 102 that exists at, for example, a base station of a network 110), as illustrated.

[0022] The UE side 104 of the AI / ML model 102 that operates at the UE 106 includes an encoder 112 (also referred to herein as a “CSI encoder’"). The encoder 112 is configured to accept an input 116 and to provide a bitstream 118 that is based on that34938-8423-3285' 1 P71483WO1input 116 as output. As illustrated, in some cases, the input 116 may include CSI, such as a raw downlink (DL) channel estimate or precoder information (such as a codebookbased indication of a precoder W and / or a set of eigenvectors corresponding to a precoder W).

[0023] The bitstream 118 is then transmitted from the UE 106 to the network 110.

[0024] The network side 108 of the AI / ML model 102 that operates at the network 110 (e.g., a base station of the network 110) includes a decoder 114 (also referred to herein as a "CSI decoder'). The decoder 114 is configured to accept the bitstream 118 as input and to decode information therein as output 120 to the network 110 for further processing. In this way, the information from the input 116 is made known to the network 110. Accordingly, in some cases, the output 120 may be understood to include CSI, such as a raw DL channel estimate or precoder information (such as a codebookbased indication of a precoder IT and / or a set of eigenvectors corresponding to a precoder IT), corresponding to the format of the input 116.

[0025] Based on the encoding mechanism used by the encoder 112, the bitstream 118 may be smaller than the raw or data as presented in the input 116. The encoder 112 may thus be understood to “compress” the input 116 into the bitstream, which is correspondingly understood to represent “compressed” information.

[0026] The result is that the transmission of the bitstream 118 to from the UE 106 to the network 110 results in the use of fewer radio resources than an alternative case where the input 116 is itself sent from the UE 106 to the network side 108 without such encoding / compression.

[0027] The output 120 may correspondingly be referred to variously herein as “decoded,” “decompressed,” “recovered,” etc.

[0028] Various methods for training and alignment of Al / ML models between a network and a UE may be referred to generally as collaboration directions. For example, in a collaboration Direction A, the network-side trains a decoder on dataset A, and the UE-side trains an encoder on dataset A, dataset B, or both dataset A and dataset B. In a collaboration Direction B, the network-side trains an encoder-decoder pair on dataset A and transfers decoder parameters to the UE. In a collaboration Direction C, fully standardized reference model(s) and parameters with specified CSI generation part and / or CSI reconstruction part are used for CSI generation.

[0029] CSI Inter-vendor Collaboration Examples (Direction A)44938-8423-3285' 1 P71483WO1

[0030] Various embodiments herein relate to mechanisms for AI / ML model training with the goal of developing a decoder / encoder pairing corresponding to an AI / ML model for CSI compression and decompression. In such training contexts, it will be understood that, in addition to an encoder at a UE and a decoder at the network, other entities associated with the AI / ML model could be used / developed in an intermediate fashion (e.g., to facilitate the training of the AI / ML model). For example, there may be one or more decoders at a UE and / or one or more encoders at a base station that are associated with training the AI / ML model (e g., as will now be discussed in FIG. 2 and FIG. 3).

[0031] FIG. 2 is a diagram illustrating a procedure 200 for an inter-vendor training collaboration procedure for training an AI / ML model for CSI compression and decompression that relies on a model / parameter exchange from a base station to a UE along with a target CSI. As illustrated, at the network 202, a joint training 206 for a network-side encoder Ei 208 and a network-side decoder Di 210 is performed. Then, the network 202 sends the encoder model / parameter(s) 212 to the UE 204 along with the target CSI.

[0032] Training at the UE 204 proceeds according to one of a first alternative 214 and a second alternative 216.

[0033] In the first alternative 214, which may be referred to generally as Direction A Option 3a-l Alternative 1, the UE 204 performs training in multiple steps. In a first step 224, the UE 204 trains a UE-side decoder D2218 using the model / parameter(s) 212 and the target CSI received from the network 202. Note that as part of this process, the UE-side decoder D2218 may have a same or a different structure than the structure of the network-side decoder Di 210. Then, in a second step 226, the UE 204 freezes the model for the UE-side decoder D2218 and performs a joint training procedure with the UE-side decoder D2218 to train a UE-side encoder E2220. Note that as part of this process, the UE-side encoder E2220 may have a same or a different structure than the structure of the network-side encoder Ei 208. The UE 204 may use its own dataset for this joint training. Inter-vendor compatibility may be achieved when the UE-side encoder E2220 performs the same as or better when tested with the network-side decoder Di 210.

[0034] In the second alternative 216, which may be referred to generally as Direction A Option 3a- 1 Alternative 2, upon receiving the model / parameter(s) 212 along with the target CSI. the UE 204 generates CSI feedback based on the model / parameter(s) 212 and the target CSI. The UE 204 trains a UE-side encoder E2222 according to the generated54938-8423-3285' 1 P71483WO1CSI feedback. Note that as part of this process, the UE-side encoder E2222 may have a same or a different structure than the structure of the network-side encoder Ei 208. Intervendor compatibility may be achieved when the UE-side encoder E2222 performs the same as or better when tested with the network-side decoder Di 210.

[0035] Accordingly, it may be understood that in cases corresponding to such examples, a network vendor first trains an AI / ML model (i.e. , CSI generation model (a network-side encoder Ei 208), CSI reconstruction model (network-side decoder Di 210)). Then, corresponding model / parameter(s) 212 is sent to a UE 204 using the standardized dataset format.

[0036] FIG. 3 is a diagram illustrating an inter-vendor training collaboration procedure 300 for training an AI / ML model for CSI compression and decompression that relies on a standardized dataset format and a dataset exchange from a base station to a UE. As illustrated, at the network 302. a joint training 306 for a network-side encoder Ei 308 and a network-side decoder Di 310 is performed.

[0037] Then, the network 302 sends a dataset 312 to the UE 304. The dataset 312 includes target CSI and CSI feedback information. The CSI feedback information may include model parameters for the AI / ML model at the network 302 (as represented by the network-side encoder Ei 308 and / or the network-side decoder Di 310). Note that the dataset 312 may be a partial dataset.

[0038] Training at the UE 304 then proceeds according to one of a first alternative 314 and a second alternative 316.

[0039] In the first alternative 314, which may be referred to generally as Direction A Option 4-1 Alternative 1, the UE 304 performs training in multiple steps. In a first step 324, the UE 304 trains a UE-side decoder D2318 using the dataset 312 (e.g., the target CSI and the CSI feedback information) received from the network 302. Note that as part of this process, the UE-side decoder D2318 may have a same or a different structure than the structure of the network-side decoder Di 310. Then, in a second step 326, the UE 304 freezes the model for the UE-side decoder D2318 and performs a joint training procedure with the UE-side decoder D2318 to train a UE-side encoder E2320. Note that as part of this process, the UE-side encoder E2320 may have a same or a different structure than the structure of the network-side encoder Ei 308. The UE 304 may use its own dataset for this joint training.64938-8423-3285' 1 P71483WO1

[0040] In a second alternative 316, which may be referred to generally as Direction A Option 4-1 Alternative 2, upon receiving the dataset 312 (e.g.. the target CSI and the CSI feedback information), the UE 304 uses the dataset 312 to train a UE-side encoder E2 322. Note that as part of this process, the UE-side encoder E2322 may have a same or a different structure than the structure of the network-side encoder Ei 308.

[0041] Accordingly, it may be understood that in cases corresponding to such examples, a network vendor first trains an AI / ML model (i.e., CSI generation model (a network-side encoder Ei 308), CSI reconstruction model (network-side decoder Di 310)). Then, a corresponding dataset 312 is sent to a UE 304 using the standardized dataset format.

[0042] Note that, corresponding to a case of AI / ML model use for active inferencing as was described according to the AI / ML model 102 described in FIG. 1, the UE-side encoder E2320 (in the case the first alternative 314 was used) or the UE-side encoder E2 322 (in the case the second alternative 316 was used) may be understood to correspond to the encoder 112, while the network-side decoder Di 310 may be understood to correspond to the decoder 114. The network-side encoder Ei 308 and the UE-side decoder D2318 (in the case that the first alternative 314 was used) may be understood to have been used for purposes other than active inferencing (e.g., training purposes, as just described) but not ultimately used corresponding to active inferencing using the AI / ML model.

[0043] Note that the notation for a network-side encoder (Ei), a network-side decoder (Di), a UE-side encoder (E2), and a UE-side decoder (D2) will be followed throughout this disclosure.

[0044] Associated ID Framework for Multiple Models

[0045] In certain embodiments, an associated identifier (ID) framework includes assigning unique IDs to predefined network-side additional conditions (e.g., beam configurations, antenna settings, and / or network state parameters). The associated IDs encode the network additional conditions and may be standardized and known to both the UE and the network. The associated ID framework acts as a shared reference to synchronize assumptions about network-side conditions during data collection, training, and inference.

[0046] However, for site specific models, different deployments may use different AI / ML models that may be more efficient and / or accurate depending on the74938-8423-3285' 1 P71483WO1implementation scenario. As a result, it may be beneficial to implement procedures for aligning encoders and decoders (e.g., Di (Ei). D2 (E2), ... Di (Ei)) of AI / ML models sited at the UE and AI / ML models sited at the network for various implementation scenarios. Certain embodiments herein introduce an associated ID framework for aligning the encoders and the decoders of UE-side AI / ML models and network-side AI / ML models for AI / ML-based CSI compression.

[0047] Alignment w ith Associated ID Framework

[0048] FIG. 4 illustrates an example of an associated ID framework, according to embodiments herein. For target CSI collection 402 (e.g., data collection), the network 408 signals 416 a configuration for data collection including the current associated ID to the UE 410 when requesting CSI reports. The UE 410 collects CSI data under the specific network conditions linked to the associated ID, ensuring alignment with the network's decoder training data. The UE 410 performs categorization and reporting by tagging collected target CSI data with the associated ID before reporting 418 to it to the network 408. This enables the netw ork 408 to map UE-reported data to the correct network-side conditions during model updates. Note that the configuration with the associate ID signaled by the network 408 may be stored at UE-side over-the-top (OTT) servers such as the UE-side OTT server 412 and the reported target CSI data with the associated data reported to the UE 410 may be stored at network-side OTT servers such as the netw ork-side OTT server 414.

[0049] In certain embodiments, for consistency across training and inference, during the training 404 phase, when the network 408 transfers 420 a reference AI / ML model (e.g., model parameters) and / or a dataset to the UE 410, it includes the associated ID used for training 404 the reference AI / ML model. Then, the UE 410 trains its local AI / ML model using data aligned with the associated ID’s conditions, maintaining alignment wdth the network’s decoder 424. Note that the transferred reference model may be stored at a UE-side OTT server 412.

[0050] During the inference 406 phase, the network 408 specifies 422 the associated ID to the UE 410, for example, in a CSI report configuration message (e.g., CSIReportConfig). This ensures that the UE 410 performs AI / ML model predictions by aligning the UE 410 encoder 426 with same associated ID (e.g., implies that the correct network-side assumptions are used) with the network's decoder 424.84938-8423-3285' 1 P71483WO1

[0051] For example, the network may signal an associated ID "X" to the UE during CSI data collection. Then, the UE collects CSI data under the conditions tied to the associated ID “X." The UE tags the data with the associated ID "X", and reports the CSI data to the network. Subsequently, the network trains its decoder using CSI data linked to associated ID “X”. The UE trains a local model using reference data and / or an AI / ML model that is sent from the network marked with the associated ID “‘X”. Then, during inference, the network instructs the UE to use associated ID “X,” ensuring consistency and encoder / decoder alignment. By centralizing coordination through the associated ID framework, both the UE and the network maintain aligned assumptions of network-side conditions / parameters, enabling robust two-sided CSI model training and reliable inference.

[0052] Model Identification ID Framework

[0053] In various embodiments, a model identification ID framework is used, for example, for aligning UE-sided AI / ML models and network-sided AI / ML models, life cycle management (LCM) procedures, and cooperative inference procedures. The model identification ID framework uses unique pair of model IDs that are known to both the UE and the network (e.g., the unique pair of model IDs are standardized). For example, globally standardized identifiers are assigned to CSI generation models (e.g., at the UE-side) and CSI reconstruction models (e.g., at the network-side). In certain embodiments, the model IDs encode various AI / ML model information such as a training collaboration direction (e.g.. Direction A, Direction B, Direction C), inter-vendor collaboration options (e.g.. Option 3a-l), AI / ML model architecture / parameters (e.g.. neural network structure, quantization), version for the AI / ML model updates, and / or an associated ID (e.g., encodes for the network side conditions and deployment / scenario). By way of example, the hierarchal structure of the model identification ID framework may take the form of: collaboration direction, training option, sub-option, model variant, and associated ID (e.g., Direction A / Option 3a- 1 / Variant 2 / vl 2 / Associated ID).

[0054] The model identification ID framework may be used in various alignment mechanisms. For example, in UE capability reporting (initial), a UE may signal a capability report that includes supported model IDs specifying, e g., supported collaboration directions (e.g., Direction A, Direction B, Direction C), training options / sub-options (e.g., federated learning, transfer learning), model variants (e.g.. lightweight vs. high-accuracy models), and / or an associated ID. As another example, the94938-8423-3285' 1 P71483WO1model identification framework may be used to perform another training collaboration, wherein the network signals the model ID and the UE trains its own encoder (model ID) and decoder (model ID). In another example, the model identification framework may be used in a network configuration for inference, wherein the network uses UE-reported IDs to match its CSI reconstruction model (e.g., decoder) with the UE’s CSI generation model and / or to configure LCM commands (e.g., activation / deactivation / switching). Note that the above example mechanisms may be implemented independently of each other or in combination with each other.

[0055] Lifecycle Management (LCM) Procedures

[0056] As indicated above, in certain embodiments, the model identification ID framework may be used for LCM procedures . For example, for model activation, the network signals a model ID (e.g., A / 3a-l / V2) to activate a UE-trained model and then the UE validates and switches to the specified model. In some embodiments, for AI / ML model fallback, if the UE fails to use the AI / ML model (e g., outdated version), the UE reverts to a default AI / ML model (e.g., a model corresponding to a predefined ID).

[0057] In certain embodiments, the model identification ID framework may be used for cooperative inference procedures. For example, for UE capability exchange, the UE reports supported model IDs during registration. For network configuration for inference, the network selects a compatible model ID (e.g., A / 3a-l / V2) and signals it to the UE. For model alignment, the UE loads the corresponding CSI generation model and the network loads the paired CSI reconstruction model. For inference execution, the UE compresses CSI using the active model and the network decodes the CSI using the matched model.

[0058] Example Use Case for Model ID Framework

[0059] By way of an example scenario, and not by limitation, a UE may support intervendor collaboration model Direction A, option 3a-l with two model variants: A / 3a-1 / V1 (e.g.. a baseline AI / ML model); and A / 3a-l / V2 (e.g.. AI / ML model optimized for high mobility). The network may detect that the UE is undergoing a high mobility scenario and may signal A / 3a-l / V2 for model activation. Accordingly, the network may configure its decoder to use the network-side model paired with model variant V2 (e.g., the model optimized for high mobility). As a result, improved CSI accuracy may be achieved as the UE is experiencing mobility and, accordingly, the AI / ML model selected / configured is optimized for high mobility.104938-8423-3285' 1 P71483WO1

[0060] As another example, an AI / ML model ID "A-3al -V2-X- 1.2" may be signaled from the network to the UE. In this example, "‘A” is the direction (e.g., UE-driven training), “3al” is the training option (e.g., federated learning with specific hyperparameters), “V2” is the model variant (e.g., optimized for mobility ), “X” is the associated ID (e.g., specific network-side conditions such as beam configuration, antenna settings, and / or network state parameters), andL‘1.2” is the version (e.g., AI / ML model / condition updates). In certain embodiments, the indicated model ID (e.g., model ID '‘A-3al-V2-X-1.2’’) may be linked to a network-side model (e.g., decoder), a UE-side model (e.g., encoder), and associated conditions. An example mapping table is shown in Table 1, wherein the associated conditions include a beam configuration X and a subcarrier spacing (SCS) of 30 kHz.Table 1 UE-Network Mapping Table

[0061] Example Model ID Framework Processes

[0062] In certain embodiments, in a training phase, a UE trains a model (e.g.. A-3al-V2) under network conditions tagged with Associated ID X. A model ID (e.g., A-3al-V2-X-1.0) is stored in the UE and network mapping tables. In an inference phase, the network signals the model ID (e.g., A-3al-V2-X-1.0) to the UE. The UE generates CSI using model A-3al-V2 while assuming network conditions X. The network decodes CSI using the paired decoder (e.g., NW-Decoder-A3al-V2) configured for conditions X. Certain such embodiments further include dynamic updating and / or training wherein the network deploys a new beam configuration (Associated ID Y), retrains and / or updates models to create a new Model ID (e.g., A-3al-V2-Y-l.l), and signals the new ID to UEs supporting Associated ID Y.

[0063] FIG. 5 is a flow diagram of an example process 500 of the model ID framework, according to embodiments herein. The process 500 begins with a network 502 signaling 506 a UE capability inquiry' to a UE 504. In response, the UE 504 transmits 508 its UE capability to the network 502. This may include an indication that the UE 504 supports CSI feedback based on UE-side offline engineering. The UE may include supported model IDs in its capability reporting to specify supported collaboration directions, training options / sub-options, model variants, and / or associated ID.114938-8423-3285' 1 P71483WO1

[0064] Based on the reported UE capability, the network 502 determines availability information of AI / ML model(s) (e.g., training collaboration method, information of model / parameter / dataset), which it signals 510 to the UE 504 in a network-side model ID. Based on the network-side model ID, the UE determines availability of UE-side model(s) (e.g., model / parameter ID). The UE 504 generates CSI data and tags the CSI data with the corresponding model IDs. The UE 504 then sends 512 the tagged CSI data to the network 502. In an inference phase, the network 502 sends 514 an inference configuration to the UE 504 that includes, e.g., selected AI / ML model(s), application duration, performance monitoring, method, etc. The UE 504 uses the inference configuration to perform inference and provide compressed CSI to the network 502.

[0065] FIG. 6 is a flowchart illustrating a method for a UE, according to embodiments herein. The illustrated method 600 includes receiving 602, from a wireless network, a data collection request including an associated ID corresponding to network-side conditions comprising one or more of a beam configuration, an antenna setting, and a network state parameter. The method 600 further includes generating 604, based on a measurement of one or more reference signals received from the wireless network, CSI feedback based on the network-side conditions corresponding to the associated ID. In tagging 606, method 600 tags the CSI feedback with the associated ID. The method 600 further includes transmitting 608, to the wireless network, a CSI report comprising the CSI feedback tagged with the associated ID. The method 600 further includes receiving 610, from the wireless network, network-side model information marked with the associated ID, the network-side model information comprising at least one of a model ID and a reference data set. The method 600 further includes training 612 an AI / ML model, at the UE, using the network-side model information to align an AI / ML CSI encoder model of the UE with an AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

[0066] In some embodiments, the method 600 further comprises receiving, from the wireless network, a CSI report configuration comprising the associated ID, in response to the CSI report configuration, performing inference to generate compressed CSI feedback using the AI / ML CSI encoder model of the UE under the network-side conditions corresponding to the associated ID, and reporting the compressed CSI feedback to the wireless network.124938-8423-3285' 1 P71483WO1

[0067] In some embodiments of the method 600, the model ID comprises one or more of a training collaboration direction, an inter-vendor collaboration option, AI / ML model parameters, AI / ML model update version, and the associated ID.

[0068] In some embodiments, the method 600 further comprises receiving, from the wireless network, a capability inquiry, and in response to the capability inquiry, transmitting a capability report to the wireless network comprising one or more supported model IDs indicating supported collaboration directions, training options, AI / ML model variants, and one or more supported associated IDs. Some such embodiments further comprise, in a cooperative inference procedure, receiving, from the wireless network, a compatible model ID corresponding a selected one of the one or more supported model IDs in the capability report, loading, as an active model, a CSI generation model corresponding to the compatible model ID, and compressing the CSI feedback using the active model.

[0069] In some embodiments, the method 600 further comprises, in a lifecycle management procedure receiving, from the wireless network, an indication to activate the AI / ML model, validating the network-side model information marked with the associated ID corresponding to the indication, and switching to the AI / ML model based on the network-side model information. Some such embodiments further comprise determining that the AI / ML model is outdated, and switching to a default AI / ML model corresponding to a predefined associated ID.

[0070] In some embodiments, the method 600 further comprises storing, at the UE, a mapping table based on the network-side model information marked with the associated ID, wherein the mapping table comprises the model ID, the AI / ML CSI encoder model of the UE, the AI / ML CSI decoder model of the wireless network, and the network-side conditions.

[0071] In some embodiments, the method 600 further comprises receiving, from the wireless network, an updated data collection request including an updated associated ID corresponding to updated network-side conditions, generating updated CSI feedback based on the updated network-side conditions corresponding to the updated associated ID, tagging the updated CSI feedback with the updated associated ID, transmitting, to the wireless network, an updated CSI report comprising the updated CSI feedback tagged with the updated associated ID, receiving, from the wireless network, updated networkside model information marked with the updated associated ID, the updated network-side134938-8423-3285' 1 P71483WO1model information comprising an updated model ID, and training an additional AI / ML model, at the UE, using the updated network-side model information to align an additional AI / ML CSI encoder model of the UE with an additional AI / ML CSI decoder model of the wireless network under the updated network-side conditions corresponding to the updated associated ID.

[0072] FIG. 7 is a flowchart illustrating a method 700 for a base station in a wireless network, according to embodiments herein. The illustrated method 700 includes transmitting 702, to a UE, a data collection request including an associated ID corresponding to network-side conditions comprising one or more of a beam configuration, an antenna setting, and a network state parameter. The method 700 further includes receiving 704, from the UE, CSI feedback based on the network-side conditions, wherein the CSI feedback is tagged with the associated ID. The method 700 further includes training 706 an AI / ML model, at the base station, using the CSI feedback tagged with the associated ID. The method 700 further includes transmitting 708, to the UE, network-side model information marked with the associated ID, the network-side model information comprising at least one of a model ID and a reference data set to align an AI / ML CSI encoder model of the UE with an AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

[0073] In some embodiments, the method 700 further comprises transmitting, to the UE, a CSI report configuration comprising the associated ID, receiving, in response to the CSI report configuration, compressed CSI feedback, and performing inference to decode the compressed CSI feedback using the AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

[0074] In some embodiments of the method 700, the model ID comprises one or more of a training collaboration direction, an inter-vendor collaboration option, AI / ML model parameters, AI / ML model update version, and the associated ID.

[0075] In some embodiments, the method 700 further comprises transmitting, to the UE, a capability inquiry, and receiving, from the UE in response to the capability inquiry', a capability report comprising one or more supported model IDs indicating supported collaboration directions, training options, AI / ML model variants, and one or more supported associated IDs. Some such embodiments further comprise, in a cooperative inference procedure selecting, a compatible model ID corresponding one of144938-8423-3285' 1 P71483WO1the one or more supported model IDs in the capability report, transmitting, to the UE, the compatible model ID. loading, as a paired model, a CSI reconstruction model corresponding to the compatible model ID, and decoding the CSI feedback using the paired model.

[0076] In some embodiments, the method 700 further comprises, in a lifecycle management procedure, transmitting, to the UE, an indication to activate a UE-trained AI / ML model.

[0077] In some embodiments, the method 700 further comprises storing, at the base station, a mapping table based on the network-side model information marked with the associated ID, wherein the mapping table comprises the model ID, the AI / ML CSI encoder model of the UE, the AI / ML CSI decoder model of the wireless network, and the network-side conditions.

[0078] In some embodiments, the method 700 further comprises transmitting, to the UE, an updated data collection request including an updated associated ID corresponding to updated network-side conditions, receiving, from the UE, an updated CSI report comprising updated CSI feedback based on the updated network-side conditions corresponding to the updated associated ID, wherein the updated CSI feedback is tagged with the updated associated ID, and transmitting, to the UE, updated network-side model information marked with the updated associated ID, the updated network-side model information comprising an updated model ID.

[0079] FIG. 8 illustrates an example architecture of a wireless communication system 800, according to embodiments disclosed herein. The following description is provided for an example wireless communication system 800 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. 8. the wireless communication system 800 includes UE 802 and UE 804 (although any number of UEs may be used). In this example, the UE 802 and the UE 804 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 802 and UE 804 may be configured to communicatively couple with a RAN 806. In embodiments, the RAN 806 may be NG-RAN, E-UTRAN, etc. The UE 802 and UE 804 utilize connections (or channels) (shown as connection 808 and connection 154938-8423-3285' 1 P71483WO1810, respectively) with the RAN 806, each of which comprises a physical communications interface. The RAN 806 can include one or more base stations (such as base station 812 and base station 814) that enable the connection 808 and connection 810.

[0082] In this example, the connection 808 and connection 810 are air interfaces to enable such communicative coupling, and may be consistent with RAT(s) used by the RAN 806, such as, for example, an LTE and / or NR.

[0083] In some embodiments, the UE 802 and UE 804 may also directly exchange communication data via a sidelink interface 816. The UE 804 is shown to be configured to access an access point (shown as AP 818) via connection 820. By way of example, the connection 820 can comprise a local wireless connection, such as a connection consistent with any IEEE 802.11 protocol, wherein the AP 818 may comprise a Wi-Fi® router. In this example, the AP 818 may be connected to another network (for example, the Internet) without going through a CN 824.

[0084] In embodiments, the UE 802 and UE 804 can be configured to communicate using orthogonal frequency division multiplexing (OFDM) communication signals with each other or with the base station 812 and / or the base station 814 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 plurality7of orthogonal subcarriers.

[0085] In some embodiments, all or parts of the base station 812 or base station 814 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 812 or base station 814 may be configured to communicate with one another via interface 822. In embodiments where the wireless communication system 800 is an LTE system (e.g., when the CN 824 is an EPC), the interface 822 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 800 is an NR system (e.g., when164938-8423-3285' 1 P71483WO1CN 824 is a 5GC), the interface 822 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 812 (e.g., a gNB) connecting to 5GC and an eNB, and / or between two eNBs connecting to 5GC (e.g., CN 824).

[0086] The RAN 806 is shown to be communicatively coupled to the CN 824. The CN 824 may comprise one or more network elements 826, which are configured to offer various data and telecommunications services to customers / subscribers (e.g., users of UE 802 and UE 804) who are connected to the CN 824 via the RAN 806. The components of the CN 824 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 824 may be an EPC, and the RAN 806 may be connected with the CN 824 via an S I interface 828. In embodiments, the S I interface 828 may be split into two parts, an SI user plane (Sl-U) interface, which carries traffic data between the base station 812 or base station 814 and a serving gateway (S-GW), and the SI -MME interface, which is a signaling interface between the base station 812 or base station 814 and mobility management entities (MMEs).

[0088] In embodiments, the CN 824 may be a 5GC, and the RAN 806 may be connected with the CN 824 via an NG interface 828. In embodiments, the NG interface 828 may be split into two parts, an NG user plane (NG-U) interface, which carries traffic data between the base station 812 or base station 814 and a user plane function (UPF), and the SI control plane (NG-C) interface, which is a signaling interface between the base station 812 or base station 814 and access and mobility management functions (AMFs).

[0089] Generally, an application server 830 may be an element offering applications that use internet protocol (IP) bearer resources with the CN 824 (e.g., packet switched data services). The application server 830 can also be configured to support one or more communication services (e.g., VoIP sessions, group communication sessions, etc.) for the UE 802 and UE 804 via the CN 824. The application server 830 may communicate with the CN 824 through an IP communications interface 832.

[0090] FIG. 9 illustrates a system 900 for performing signaling 934 between a wireless device 902 and a network device 918, according to embodiments disclosed herein. The system 900 may be a portion of a wireless communications system as herein described.174938-8423-3285' 1 P71483WO1The wireless device 902 may be, for example, a UE of a wireless communication system. The network device 918 may be. for example, a base station (e.g.. an eNB or a gNB) of a wireless communication system.

[0091] The wireless device 902 may include one or more processor(s) 904. The processor(s) 904 may execute instructions such that various operations of the wireless device 902 are performed, as described herein. The processor(s) 904 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 902 may include a memory 906. The memory 906 may be a non-transitory computer-readable storage medium that stores instructions 908 (which may include, for example, the instructions being executed by the processor(s) 904). The instructions 908 may also be referred to as program code or a computer program. The memory 906 may also store data used by, and results computed by, the processor(s) 904.

[0093] The wireless device 902 may include one or more transceiver(s) 910 that may include radio frequency (RF) transmitter circuitry' and / or receiver circuitry that use the antenna(s) 912 of the wireless device 902 to facilitate signaling (e.g., the signaling 934) to and / or from the wireless device 902 with other devices (e.g., the network device 918) according to corresponding RATs.

[0094] The wireless device 902 may include one or more antenna(s) 912 (e.g., one, two, four, or more). For embodiments with multiple antenna(s) 912, the wireless device 902 may leverage the spatial diversity of such multiple antenna(s) 912 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 902 may be accomplished according to precoding (or digital beamforming) that is applied at the wireless device 902 that multiplexes the data streams across the antenna(s) 912 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 data184938-8423-3285' 1 P71483WO1stream). 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 902 may implement analog beamforming techniques, whereby phases of the signals sent by the antenna(s) 912 are relatively adjusted such that the (joint) transmission of the antenna(s) 912 can be directed (this is sometimes referred to as beam steering).

[0096] The wireless device 902 may include one or more interface(s) 914. The interface(s) 914 may be used to provide input to or output from the wireless device 902. For example, a wireless device 902 that is a UE may include interface(s) 914 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) 910 / antenna(s) 912 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 902 may include a CSI module 916. The CSI module 916 may be implemented via hardware, software, or combinations thereof. For example, the CSI module 916 may be implemented as a processor, circuit, and / or instructions 908 stored in the memory 906 and executed by the processor(s) 904. In some examples, the CSI module 916 may be integrated within the processor(s) 904 and / or the transceiver(s) 910. For example, the CSI module 916 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) 904 or the transceiver(s) 910.

[0098] The CSI module 916 may be used for various aspects of the present disclosure, for example, aspects of FIG. 1, FIG. 2, FIG. 3, FIG. 4, FIG. 5 and FIG. 6. The CSI module 916 is configured to cause the wireless device 902 to receive, from a network device 918, a data collection request including an associated ID corresponding to network-side conditions comprising one or more of a beam configuration, an antenna setting, and a network state parameter. The CSI module 916 is further configured to cause the wireless device 902 to generate, based on a measurement of one or more194938-8423-3285' 1 P71483WO1reference signals received from the network device 918, CSI feedback based on the network-side conditions corresponding to the associated ID. The CSI module 916 is further configured to cause the network device 918 to tag the CSI feedback with the associated ID. The CSI module 916 is further configured to cause the wireless device 902 to transmit, to the network device 918, a CSI report comprising the CSI feedback tagged with the associated ID. The CSI module 916 is further configured to cause the wireless device 902 to receive, from the network device 918, network-side model information marked with the associated ID, the network-side model information comprising at least one of a model ID and a reference data set. The CSI module 916 is further configured to cause the wireless device 902 to train an AI / ML model, at the wireless device 902, using the network-side model information to align an AI / ML CSI encoder model of the wireless device 902 with an AI / ML CSI decoder model of the network device 918 under the network-side conditions corresponding to the associated ID.

[0099] The network device 918 may include one or more processor(s) 920. The processor(s) 920 may execute instructions such that various operations of the network device 918 are performed, as described herein. The processor(s) 920 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 918 may include a memory 922. The memory 922 may be a non-transitory computer-readable storage medium that stores instructions 924 (which may include, for example, the instructions being executed by the processor(s) 920). The instructions 924 may also be referred to as program code or a computer program. The memory 922 may also store data used by, and results computed by, the processor(s) 920.

[0101] The network device 918 may include one or more transceiver(s) 926 that may include RF transmitter circuitry and / or receiver circuitry7that use the antenna(s) 928 of the network device 918 to facilitate signaling (e.g., the signaling 934) to and / or from the network device 918 with other devices (e.g., the wireless device 902) according to corresponding RATs.

[0102] The network device 918 may include one or more antenna(s) 928 (e.g., one, two, four, or more). In embodiments having multiple antenna(s) 928. the network device 918204938-8423-3285' 1 P71483WO1may perform MIMO, digital beamforming, analog beamforming, beam steering, etc., as has been described.

[0103] The network device 918 may include one or more interface(s) 930. The interface(s) 930 may be used to provide input to or output from the network device 918. For example, a network device 918 that is a base station may include interface(s) 930 made up of transmitters, receivers, and other circuitry (e.g., other than the transceiver(s) 926 / antenna(s) 928 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 918 may include a CSI module 932. The CSI module 932 may be implemented via hardware, software, or combinations thereof. For example, the CSI module 932 may be implemented as a processor, circuit, and / or instructions 924 stored in the memory 922 and executed by the processor(s) 920. In some examples, the CSI module 932 may be integrated within the processor(s) 920 and / or the transceiver(s) 926. For example, the CSI module 932 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) 920 or the transceiver(s) 926.

[0105] The CSI module 932 may be used for various aspects of the present disclosure, for example, aspects of FIG. 1, FIG. 2, FIG. 3, FIG. 4, FIG. 5 and FIG. 7. The CSI module 932 is configured to cause the network device 918 to transmit, to a wireless device 902, a data collection request including an associated ID corresponding to network-side conditions comprising one or more of a beam configuration, an antenna setting, and a network state parameter. The CSI module 932 is further configured to cause the network device 918 to receive, from the wireless device 902, CSI feedback based on the network-side conditions, wherein the CSI feedback is tagged with the associated ID. The CSI module 932 is further configured to cause the network device 918 to train an AI / ML model, at the base station, using the CSI feedback tagged with the associated ID. The CSI module 932 is further configured to cause the network device 918 to transmit, to the wireless device 902, network-side model information marked with the associated ID, the network-side model information comprising at least one of a214938-8423-3285' 1 P71483WO1model ID and a reference data set to align an AI / ML CSI encoder model of the wireless device 902 with an AI / ML CSI decoder model of the network device 918 under the network-side conditions corresponding to the associated ID.

[0106] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 600. This apparatus may be. for example, an apparatus of a UE (such as a wireless device 902 that is a UE, as described herein).

[0107] 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 600. This non-transitory computer-readable media may be, for example, a memory of a UE (such as a memory 906 of a wireless device 902 that is a UE, as described herein).

[0108] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 600. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 902 that is a UE, as described herein).

[0109] 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 600. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 902 that is a UE, as described herein).

[0110] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 600.[OHl] 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 600. The processor may be a processor of a UE (such as a processor(s) 904 of a wireless device 902 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 906 of a wireless device 902 that is a UE, as described herein).

[0112] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 700. This apparatus may be. for example,224938-8423-3285' 1 P71483WO1an apparatus of a base station (such as a network device 918 that is a base station, as described herein).

[0113] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 700. This non-transitory computer-readable media may be, for example, a memory of a base station (such as a memory 922 of a network device 918 that is a base station, as described herein).

[0114] 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 base station (such as a network device 918 that is a base station, as described herein).

[0115] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 700. This apparatus may be. for example, an apparatus of a base station (such as a network device 918 that is a base station, as described herein).

[0116] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 700.

[0117] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processing element is to cause the processing element to carry out one or more elements of the method 700. The processor may be a processor of a base station (such as a processor(s) 920 of a network device 918 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 922 of a network device 918 that is a base station, as described herein).

[0118] 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 234938-8423-3285' 1 P71483WO1herein. 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.

[0119] 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.

[0120] 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.

[0121] 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.

[0122] 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.244938-8423-3285' 1 P71483WO1

[0123] 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.254938-8423-3285' 1 P71483WO1

Claims

CLAIMS1. A method for a user equipment (UE), the method comprising:receiving, from a wireless network, a data collection request including an associated identifier (ID) corresponding to network-side conditions comprising one or more of a beam configuration, an antenna setting, and a network state parameter:generating, based on a measurement of one or more reference signals received from the wireless network, channel state information (CSI) feedback based on the network-side conditions corresponding to the associated ID;tagging the CSI feedback with the associated ID;transmitting, to the wireless network, a CSI report comprising the CSI feedback tagged with the associated ID;receiving, from the wireless network, network-side model information marked with the associated ID, the network-side model information comprising at least one of a model ID and a reference data set; andtraining an artificial intelligence (AI) / machine learning (ML) model, at the UE, using the network-side model information to align an AI / ML CSI encoder model of the UE with an AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

2. The method of claim 1, further comprising:receiving, from the wireless network, a CSI report configuration comprising the associated ID;in response to the CSI report configuration, performing inference to generate compressed CSI feedback using the AI / ML CSI encoder model of the UE under the network-side conditions corresponding to the associated ID; andreporting the compressed CSI feedback to the wireless network.

3. The method of claim 1, wherein the model ID comprises one or more of a training collaboration direction, an inter-vendor collaboration option, AI / ML model parameters, AI / ML model update version, and the associated ID.

4. The method of claim 1, further comprising:receiving, from the wireless network, a capability inquiry'; and264938-8423-3285' 1 P71483WO1in response to the capability inquiry, transmitting a capability' report to the wireless network comprising one or more supported model IDs indicating supported collaboration directions, training options, AI / ML model variants, and one or more supported associated IDs.

5. The method of claim 4, further comprising, in a cooperative inference procedure: receiving, from the wireless network, a compatible model ID corresponding a selected one of the one or more supported model IDs in the capability' report;loading, as an active model, a CSI generation model corresponding to the compatible model ID; andcompressing the CSI feedback using the active model.

6. The method of claim 1, further comprising, in a lifecycle management procedure: receiving, from the wireless network, an indication to activate the AI / ML model; validating the network-side model information marked with the associated ID corresponding to the indication; andswitching to the AI / ML model based on the network-side model information.

7. The method of claim 6, further comprising:determining that the AI / ML model is outdated; andswitching to a default AI / ML model corresponding to a predefined associated ID.

8. The method of claim 1, further comprising storing, at the UE. a mapping table based on the network-side model information marked with the associated ID, wherein the mapping table comprises the model ID, the AI / ML CSI encoder model of the UE, the AI / ML CSI decoder model of the wireless network, and the network-side conditions.

9. The method of claim 1, further comprising:receiving, from the wireless network, an updated data collection request including an updated associated ID corresponding to updated network-side conditions;generating updated CSI feedback based on the updated network-side conditions corresponding to the updated associated ID;tagging the updated CSI feedback with the updated associated ID; transmitting, to the wireless network, an updated CSI report comprising the updated CSI feedback tagged with the updated associated ID;274938-8423-3285' 1 P71483WO1receiving, from the wireless network, updated network-side model information marked with the updated associated ID, the updated network-side model information comprising an updated model ID; andtraining an additional AI / ML model, at the UE, using the updated network-side model information to align an additional AI / ML CSI encoder model of the UE with an additional AI / ML CSI decoder model of the wireless network under the updated network-side conditions corresponding to the updated associated ID.

10. A method for a base station in a wireless network, the method comprising:transmitting, to a user equipment (UE), a data collection request including an associated identifier (ID) corresponding to network-side conditions comprising one or more of a beam configuration, an antenna setting, and a network state parameter;receiving, from the UE, channel state information (CSI) feedback based on the network-side conditions, wherein the CSI feedback is tagged with the associated ID; training an artificial intelligence (AI) / machine learning (ML) model, at the base station, using the CSI feedback tagged with the associated ID; andtransmitting, to the UE, network-side model information marked with the associated ID, the network-side model information comprising at least one of a model ID and a reference data set to align an AI / ML CSI encoder model of the UE with an AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

11. The method of claim 10, further comprising:transmitting, to the UE. a CSI report configuration comprising the associated ID; receiving, in response to the CSI report configuration, compressed CSI feedback; andperforming inference to decode the compressed CSI feedback using the AI / ML CSI decoder model of the wireless network under the network-side conditions corresponding to the associated ID.

12. The method of claim 10, wherein the model ID comprises one or more of a training collaboration direction, an inter-vendor collaboration option, AI / ML model parameters, AI / ML model update version, and the associated ID.

13. The method of claim 10, further comprising:284938-8423-3285' 1 P71483WO1transmitting, to the UE, a capability inquiry; andreceiving, from the UE in response to the capability inquiry, a capability report comprising one or more supported model IDs indicating supported collaboration directions, training options, AI / ML model variants, and one or more supported associated IDs.

14. The method of claim 13, further comprising, in a cooperative inference procedure:selecting, a compatible model ID corresponding one of the one or more supported model IDs in the capability7report;transmitting, to the UE, the compatible model ID;loading, as a paired model, a CSI reconstruction model corresponding to the compatible model ID; anddecoding the CSI feedback using the paired model.

15. The method of claim 10, further comprising, in a lifecycle management procedure, transmitting, to the UE, an indication to activate a UE-trained AI / ML model.

16. The method of claim 10, further comprising storing, at the base station, a mapping table based on the network-side model information marked with the associated ID. wherein the mapping table comprises the model ID, the AI / ML CSI encoder model of the UE, the AI / ML CSI decoder model of the wireless network, and the network-side conditions.

17. The method of claim 10, further comprising:transmitting, to the UE, an updated data collection request including an updated associated ID corresponding to updated network-side conditions;receiving, from the UE, an updated CSI report comprising updated CSI feedback based on the updated network-side conditions corresponding to the updated associated ID, wherein the updated CSI feedback is tagged with the updated associated ID; and transmitting, to the UE, updated network-side model information marked with the updated associated ID, the updated network-side model information comprising an updated model ID.

18. An apparatus comprising means to perform the method of any one of claim 1 to claim 17.294938-8423-3285' 1 P71483WO119. 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 one of claim 1 to claim 17.

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

21. 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 9.

22. 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 10 to claim 17.304938-8423-3285' 1 P71483WO1

Images