Over-the-air signaling enhancements for data collection for two-sided model training
OTA signaling techniques streamline AI/ML model training and deployment in wireless communications systems, addressing computational complexity and enhancing channel state feedback performance.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-09-26
- Publication Date
- 2026-04-09
AI Technical Summary
Existing wireless communications systems face challenges in efficiently training, transferring, and deploying AI/ML models for channel state feedback, leading to high computational complexity and suboptimal performance.
Implementing over-the-air (OTA) signaling techniques for AI/ML model transfer and data collection, enabling efficient training, collection, and deployment of AI/ML models for channel state feedback, including the use of UE and network entities to signal model information and training data to a server for model training and deployment.
Reduces computational complexity and enhances the performance of AI/ML-based CSI processing by facilitating efficient AI/ML model training, collection, and deployment, thereby improving the accuracy and efficiency of channel state feedback.
Smart Images

Figure CN2025124205_09042026_PF_FP_ABST
Abstract
Description
OVER-THE-AIR SIGNALING ENHANCEMENTS FOR DATA COLLECTION FOR TWO-SIDED MODEL TRAININGCross-Reference to Related Application (s)
[0001] This application claims benefit of and priority to International Patent Cooperation Treaty Application No. PCT / CN2024 / 123195, filed October 3, 2024, which is hereby assigned to the assignee hereof and hereby expressly incorporated by reference herein in its entirety as if fully set forth below and for all applicable purposes. Field of the Disclosure
[0002] Aspects of the present disclosure relate to wireless communications, and more particularly, to techniques for artificial intelligence (AI) and / or machine learning (ML) model transfer and data collection via over-the-air (OTA) signaling. Description of Related Art
[0003] Wireless communications systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, broadcasts, or other similar types of services. These wireless communications systems may employ multiple-access technologies capable of supporting communications with multiple users by sharing available wireless communications system resources with those users.
[0004] Although wireless communications systems have made great technological advancements over many years, challenges still exist. For example, complex and dynamic environments can still attenuate or block signals between wireless transmitters and wireless receivers. Accordingly, there is a continuous desire to improve the technical performance of wireless communications systems, including, for example: improving speed and data carrying capacity of communications, improving efficiency of the use of shared communications mediums, reducing power used by transmitters and receivers while performing communications, improving reliability of wireless communications, avoiding redundant transmissions and / or receptions and related processing, improving the coverage area of wireless communications, increasing the number and types of devices that can access wireless communications systems, increasing the ability for different types of devices to intercommunicate, increasing the number and type of wireless communications mediums available for use, and the like. Consequently, there exists a need for further improvements in wireless communications systems to overcome the aforementioned technical challenges and others.SUMMARY
[0005] One aspect provides a method for wireless communication by a user equipment (UE) . The method includes transmitting signaling to a network entity, the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity to a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information.
[0006] Another aspect provides a method for wireless communication by a network entity. The method includes receiving signaling from a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity and a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information; and transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, to the server based on the received information from the UE.
[0007] Another aspect provides a method for wireless communication by a training server. The method includes transmitting signaling to a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination thereof, with the training server; receiving channel state feedback (CSF) information training data; receiving the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, from a network entity; training, based on the CSF information training data and on the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with a CSF information encoder model, or a trained parameter set associated with a CSF information decoder model, or any combination thereof; and providing, to the UE, the trained CSF information encoder model, the trained CSF information decoder model, the trained parameter set associated with the CSF information encoder model, or the trained parameter set associated with the CSF information decoder model, or the combination thereof.
[0008] Another aspect provides a method for wireless communication by an operations, administration, and maintenance (OAM) entity. The method includes receiving signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, with a server that is configured to train one or more user equipment (UE) machine learning (ML) models for channel state feedback (CSF) information; and transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the server based on the received information from the UE.
[0009] Other aspects provide: an apparatus operable, configured, or otherwise adapted to perform any one or more of the aforementioned methods and / or those described elsewhere herein; a non-transitory, computer-readable media comprising instructions that, when executed (e.g., directly, indirectly, after pre-processing, without pre-processing) by one or more processors of an apparatus, cause the apparatus to perform the aforementioned methods as well as those described elsewhere herein; a computer program product embodied on a computer-readable storage medium comprising code for performing the aforementioned methods as well as those described elsewhere herein; and / or an apparatus comprising means for performing the aforementioned methods as well as those described elsewhere herein. By way of example, an apparatus may comprise a processing system, a device with a processing system, or processing systems cooperating over one or more networks.
[0010] The following description and the appended figures set forth certain features for purposes of illustration.BRIEF DESCRIPTION OF DRAWINGS
[0011] The appended figures depict certain features of the various aspects described herein and are not to be considered limiting of the scope of this disclosure.
[0012] FIG. 1 depicts an example wireless communications network.
[0013] FIG. 2 depicts an example disaggregated base station architecture.
[0014] FIG. 3 depicts aspects of an example base station and an example user equipment.
[0015] FIGS. 4A, 4B, 4C, and 4D depict various example aspects of data structures for a wireless communications network.
[0016] FIG. 5 depicts a general functional framework applied for artificial intelligence-enabled radio access network (RAN) intelligence.
[0017] FIG. 6 depicts an example of a machine learning based channel state information (CSI) feedback mechanism.
[0018] FIG. 7 is a table depicting an example of encoder input and decoder output for the ML-based CSI feedback mechanism of FIG. 6.
[0019] FIG. 8 depicts an example system for model training with a single training entity.
[0020] FIG. 9 depicts an example system for model training with separate training entities.
[0021] FIG. 10 depicts another example system for model training with separate training entities.
[0022] FIG. 11 depicts an example system for two-sided model training with a standardized reference model.
[0023] FIG. 12 depicts an example system for two-sided model training with a standardized dataset.
[0024] FIG. 13 depicts standardized signaling for a plug-and-play approach to developing a model.
[0025] FIG. 14 depicts standardized signaling for an offline engineering approach to developing a model.
[0026] FIG. 15 depicts a table depicting example configurations for first and second options for data collection and transfer for a two-side model.
[0027] FIG. 16 depicts a table depicting example configurations for third and fourth options for data collection and transfer for a two-side model.
[0028] FIG. 17 is a call flow diagram illustrating example operations for providing model sharing information from a UE to the network.
[0029] FIG. 18 is a call flow diagram illustrating example operations for signaling a parameter set, training data, and / or reference model from the network side to the UE side.
[0030] FIG. 19 is a call flow diagram illustrating example operations for retrieving a parameter set, training data, and / or reference model from the network-side.
[0031] FIG. 20 is a call flow diagram illustrating example operations for retrieving UE-side training data for encoder training.
[0032] FIG. 21 is a call flow diagram illustrating example operations for signaling a parameter set, training data, and / or reference model to the UE-side.
[0033] FIG. 22 is a call flow diagram illustrating example operations for configuring a base station in an open radio access network (O-RAN) to report a parameter set, training data, and / or reference model.
[0034] FIG. 23 is another call flow diagram illustrating example operations for configuring a base station in an O-RAN to report a parameter set, training data, and / or reference model.
[0035] FIG. 24 depicts a method for wireless communications by a user equipment.
[0036] FIG. 25 depicts a method for wireless communications by a network entity.
[0037] FIG. 26 depicts a method for wireless communications by a training server.
[0038] FIG. 27 depicts a method for wireless communications by an Operations, Administration, and Maintenance (OAM) entity.
[0039] FIG. 28 depicts aspects of an example communications device.
[0040] FIG. 29 depicts aspects of another example communications device.
[0041] FIG. 30 depicts aspects of yet another example communications device.
[0042] FIG. 31 depicts aspects of yet another example communications device.DETAILED DESCRIPTION
[0043] Aspects of the present disclosure provide apparatuses, methods, processing systems, and computer-readable mediums for artificial intelligence (AI) and / or machine learning (ML) model transfer and data collection via over-the-air (OTA) signaling.
[0044] Understanding the channel state between devices communicating in a wireless communications system is one aspect of improving the performance of wireless communications. Various techniques have been employed for measuring the channel state and for reporting feedback to improve the performance. The feedback may include channel station information (CSI) , which is information regarding the channel properties (or channel conditions) of a communication link. In some examples, the CSI may indicate how a signal propagates from a transmitter to the receiver and represents the combined effect of various phenomena, such as scattering, fading, and power decay with distance.
[0045] CSI feedback reporting may be based on a codebook. The codebook may be used as a precoding matrix indicator (PMI) dictionary from which a user equipment (UE) may report the best PMI codewords for a given channel condition, and use a sequence of bits to report the PMI. In other words, rather than feedback actual values as PMI codewords, a UE may just feedback a set of bits that represents an entry into the codebook. Codebook based CSI reporting reduces signaling overhead. Upon receiving the feedback, the network entity may retrieve the PMI codewords from the codebook. Such techniques are often relatively slow, power hungry, and static in approach.
[0046] AI / ML represents an opportunity to potentially improve techniques for measuring channel state and reporting feedback. ML generally refers to a subset of AI that involves algorithms and models that enable computers / processors to learn from and make predictions or decisions based on data. ML typically focuses on creating systems that can improve their performance on a specific task by recognizing patterns and making adjustments through iterative learning, without being explicitly programmed. ML is used in various applications, including image and speech recognition, recommendation systems, and predictive analytics.
[0047] AI / ML may be deployed to perform certain functions in certain wireless communications systems, such as acquisition and feedback of CSI (collectively referred to herein as CSI processing) . For example, AI / ML models may reduce the number of resource elements needed for estimating a channel state (reducing signaling overhead) , and improve the estimates of values used in reporting the estimated channel state (e.g., by training AI / ML models to improve accuracy of channel estimation) . In some cases, AI / ML-based CSI feedback may replace codebook-based processing by a CSI encoder (also known as a CSI generation model) , a CSI decoder (also known as a CSI reconstruction model) , or a combination thereof. In such cases, the CSI encoder may be analogous to the PMI algorithm, while the CSI decoder may be analogous to the PMI codebook used to translate the CSI reporting bits to a PMI codeword. Thus, a CSI encoder at the UE may be trained to compress the channel estimate to a few bits that are then reported to a network entity (e.g., a base station) , while the CSI decoder at the network entity is trained to recover the channel or the precoding matrix using the reported bits.
[0048] However, while AI / ML models may be trained to perform many functions related to channel estimation and feedback, one potential challenge to the use of such models relates to deriving and communicating relations and characteristics for a set of AI / ML models supported by UEs and / or network entities. For example, there may be a significant amount of complexity involved in AI / ML model training, AI / ML model data collection, AI / ML model transferring, and / or AI / ML model deployment. It can be a significant burden to train, transfer, and deploy an AI / ML-based CSI processing model for each communication scenario and configuration.
[0049] Aspects of the present disclosure provide techniques and apparatus for OTA signaling that can be used for various aspects of AI / ML model training, AI / ML model data collection, AI / ML model transferring, and / or AI / ML model deployment. The techniques described herein may help alleviate (or reduce) computational complexity involved in AI / ML model training, AI / ML model data collection, AI / ML model transferring, and / or AI / ML model deployment, thereby improving overall performance of AI / ML-based CSI processing models.
[0050] As used herein, the term “model” may be used to refer to an AI model, a ML model, or a combination thereof. Similarly, as used herein, the term “encoder” may also be referred to as a CSI generation model or a CSI encoder, the term “decoder” may also be referred to as a CSI reconstruction model or a CSI decoder, and the term “encoder-decoder pair” may also be referred to as a CSI generation-reconstruction pair or combined encoder-decoder. As used herein, the term “CSI processing” may be used to refer to one or more (or a combination of) aspects involved in the acquisition (or estimation) of CSI and feedback of CSI.
[0051] According to certain aspects, the UE may signal a location of a server where model information can be sent to and retrieved from. Model information may include model parameter sets, training datasets, and / or reference models.
[0052] According to certain aspects, the UE may signal one or more identifiers (IDs) associated with the model information that may be used for retrieving the model information.
[0053] In some aspects, models are trained at the server based on the model information and training data. In some aspects, the model information is shared to the server from a base station, an operations, administration, and maintenance (OAM) entity, and / or a network data analytics function (NWDAF) of a core network. In some aspects, the training data is shared to the server from the UE or the base station.
[0054] In some aspects, models are trained at the NWDAF and exposed to the server. The NWDAF may obtain the model information and the training data from the UE, the base station, and / or the OAM. Introduction to Wireless Communications Networks
[0055] The techniques and methods described herein may be used for various wireless communications networks. While aspects may be described herein using terminology commonly associated with 3G, 4G, and / or 5G wireless technologies, aspects of the present disclosure may likewise be applicable to other communications systems and standards not explicitly mentioned herein.
[0056] FIG. 1 depicts an example of a wireless communications network 100, in which aspects described herein may be implemented.
[0057] Generally, wireless communications network 100 includes various network entities (alternatively, network elements or network nodes) . A network entity is generally a communications device and / or a communications function performed by a communications device (e.g., a user equipment (UE) , a base station (BS) , a component of a BS, a server, etc. ) . For example, various functions of a network as well as various devices associated with and interacting with a network may be considered network entities. Further, wireless communications network 100 includes terrestrial aspects, such as ground-based network entities (e.g., BSs 102) , and non-terrestrial aspects, such as satellite 140 and aircraft 145, which may include network entities on-board (e.g., one or more BSs) capable of communicating with other network elements (e.g., terrestrial BSs) and user equipments.
[0058] In the depicted example, wireless communications network 100 includes BSs 102, UEs 104, and one or more core networks, such as an Evolved Packet Core (EPC) 160 and 5G Core (5GC) network 190, which interoperate to provide communications services over various communications links, including wired and wireless links.
[0059] FIG. 1 depicts various example UEs 104, which may more generally include: a cellular phone, smart phone, session initiation protocol (SIP) phone, laptop, personal digital assistant (PDA) , satellite radio, global positioning system, multimedia device, video device, digital audio player, camera, game console, tablet, smart device, wearable device, vehicle, electric meter, gas pump, large or small kitchen appliance, healthcare device, implant, sensor / actuator, display, internet of things (IoT) devices, always on (AON) devices, edge processing devices, or other similar devices. UEs 104 may also be referred to more generally as a mobile device, a wireless device, a wireless communications device, a station, a mobile station, a subscriber station, a mobile subscriber station, a mobile unit, a subscriber unit, a wireless unit, a remote unit, a remote device, an access terminal, a mobile terminal, a wireless terminal, a remote terminal, a handset, and others.
[0060] BSs 102 wirelessly communicate with (e.g., transmit signals to or receive signals from) UEs 104 via communications links 120. The communications links 120 between BSs 102 and UEs 104 may include uplink (UL) (also referred to as reverse link) transmissions from a UE 104 to a BS 102 and / or downlink (DL) (also referred to as forward link) transmissions from a BS 102 to a UE 104. The communications links 120 may use multiple-input and multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming, and / or transmit diversity in various aspects.
[0061] BSs 102 may generally include: a NodeB (NB) , enhanced NodeB (eNB) , next generation enhanced NodeB (ng-eNB) , next generation NodeB (gNB or gNodeB) , access point, base transceiver station, radio base station, radio transceiver, transceiver function, transmission reception point, and / or others. Each of BSs 102 may provide communications coverage for a respective geographic coverage area 110, which may sometimes be referred to as a cell, and which may overlap in some cases (e.g., small cell 102’ may have a coverage area 110’ that overlaps the coverage area 110 of a macro cell) . A BS may, for example, provide communications coverage for a macro cell (covering relatively large geographic area) , a pico cell (covering relatively smaller geographic area, such as a sports stadium) , a femto cell (relatively smaller geographic area (e.g., a home) ) , and / or other types of cells.
[0062] While BSs 102 are depicted in various aspects as unitary communications devices, BSs 102 may be implemented in various configurations. For example, one or more components of a base station may be disaggregated, including a central unit (CU) , one or more distributed units (DUs) , one or more radio units (RUs) , a Near-Real Time (Near-RT) RAN Intelligent Controller (RIC) , or a Non-Real Time (Non-RT) RIC, to name a few examples. In another example, various aspects of a base station may be virtualized. More generally, a base station (e.g., BS 102) may include components that are located at a single physical location or components located at various physical locations. In examples in which a base station includes components that are located at various physical locations, the various components may each perform functions such that, collectively, the various components achieve functionality that is similar to a base station that is located at a single physical location. In some aspects, a base station including components that are located at various physical locations may be referred to as a disaggregated radio access network architecture, such as an Open RAN (O-RAN) or Virtualized RAN (VRAN) architecture. FIG. 2 depicts and describes an example disaggregated base station architecture.
[0063] Different BSs 102 within wireless communications network 100 may also be configured to support different radio access technologies, such as 3G, 4G, and / or 5G. For example, BSs 102 configured for 4G LTE (collectively referred to as Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN) ) may interface with the EPC 160 through first backhaul links 132 (e.g., an S1 interface) . BSs 102 configured for 5G (e.g., 5G new radio (NR) or Next Generation RAN (NG-RAN) ) may interface with 5GC 190 through second backhaul links 184. BSs 102 may communicate directly or indirectly (e.g., through the EPC 160 or 5GC 190) with each other over third backhaul links 134 (e.g., X2 interface) , which may be wired or wireless.
[0064] Wireless communications network 100 may subdivide the electromagnetic spectrum into various classes, bands, channels, or other features. In some aspects, the subdivision is provided based on wavelength and frequency, where frequency may also be referred to as a carrier, a subcarrier, a frequency channel, a tone, or a subband. For example, the 3rd Generation Partnership Project (3GPP) currently defines Frequency Range 1 (FR1) as including 410 megahertz (MHz) –7125 MHz, which is often referred to (interchangeably) as “Sub-6 GHz” . Similarly, 3GPP currently defines Frequency Range 2 (FR2) as including 24, 250 MHz –71,000 MHz, which is sometimes referred to (interchangeably) as a “millimeter wave” ( “mmW” or “mmWave” ) . In some cases, FR2 may be further defined in terms of sub-ranges, such as a first sub-range FR2-1 including 24,250 MHz –52, 600 MHz and a second sub-range FR2-2 including 52, 600 MHz –71,000 MHz. A base station configured to communicate using mmWave / near mmWave radio frequency bands (e.g., a mmWave base station such as BS 180) may utilize beamforming (e.g., 182) with a UE (e.g., 104) to improve path loss and range.
[0065] The communications links 120 between BSs 102 and, for example, UEs 104, may be through one or more carriers, which may have different bandwidths (e.g., 5, 10, 15, 20, 100, 400, and / or other MHz) , and which may be aggregated in various aspects. Carriers may or may not be adjacent to each other. Allocation of carriers may be asymmetric with respect to DL and UL (e.g., more or fewer carriers may be allocated for DL than for UL) .
[0066] Communications using higher frequency bands may have higher path loss and a shorter range compared to lower frequency communications. Accordingly, certain base stations (e.g., 180 in FIG. 1) may utilize beamforming 182 with a UE 104 to improve path loss and range. For example, BS 180 and the UE 104 may each include a plurality of antennas, such as antenna elements, antenna panels, and / or antenna arrays to facilitate the beamforming. In some cases, BS 180 may transmit a beamformed signal to UE 104 in one or more transmit directions 182’ . UE 104 may receive the beamformed signal from the BS 180 in one or more receive directions 182” . UE 104 may also transmit a beamformed signal to the BS 180 in one or more transmit directions 182” . BS 180 may also receive the beamformed signal from UE 104 in one or more receive directions 182’ . BS 180 and UE 104 may then perform beam training to determine the best receive and transmit directions for each of BS 180 and UE 104. Notably, the transmit and receive directions for BS 180 may or may not be the same. Similarly, the transmit and receive directions for UE 104 may or may not be the same.
[0067] Wireless communications network 100 further includes a Wi-Fi AP 150 in communication with Wi-Fi stations (STAs) 152 via communications links 154 in, for example, a 2.4 GHz and / or 5 GHz unlicensed frequency spectrum.
[0068] Certain UEs 104 may communicate with each other using device-to-device (D2D) communications link 158. D2D communications link 158 may use one or more sidelink channels, such as a physical sidelink broadcast channel (PSBCH) , a physical sidelink discovery channel (PSDCH) , a physical sidelink shared channel (PSSCH) , a physical sidelink control channel (PSCCH) , and / or a physical sidelink feedback channel (PSFCH) .
[0069] EPC 160 may include various functional components, including: a Mobility Management Entity (MME) 162, other MMEs 164, a Serving Gateway 166, a Multimedia Broadcast Multicast Service (MBMS) Gateway 168, a Broadcast Multicast Service Center (BM-SC) 170, and / or a Packet Data Network (PDN) Gateway 172, such as in the depicted example. MME 162 may be in communication with a Home Subscriber Server (HSS) 174. MME 162 is the control node that processes the signaling between the UEs 104 and the EPC 160. Generally, MME 162 provides bearer and connection management.
[0070] Generally, user Internet protocol (IP) packets are transferred through Serving Gateway 166, which itself is connected to PDN Gateway 172. PDN Gateway 172 provides UE IP address allocation as well as other functions. PDN Gateway 172 and the BM-SC 170 are connected to IP Services 176, which may include, for example, the Internet, an intranet, an IP Multimedia Subsystem (IMS) , a Packet Switched (PS) streaming service, and / or other IP services.
[0071] BM-SC 170 may provide functions for MBMS user service provisioning and delivery. BM-SC 170 may serve as an entry point for content provider MBMS transmission, may be used to authorize and initiate MBMS Bearer Services within a public land mobile network (PLMN) , and / or may be used to schedule MBMS transmissions. MBMS Gateway 168 may be used to distribute MBMS traffic to the BSs 102 belonging to a Multicast Broadcast Single Frequency Network (MBSFN) area broadcasting a particular service, and / or may be responsible for session management (start / stop) and for collecting eMBMS related charging information.
[0072] 5GC 190 may include various functional components, including: an Access and Mobility Management Function (AMF) 192, other AMFs 193, a Session Management Function (SMF) 194, and a User Plane Function (UPF) 195. AMF 192 may be in communication with Unified Data Management (UDM) 196.
[0073] AMF 192 is a control node that processes signaling between UEs 104 and 5GC 190. AMF 192 provides, for example, quality of service (QoS) flow and session management.
[0074] Internet protocol (IP) packets are transferred through UPF 195, which is connected to the IP Services 197, and which provides UE IP address allocation as well as other functions for 5GC 190. IP Services 197 may include, for example, the Internet, an intranet, an IMS, a PS streaming service, and / or other IP services.
[0075] In various aspects, a network entity or network node can be implemented as an aggregated base station, as a disaggregated base station, a component of a base station, an integrated access and backhaul (IAB) node, a relay node, a sidelink node, to name a few examples.
[0076] FIG. 2 depicts an example disaggregated base station 200 architecture. The disaggregated base station 200 architecture may include one or more central units (CUs) 210 that can communicate directly with a core network 220 via a backhaul link, or indirectly with the core network 220 through one or more disaggregated base station units (such as a Near-Real Time (Near-RT) RAN Intelligent Controller (RIC) 225 via an E2 link, or a Non-Real Time (Non-RT) RIC 215 associated with a Service Management and Orchestration (SMO) Framework 205, or both) . A CU 210 may communicate with one or more distributed units (DUs) 230 via respective midhaul links, such as an F1 interface. The DUs 230 may communicate with one or more radio units (RUs) 240 via respective fronthaul links. The RUs 240 may communicate with respective UEs 104 via one or more radio frequency (RF) access links. In some implementations, the UE 104 may be simultaneously served by multiple RUs 240.
[0077] Each of the units, e.g., the CUs 210, the DUs 230, the RUs 240, as well as the Near-RT RICs 225, the Non-RT RICs 215 and the SMO Framework 205, may include one or more interfaces or be coupled to one or more interfaces configured to receive or transmit signals, data, or information (collectively, signals) via a wired or wireless transmission medium. Each of the units, or an associated processor or controller providing instructions to the communications interfaces of the units, can be configured to communicate with one or more of the other units via the transmission medium. For example, the units can include a wired interface configured to receive or transmit signals over a wired transmission medium to one or more of the other units. Additionally or alternatively, the units can include a wireless interface, which may include a receiver, a transmitter or transceiver (such as a radio frequency (RF) transceiver) , configured to receive or transmit signals, or both, over a wireless transmission medium to one or more of the other units.
[0078] In some aspects, the CU 210 may host one or more higher layer control functions. Such control functions can include radio resource control (RRC) , packet data convergence protocol (PDCP) , service data adaptation protocol (SDAP) , or the like. Each control function can be implemented with an interface configured to communicate signals with other control functions hosted by the CU 210. The CU 210 may be configured to handle user plane functionality (e.g., Central Unit –User Plane (CU-UP) ) , control plane functionality (e.g., Central Unit –Control Plane (CU-CP) ) , or a combination thereof. In some implementations, the CU 210 can be logically split into one or more CU-UP units and one or more CU-CP units. The CU-UP unit can communicate bidirectionally with the CU-CP unit via an interface, such as the E1 interface when implemented in an O-RAN configuration. The CU 210 can be implemented to communicate with the DU 230, as necessary, for network control and signaling.
[0079] The DU 230 may correspond to a logical unit that includes one or more base station functions to control the operation of one or more RUs 240. In some aspects, the DU 230 may host one or more of a radio link control (RLC) layer, a medium access control (MAC) layer, and one or more high physical (PHY) layers (such as modules for forward error correction (FEC) encoding and decoding, scrambling, modulation and demodulation, or the like) depending, at least in part, on a functional split, such as those defined by the 3GPP. In some aspects, the DU 230 may further host one or more low PHY layers. Each layer (or module) can be implemented with an interface configured to communicate signals with other layers (and modules) hosted by the DU 230, or with the control functions hosted by the CU 210.
[0080] Lower-layer functionality can be implemented by one or more RUs 240. In some deployments, an RU 240, controlled by a DU 230, may correspond to a logical node that hosts RF processing functions, or low-PHY layer functions (such as performing fast Fourier transform (FFT) , inverse FFT (iFFT) , digital beamforming, physical random access channel (PRACH) extraction and filtering, or the like) , or both, based at least in part on the functional split, such as a lower layer functional split. In such an architecture, the RU (s) 240 can be implemented to handle over the air (OTA) communications with one or more UEs 104. In some implementations, real-time and non-real-time aspects of control and user plane communications with the RU (s) 240 can be controlled by the corresponding DU 230. In some scenarios, this configuration can enable the DU (s) 230 and the CU 210 to be implemented in a cloud-based RAN architecture, such as a vRAN architecture.
[0081] The SMO Framework 205 may be configured to support RAN deployment and provisioning of non-virtualized and virtualized network elements. For non-virtualized network elements, the SMO Framework 205 may be configured to support the deployment of dedicated physical resources for RAN coverage requirements which may be managed via an operations and maintenance interface (such as an O1 interface) . For virtualized network elements, the SMO Framework 205 may be configured to interact with a cloud computing platform (such as an open cloud (O-Cloud) 290) to perform network element life cycle management (such as to instantiate virtualized network elements) via a cloud computing platform interface (such as an O2 interface) . Such virtualized network elements can include, but are not limited to, CUs 210, DUs 230, RUs 240 and Near-RT RICs 225. In some implementations, the SMO Framework 205 can communicate with a hardware aspect of a 4G RAN, such as an open eNB (O-eNB) 211, via an O1 interface. Additionally, in some implementations, the SMO Framework 205 can communicate directly with one or more RUs 240 via an O1 interface. The SMO Framework 205 also may include a Non-RT RIC 215 configured to support functionality of the SMO Framework 205.
[0082] The Non-RT RIC 215 may be configured to include a logical function that enables non-real-time control and optimization of RAN elements and resources, Artificial Intelligence / Machine Learning (AI / ML) workflows including model training and updates, or policy-based guidance of applications / features in the Near-RT RIC 225. The Non-RT RIC 215 may be coupled to or communicate with (such as via an A1 interface) the Near-RT RIC 225. The Near-RT RIC 225 may be configured to include a logical function that enables near-real-time control and optimization of RAN elements and resources via data collection and actions over an interface (such as via an E2 interface) connecting one or more CUs 210, one or more DUs 230, or both, as well as an O-eNB, with the Near-RT RIC 225.
[0083] In some implementations, to generate AI / ML models to be deployed in the Near-RT RIC 225, the Non-RT RIC 215 may receive parameters or external enrichment information from external servers. Such information may be utilized by the Near-RT RIC 225 and may be received at the SMO Framework 205 or the Non-RT RIC 215 from non-network data sources or from network functions. In some examples, the Non-RT RIC 215 or the Near-RT RIC 225 may be configured to tune RAN behavior or performance. For example, the Non-RT RIC 215 may monitor long-term trends and patterns for performance and employ AI / ML models to perform corrective actions through the SMO Framework 205 (such as reconfiguration via O1) or via creation of RAN management policies (such as A1 policies) .
[0084] FIG. 3 depicts aspects of an example BS 102 and a UE 104.
[0085] Generally, BS 102 includes various processors (e.g., 320, 330, 338, and 340) , antennas 334a-t (collectively 334) , transceivers 332a-t (collectively 332) , which include modulators and demodulators, and other aspects, which enable wireless transmission of data (e.g., data source 312) and wireless reception of data (e.g., data sink 339) . For example, BS 102 may send and receive data between BS 102 and UE 104. BS 102 includes controller / processor 340, which may be configured to implement various functions described herein related to wireless communications.
[0086] Generally, UE 104 includes various processors (e.g., 358, 364, 366, and 380) , antennas 352a-r (collectively antennas 352) , transceivers 354a-r (collectively transceivers 354) , which include modulators and demodulators, and other aspects, which enable wireless transmission of data (e.g., retrieved from data source 362) and wireless reception of data (e.g., provided to data sink 360) . UE 104 includes controller / processor 380, which may be configured to implement various functions described herein related to wireless communications.
[0087] In regards to an example downlink transmission, BS 102 includes a transmit processor 320 that may receive data from a data source 312 and control information from a controller / processor 340. The control information may be for the physical broadcast channel (PBCH) , physical control format indicator channel (PCFICH) , physical hybrid automatic repeat request indicator channel (PHICH) , physical downlink control channel (PDCCH) , group common PDCCH (GC PDCCH) , and / or others. The data may be for the physical downlink shared channel (PDSCH) , in some examples.
[0088] Transmit processor 320 may process (e.g., encode and symbol map) the data and control information to obtain data symbols and control symbols, respectively. Transmit processor 320 may also generate reference symbols, such as for the primary synchronization signal (PSS) , secondary synchronization signal (SSS) , PBCH demodulation reference signal (DMRS) , and channel state information reference signal (CSI-RS) .
[0089] Transmit (TX) multiple-input multiple-output (MIMO) processor 330 may perform spatial processing (e.g., precoding) on the data symbols, the control symbols, and / or the reference symbols, if applicable, and may provide output symbol streams to the modulators (MODs) in transceivers 332a-332t. Each modulator in transceivers 332a-332t may process a respective output symbol stream to obtain an output sample stream. Each modulator may further process (e.g., convert to analog, amplify, filter, and upconvert) the output sample stream to obtain a downlink signal. Downlink signals from the modulators in transceivers 332a-332t may be transmitted via the antennas 334a-334t, respectively.
[0090] In order to receive the downlink transmission, UE 104 includes antennas 352a-352r that may receive the downlink signals from the BS 102 and may provide received signals to the demodulators (DEMODs) in transceivers 354a-354r, respectively. Each demodulator in transceivers 354a-354r may condition (e.g., filter, amplify, downconvert, and digitize) a respective received signal to obtain input samples. Each demodulator may further process the input samples to obtain received symbols.
[0091] MIMO detector 356 may obtain received symbols from all the demodulators in transceivers 354a-354r, perform MIMO detection on the received symbols if applicable, and provide detected symbols. Receive processor 358 may process (e.g., demodulate, deinterleave, and decode) the detected symbols, provide decoded data for the UE 104 to a data sink 360, and provide decoded control information to a controller / processor 380.
[0092] In regards to an example uplink transmission, UE 104 further includes a transmit processor 364 that may receive and process data (e.g., for the physical uplink shared channel (PUSCH) ) from a data source 362 and control information (e.g., for the physical uplink control channel (PUCCH) ) from the controller / processor 380. Transmit processor 364 may also generate reference symbols for a reference signal (e.g., for the sounding reference signal (SRS) ) . The symbols from the transmit processor 364 may be precoded by a TX MIMO processor 366 if applicable, further processed by the modulators in transceivers 354a-354r (e.g., for single-carrier frequency division multiplexing (SC-FDM) ) , and transmitted to BS 102.
[0093] At BS 102, the uplink signals from UE 104 may be received by antennas 334a-t, processed by the demodulators in transceivers 332a-332t, detected by a MIMO detector 336 if applicable, and further processed by a receive processor 338 to obtain decoded data and control information sent by UE 104. Receive processor 338 may provide the decoded data to a data sink 339 and the decoded control information to the controller / processor 340.
[0094] Memories 342 and 382 may store data and program codes for BS 102 and UE 104, respectively.
[0095] Scheduler 344 may schedule UEs for data transmission on the downlink and / or uplink.
[0096] In various aspects, BS 102 may be described as transmitting and receiving various types of data associated with the methods described herein. In these contexts, “transmitting” may refer to various mechanisms of outputting data, such as outputting data from data source 312, scheduler 344, memory 342, transmit processor 320, controller / processor 340, TX MIMO processor 330, transceivers 332a-t, antenna 334a-t, and / or other aspects described herein. Similarly, “receiving” may refer to various mechanisms of obtaining data, such as obtaining data from antennas 334a-t, transceivers 332a-t, receive (RX) MIMO detector 336, controller / processor 340, receive processor 338, scheduler 344, memory 342, and / or other aspects described herein.
[0097] In various aspects, UE 104 may likewise be described as transmitting and receiving various types of data associated with the methods described herein. In these contexts, “transmitting” may refer to various mechanisms of outputting data, such as outputting data from data source 362, memory 382, transmit processor 364, controller / processor 380, TX MIMO processor 366, transceivers 354a-t, antenna 352a-t, and / or other aspects described herein. Similarly, “receiving” may refer to various mechanisms of obtaining data, such as obtaining data from antennas 352a-t, transceivers 354a-t, RX MIMO detector 356, controller / processor 380, receive processor 358, memory 382, and / or other aspects described herein.
[0098] In some aspects, one or more processors may be configured to perform various operations, such as those associated with the methods described herein, and transmit (output) to or receive (obtain) data from another interface that is configured to transmit or receive, respectively, the data.
[0099] FIGS. 4A, 4B, 4C, and 4D depict aspects of data structures for a wireless communications network, such as wireless communications network 100 of FIG. 1.
[0100] In particular, FIG. 4A is a diagram 400 illustrating an example of a first subframe within a 5G (e.g., 5G NR) frame structure, FIG. 4B is a diagram 430 illustrating an example of DL channels within a 5G subframe, FIG. 4C is a diagram 450 illustrating an example of a second subframe within a 5G frame structure, and FIG. 4D is a diagram 480 illustrating an example of UL channels within a 5G subframe.
[0101] Wireless communications systems may utilize orthogonal frequency division multiplexing (OFDM) with a cyclic prefix (CP) on the uplink and downlink. Such systems may also support half-duplex operation using time division duplexing (TDD) . OFDM and SC-FDM partition the system bandwidth (e.g., as depicted in FIGS. 4B and 4D) into multiple orthogonal subcarriers. Each subcarrier may be modulated with data. Modulation symbols may be sent in the frequency domain with OFDM and / or in the time domain with SC-FDM.
[0102] A wireless communications frame structure may be frequency division duplex (FDD) , in which, for a particular set of subcarriers, subframes within the set of subcarriers are dedicated for either DL or UL. Wireless communications frame structures may also be time division duplex (TDD) , in which, for a particular set of subcarriers, subframes within the set of subcarriers are dedicated for both DL and UL.
[0103] In FIG. 4A and 4C, the wireless communications frame structure is TDD where D is DL, U is UL, and X is flexible for use between DL / UL. UEs may be configured with a slot format through a received slot format indicator (SFI) (dynamically through DL control information (DCI) , or semi-statically / statically through radio resource control (RRC) signaling) . In the depicted examples, a 10 ms frame is divided into 10 equally sized 1 ms subframes. Each subframe may include one or more time slots. In some examples, each slot may include 7 or 14 symbols, depending on the slot format. Subframes may also include mini-slots, which generally have fewer symbols than an entire slot. Other wireless communications technologies may have a different frame structure and / or different channels.
[0104] In certain aspects, the number of slots within a subframe is based on a slot configuration and a numerology. For example, for slot configuration 0, different numerologies (μ) 0 to 6 allow for 1, 2, 4, 8, 16, 32, and 64 slots, respectively, per subframe. For slot configuration 1, different numerologies 0 to 2 allow for 2, 4, and 8 slots, respectively, per subframe. Accordingly, for slot configuration 0 and numerology μ, there are 14 symbols / slot and 2μ slots / subframe. The subcarrier spacing and symbol length / duration are a function of the numerology. The subcarrier spacing may be equal to 2μ×15 kHz, where μ is the numerology 0 to 6. As such, the numerology μ=0 has a subcarrier spacing of 15 kHz and the numerology μ=6 has a subcarrier spacing of 960 kHz. The symbol length / duration is inversely related to the subcarrier spacing. FIGS. 4A, 4B, 4C, and 4D provide an example of slot configuration 0 with 14 symbols per slot and numerology μ=2 with 4 slots per subframe. The slot duration is 0.25 ms, the subcarrier spacing is 60 kHz, and the symbol duration is approximately 16.67 μs.
[0105] As depicted in FIGS. 4A, 4B, 4C, and 4D, a resource grid may be used to represent the frame structure. Each time slot includes a resource block (RB) (also referred to as physical RBs (PRBs) ) that extends, for example, 12 consecutive subcarriers. The resource grid is divided into multiple resource elements (REs) . The number of bits carried by each RE depends on the modulation scheme.
[0106] As illustrated in FIG. 4A, some of the REs carry reference (pilot) signals (RS) for a UE (e.g., UE 104 of FIGS. 1 and 3) . The RS may include demodulation RS (DMRS) and / or channel state information reference signals (CSI-RS) for channel estimation at the UE. The RS may also include beam measurement RS (BRS) , beam refinement RS (BRRS) , and / or phase tracking RS (PT-RS) .
[0107] FIG. 4B illustrates an example of various DL channels within a subframe of a frame. The physical downlink control channel (PDCCH) carries DCI within one or more control channel elements (CCEs) , each CCE including, for example, nine RE groups (REGs) , each REG including, for example, four consecutive REs in an OFDM symbol.
[0108] A primary synchronization signal (PSS) may be within symbol 2 of particular subframes of a frame. The PSS is used by a UE (e.g., 104 of FIGS. 1 and 3) to determine subframe / symbol timing and a physical layer identity.
[0109] A secondary synchronization signal (SSS) may be within symbol 4 of particular subframes of a frame. The SSS is used by a UE to determine a physical layer cell identity group number and radio frame timing.
[0110] Based on the physical layer identity and the physical layer cell identity group number, the UE can determine a physical cell identifier (PCI) . Based on the PCI, the UE can determine the locations of the aforementioned DMRS. The physical broadcast channel (PBCH) , which carries a master information block (MIB) , may be logically grouped with the PSS and SSS to form a synchronization signal (SS) / PBCH block. The MIB provides a number of RBs in the system bandwidth and a system frame number (SFN) . The physical downlink shared channel (PDSCH) carries user data, broadcast system information not transmitted through the PBCH such as system information blocks (SIBs) , and / or paging messages.
[0111] As illustrated in FIG. 4C, some of the REs carry DMRS (indicated as R for one particular configuration, but other DMRS configurations are possible) for channel estimation at the base station. The UE may transmit DMRS for the PUCCH and DMRS for the PUSCH. The PUSCH DMRS may be transmitted, for example, in the first one or two symbols of the PUSCH. The PUCCH DMRS may be transmitted in different configurations depending on whether short or long PUCCHs are transmitted and depending on the particular PUCCH format used. UE 104 may transmit sounding reference signals (SRS) . The SRS may be transmitted, for example, in the last symbol of a subframe. The SRS may have a comb structure, and a UE may transmit SRS on one of the combs. The SRS may be used by a base station for channel quality estimation to enable frequency-dependent scheduling on the UL.
[0112] FIG. 4D illustrates an example of various UL channels within a subframe of a frame. The PUCCH may be located as indicated in one configuration. The PUCCH carries uplink control information (UCI) , such as scheduling requests, a channel quality indicator (CQI) , a precoding matrix indicator (PMI) , a rank indicator (RI) , and hybrid automatic repeat request (HARQ) acknowledgment (ACK) / negative acknowledgement (NACK) feedback. The PUSCH carries data, and may additionally be used to carry a buffer status report (BSR) , a power headroom report (PHR) , and / or UCI. Overview of AI / ML Functional Framework for RAN Intelligence
[0113] FIG. 5 illustrates an AI / ML functional framework 500 applied for AI-enabled RAN intelligence. The AI / ML functional framework 500 includes a data collection function 502, a model training function 504, a model inference function 506, and an actor function 508, which interoperate to provide a platform for collaboratively applying AI / ML to various procedures in RAN. In some aspects, an ML model may be a neural network to implement a function.
[0114] The data collection function 502 generally provides input data to the model training function 504 and the model inference function 506. AI / ML algorithm specific data preparation (e.g., data pre-processing and cleaning, formatting, and transformation) may not be carried out in the data collection function 502. Examples of input data to the data collection function 502 (or other functions) may include measurements from UEs or different network entities, feedback from the actor function 508, and output from an AI / ML model. In some cases, analysis of data needed at the model training function 504 and at the model inference function 506 may be performed at the data collection function 502. As illustrated, the data collection function 502 may deliver training data to the model training function 504 and inference data to the model inference function 506.
[0115] The model training function 504 may perform AI / ML model training, validation, and testing, which may generate model performance metrics as part of the model testing procedure. The model training function 504 may also be responsible for data preparation (e.g., data pre-processing and cleaning, data formatting, and data transformation) based on the training data delivered by the data collection function 502.
[0116] The model training function 504 may provide model deployment / update data to the model interface function 506. The model deployment / update data may be used to initially deploy a trained, validated, and tested AI / ML model to the model inference function 506 or to deliver an updated model to the model inference function 506.
[0117] As illustrated, the model inference function 506 may provide AI / ML model inference output (e.g., predictions or decisions) to the actor function 508 and may also provide model performance feedback to the model training function 504. The model inference function 506 may also be responsible for data preparation (e.g., data pre-processing and cleaning, data formatting, and data transformation) based on inference data delivered by the data collection function 502, at times.
[0118] The inference output of the AI / ML model may be produced by the model inference function 506. Specific details of this output may be specific in terms of use cases. The model performance feedback may be used for monitoring the performance of the AI / ML model, at times. In some cases, the model performance feedback may be delivered to the model training function 504, for example, if certain information derived from the model inference function is suitable for improvement of the AI / ML model trained in the model training function 504.
[0119] The model inference function 506 may signal the outputs of the model to nodes that have requested them (e.g., via subscription) , or nodes that take actions based on the output from the model inference function. An AI / ML model used in a model inference function 506 may need to be initially trained, validated and tested by a model training function before deployment. The model training function 504 and model inference function 506 may be able to request specific information to be used to train or execute the AI / ML algorithm and to avoid reception of unnecessary information. The nature of such information may depend on the use case and on the AI / ML algorithm.
[0120] The actor function 508 may receive the output from the model inference function 506, which may trigger or perform corresponding actions. The actor function 508 may trigger actions directed to other entities or to itself. The feedback generated by the actor function 508 may provide information used to derive training data and / or inference data or to monitor the performance of the AI / ML model. As noted above, input data for a data collection function 502 may include this feedback from the actor function 508. The feedback from the actor function 508 or other network entities (via data collection function 502) may also be used at the model inference function 506.
[0121] The AI / ML functional framework 500 may be deployed in various RAN intelligence-based use cases. Such use cases may include CSI feedback enhancement, enhanced beam management (BM) , positioning and location (Pos-Loc) accuracy enhancement, and various other use cases. Overview of AI / ML Model based CSF Techniques
[0122] Various techniques may be used to help determine the channel state between wireless communications devices so that those devices can optimize their wireless communications configurations (e.g., choosing the best beam for transmitting and receiving data) . For example, a channel state information reference signal (CSI-RS) may be transmitted by one device. The CSI-RS may be measured by another device in order to estimate channel state and to provide CSI feedback. The CSI feedback is useful for optimizing wireless communications between the two devices.
[0123] Due to the growing complexity and capability of wireless communication devices, such as those capable of transmitting and receiving over multiple input and output antenna ports (e.g., implementing MIMO techniques) , certain techniques may require significant processing power and time, which reduces the performance of both the devices and the overall wireless communications network. These technical problems are exacerbated by certain use cases and environments for wireless communications, which are often dynamic. In other words, because channel state is frequently changing, channel estimation and feedback procedures are often performed frequently, leading to high power use and significant network overhead (e.g., in terms of time and frequency resources dedicated to channel estimation) for the wireless communication system.
[0124] ML models may be employed to more accurately, and more efficiently, perform various functions related to channel state estimation and feedback. With an AI / ML interface, a UE (e.g., such as a UE 104) may use a trained AI / ML model to implement a function. For example, for CSI, the UE may use a neural network to derive a compressed representation of the CSI to feed back to the network (e.g., a gNB 102) . The network may use another neural network to reconstruct the CSI from the compressed representation. For the reconstruction to be accurate, the UE-side and NW-side models may be trained in a collaborative manner, so that the compressed representation created by the UE-side model is interpreted and decoded correctly by the NW-side model.
[0125] A wireless communication system may multiplex Nt ports on Nt resource elements of each resource block using, for example, time division multiplexing (TDM) , code division multiplexing (CDM) , and / or frequency division multiplexing (FDM) . Such systems may generally implement a resource block density between 0.5 and 1, such that the resource elements are transmitted in every other or every single resource block. By contrast, a machine learning model deployed by a transmitting device (e.g., a base station) may be trained to perform multiplexing of Nt ports on L resource elements of each resource block, where L < Nt, which thus reduces the number of resource elements needed for channel estimation-leaving more resource elements available for data transmission. In addition to reducing the number of resource elements needed for channel estimation, which reduces power and enhances resource utilization, such models may operate with reduced resource block density (e.g., below 0.5) and non-uniform resource block patterns may also be implemented, which further improve upon the aforementioned benefits. At a receiving device (e.g., a user equipment) side, a ML-based channel estimator may be trained to recover the full channel, e.g., Nt ports on all resource blocks while receiving the reduced number of resource, L. In various aspects, CSI-RS multiplexing models at transmitter side and receiver side may be trained jointly or sequentially.
[0126] As another example, a CSI reporting configuration may rely on a PMI searching algorithm as well as a PMI codebook for determining and reporting the best PMI codewords (e.g., CSI feedback) to a network. However, a ML-based model, such as an encoder and decoder, may be trained to generate CSI feedback directly, which obviates the need for the PMI searching algorithm (replaced by the encoder) and the PMI codebook (replaced by the decoder) . In aspects described herein, a CSI encoder at the UE side may be trained to compress the channel estimate to a few bits that are then reported to a network entity (e.g., a BS) , while the CSI decoder at the network entity side is trained to recover the channel or the precoding matrix using the reported bits.
[0127] Thus, generally speaking, ML models may be trained to perform many functions related to channel estimation and feedback, and such models may generally be more accurate, faster, more power efficient, and more capable of maintaining performance in very dynamic radio environments. There are various options for what type of information to feedback as CSF information and the decision of what to feedback may depend on a particular implementation (e.g., whether AI / ML-based or not) . The type of feedback may range (in a continuum from relatively sparse information, such as rank indicator (RI) , PMI, and channel quality indicator (CQI) , to detailed, full channel, information.
[0128] As illustrated in the block diagram of an example ML-based CSI feedback mechanism 600 in FIG. 6, in some cases, a CSI encoder 602 (at the UE) and a CSI decoder 604 at the network (e.g., gNB) may use AI / ML-based CSI feedback to replace the use of a codebook. In some cases, the CSI encoder 602 may not employ AI / ML-based CSI feedback and just the CSI decoder 604 may perform AI / ML-based processing of CSI feedback. In general, AI / ML-based CSI feedback encoding by the CSI encoder 602 is analogous to the PMI searching algorithm in current systems, while the AI / ML-based CSI processing by the CSI decoder 604 is generally analogous to the PMI codebook, which is used to translate the CSI reporting bits to a PMI codeword.
[0129] As illustrated in the table 700 of FIG. 7, input to the CSI decoder 604 may include a downlink channel matrix (H) , downlink precoders (V) , and an interference covariance matrix (Rnn) . Output of the decoder could be downlink channel matrix (H) , transmit covariance matrix, downlink precoders (V) , interference covariance matrix (Rnn) , where H or V could correspond to raw channel or channel pre-whitened by the UE based on its demodulation filter.
[0130] In certain ML models used for CSI measurement and reporting, a UE and network entity (e.g., gNB) may collaborate by sharing assistance information. For example, in one form of collaboration, inter-node assistance may help improve the respective nodes of an ML based algorithm. The inter-node assistance may involve UEs receiving assistance information from gNBs (e.g., for training, adaptation, etc. ) , as well as UEs receiving assistance information from gNBs. In some cases, assistance information may be exchanged without exchanging information about the actual ML models. In other cases (e.g., for joint ML operation) , a UE and gNB may exchange information regarding ML models or ML model instruction.
[0131] In ML based CSI schemes involve low-density CSI-RS patterns, where information may be exchanged between a UE and gNB. In some cases, due to an association between low-density CSI-RS patterns and a UE channel estimation (CHEST) ML model, the CHEST ML model may need to be changed. For example, when the environment changes, the CSI-RS pattern may change. In such cases, the UE may receive assistance information from the gNB for training of the CHEST ML model, adaptation of the CHEST ML model, and the like.
[0132] Reduction along the frequency dimension may be based on the low-density resource block (RB) pattern. For example, K RBs out of N RBs (K<N) may be selected to transmit CSI-RS. In such cases, the RB pattern may be a uniform RB pattern, a random RB pattern, or a learned RB pattern (e.g., a pattern used specifically for certain scenarios) .
[0133] Non-orthogonal cover codes may be used to multiplex Nt CSI-RS ports onto L resource elements (REs) within an RB, where L<N_t. The UE may use an ML model associated with the low-density CSI-RS pattern to recover the
[0134] A UE may receive signaling, from the network entity, indicating at least a first CSI-RS pattern. The first CSI-RS pattern may indicate time and frequency resources (e.g., RBs and / or REs within an RB) used for CSI-RS transmissions. The first CSI-RS pattern may indicate a lower density of CSI-RS resources in a channel frequency range, than a second (e.g., full-density) CSI-RS pattern.
[0135] The UE may update an ML model based on the first CSI-RS pattern. For example, during training the UE may apply a loss function by comparing output of the ML model. The output of the model is generated based on measurements taken from CSI-RS that are transmitted according to the first CSI-RS pattern. The output of the ML model may be compared to a ground-truth estimated channel obtained by using the full-density CSI-RS pattern. The first CSI-RS pattern may also be used when the UE uses the ML model for channel estimation (e.g., after the ML model is well-trained and ready to use) .
[0136] Using the updated ML model, the UE may generate CSI for the channel frequency range based on measurements taken by the UE according to the first CSI-RS pattern. The UE may input measurements taken from CSI-RS transmitted according to the first CSI-RS pattern, to the trained ML model to generate a channel estimate. The UE may then transmit a report to the network entity indicating the CSI for the channel frequency range. Overview of CSF Information AI / ML Model Training
[0137] There are various types and options for training and developing models that run at both the UE and network (NW) side. Systems that use models at both the UE and the network side for CSF information may be referred to as two-sided models.
[0138] FIG. 8 illustrates an example system 800 for first type of model training with a single training entity. According to this approach, training data may be collected and uploaded to the training entity (prior to training) , for example, by the UEs as illustrated. The single training entity handles the training for both the UE models and the network side models (e.g., the gNB as illustrated) . After model training, the trained models are transferred by the training entity to the network and / or UE (s) for inference.
[0139] FIG. 9 illustrates an example system 900 for a second type of model training with separate model training entities. For example, a UE-side training entity may send data representing ground-truth (also referred to as target output) to the NW-side training entity before model training starts. In a model training phase, the UE side training entity may send an activation signal / message to the NW-side training entity. After the model training, the NW-side training entity may send back the gradient to the UE-side training entity (e.g., for back-propagation) .
[0140] FIG. 10 illustrates a system 1000 for third type of training which may involve model training first performed at the UE (UE-first operation) or model training first performed at the NW (NW-first operation) .
[0141] For a UE-first model training operation, the UE-side training entity may first train the UE-side model (e.g., together with a private / reference decoder) . The UE may then share the latent message output by the UE-side model (z) and the ground-truth (v) or output by the UE-side private / reference decoder The NW-side training entity may then train the NW-side model using z as input and v or as target output.
[0142] For a NW-first model training operation, the UE-side may first share the ground-truth with the NW. The NW may then train the NW-side model (e.g., together with a private / reference encoder) . The NW may then send the input of the private / reference encoder (v) and latent message output by the encoder (z) to the UE. The UE-side may then train the UE-side model using v as input and z as output.
[0143] As noted, one potential challenge to the use of AI / ML-based CSI processing models is that there may be significant amount of complexity involved in training, transferring, and / or deployment of AI / ML models. Accordingly, to facilitate and support model deployment, one objective is to alleviate and reduce inter-vendor collaboration complexity for model training.
[0144] Various options may help address issues related to inter-vendor training collaboration of AI / ML-based CSI processing using a two-sided model. As noted, CSI processing may refer to one or more aspects associated with CSI acquisition / estimation, one or more aspects associated with feedback of CSI, or a combination thereof.
[0145] Such options may include, for example, a first option ( “Option 1” ) that involves a (fully) standardized reference model (defining model structure + parameters) . As used herein, parameters may refer to variables learned by the model during a training process. In some aspects, the parameters include weights assigned to connections between nodes (e.g., neurons in a neural network) , that are updated during training of the model. In some aspects, the parameters include bias values. In some aspects, parameters may include hyperparameters sets by a user. FIG. 11 depicts an example system 1100 for model training, according to the first option, with a standardized reference model. Based on original CSI, v, the UE may generate (using a UE-side encoder) output UCI payload with bitstream payload (b0, b1, b2, . . . ) that maps to reconstructed CSI The UE-side encoder may be standardized or left to UE implementation. The mapping of the encoded bitstream payload to the reconstructed CSI may be predetermined, based on a standardized decoder model.
[0146] A second option ( “Option 2” ) involves use a standardized dataset. In the context of AI / ML-based CSI compression, the term “dataset” may refer to a set of data samples of CSI feedback and associated target CSI. FIG. 12 depicts an example system 1200 for model training using a standardized dataset according to the second option. As shown, the mapping of the bitstream to the reconstructed CSI may be standardized explicitly.
[0147] Thus, with the first or second options, the PMI mapping may be predetermined, avoiding the need for inter-vendor collaboration.
[0148] A third option ( “Option 3” ) involves use of a standardized reference model structure and a standardized parameter exchange between the NW-side and the UE-side.
[0149] According to a first sub-option (sub-option 3a) , parameters received at the UE or UE-side may go through further processing (referred to herein as offline engineering) at the UE-side (e.g., at a UE-side over-the-top (OTT) server) . This may provide the opportunity for potential re-training, re-development of a different model, and / or offline testing. For the sub-options 3a, the parameter exchange may originate from the NW-side and end at the UE-side. In such cases, parameters exchanged from the NW-side to UE-side may be via CSI generation (sub-option 3a-1) , via CSI reconstruction part (sub-options 3a-2) , or via both (sub-option 3a-3) . Additional information may be shared from the NW-side to assist UE-side offline engineering and provide performance guidance. The additional information may include a performance target (s) , a dataset (s) , information related to collecting a dataset (s) , or a combination thereof.
[0150] According to a second sub-option (sub-option 3b) , parameters received at the UE may be directly used for inference at the UE without offline engineering, potentially with on-device operations. For sub-option 3b, the parameter exchange may be via an OTA-interface, offline delivery, or other mechanisms. The parameter exchange is from the NW to the UE, and includes a CSI generation part.
[0151] A fourth option ( “Option 4” ) involves the use of standardized data or dataset format in addition to a standardized dataset exchange between the NW-side and the UE-side. The dataset received at the UE or UE-side goes through offline engineering at the UE-side (e.g., at the UE-side OTT server) , where model training or offline testing may be performed. According to a first sub-option (sub-option 4-1) the dataset exchanged from the NW-side to UE-side may include target CSI and CSI feedback. According to a second sub-option (sub-option 4-2) , the dataset exchanged from the NW-side to UE-side may include CSI feedback and reconstructed target CSI. According to a third sub-option (sub-option 4-3) , the dataset exchanged from the NW-side to UE-side may include target CSI, CSI feedback, and reconstructed target CSI. In some cases, additional information (e.g., performance targets) (e.g., if necessary) , may be shared from the NW-side to help UE-side offline engineering and provide performance guidance.
[0152] A fifth option ( “Option 5” ) involves the use of a standardized model format in addition to a standardized reference model exchange between the NW-side and the UE-side. Depending on a particular deployment, one or more of these options may be used separately or together (in addition, or as an alternative, to other options) . According to a first sub-option (sub-option 5a) , a model received at the UE or UE-side may go through offline engineering at the UE-side (e.g., at the UE-side OTT server) , again providing the opportunity for potential re-training, re-development of a different model, and / or offline testing. For sub-options 5a, the model exchange may originate from the NW-side and end at the UE-side. In such cases, the model exchanged from the NW-side to UE-side may be via CSI generation (sub-option 5a-1) , via CSI reconstruction part (sub-option 5a-2) , or via both (sub-option 5a-3) . Additional information may be shared from the NW-side to help UE-side offline engineering and provide performance guidance. Such additional information may include a performance target (s) , a dataset (s) , information related to collecting a dataset (s) , or a combination thereof.
[0153] According to a second sub-option (sub-option 5b) , a model received at the UE may be used directly for inference at the UE without offline engineering (e.g., potentially with on-device operations) . For sub-option 5b, the model exchange may be via an OTA-interface, offline delivery, or other mechanisms. The model exchange may be from the NW to the UE, and may include a CSI generation part.
[0154] The third, fourth, and fifth options, described above, for developing a two-sided model may allow for vendor-specific development by using standardized signaling to address complexity. The standardized signaling can be further categorized in various sub-options (or sub-flavors) for various objectives as shown in FIGs. 13-14. FIG. 13 depicts standardized signaling 1300 for a plug-and-play approach to developing a two-sided model, for example where the UE may run untested models. FIG. 14 depicts standardized signaling 1400 for an offline engineering at the UE side approach to developing a two-sided model. For example, the standardized signaling may be used to re-develop or retrain the encoder model at a UE server using the signaled information. As illustrated in FIG. 14, the standardized signaling be used for parameter, dataset, and / or model exchange between entities to support the offline engineering work. Aspects Related to OTA Signaling for Data Collection
[0155] As noted above, there may be a significant amount of complexity involved in the training and / or transferring of AI / ML models. As one non-limiting example, there may be a significant amount of complexity associated with inter-vendor collaboration for training models and / or transferring of two-sided models that run at both the UE side and NW side.
[0156] According to certain aspects, a unified OTA signaling may be used for model development and exchange. An example of such unified OTA signaling may include RRC signaling. In some aspects, a network entity (e.g., a base station 102) may provide, via RRC signaling, information associated with parameters of a model (e.g., encoder, decoder, encoder-decoder pair) to a wireless node (e.g., a UE 104) . Such information may include a type of the model, an identifier associated with the model, one or more parameters (e.g., weights) , one or more performance targets, one or more training datasets, or any combination thereof.
[0157] For data collection for the UE-side AI / ML model training for CSF information, in one option (option 1a) , the UE collects training data and transfers the training data directly to a data collection entity outside the mobile network operator (MNO) for UE-side model training. For example, the UE may transfer the training data to an over-the-top (OTT) server. An OTT server facilitates delivery of applications and services directly over the Internet, and may bypass traditional telecommunications infrastructure, such that the applications and services can be accessed through a standard Internet connection instead of relying on dedicated network resources. Alternatively (option 1b) , the UE may transfer the training data to data collection entity inside the MNO for UE-side model training, and then optionally from the data collection entity to the OTT server outside the MNO.
[0158] In another option (option 2) for data collection for the UE-side AI / ML model training for CSF information, the UE transfers the training data to the core network (e.g., to a 5G core network) . The core network then transfers the training data to the data collection entity (e.g., the OTT server outside the MNO) .
[0159] In another option (option 3) for data collection for the UE-side AI / ML model training for CSF information, the UE transfers the training data to an operations, administration, and maintenance (OAM) entity. The OAM then transfers the training data to the data collection entity (e.g., the OTT server outside the MNO) .
[0160] FIGs. 15-16 depict tables 1500 and 1600 illustrating termination entities, data transfer path, whether control plane (CP) or user plane (UP) should be used to transfer the data, protocol layers involved, and MNO controllability and visibility over the collected data for the data collections and transfer options (options 1a, 1b, 2, and 3) described above. Regarding the table 1500, note 1 specifies full controllability in which the MNO can manage data transfer to the server for UE-side data collection, without the need of service-level agreement (SLA) , including initiating, terminating, and fully managing data transfer. Note 2 specifies visibility of data content signifies that the MNO can, at least, be aware of, access, and comprehend the data without the need of SLA. Note 3 specifies options to realize the different levels of data content visibility to the MNO include full visibility for standardized data content, partial visibility for partially standardized data content (e.g. UE proprietary information can be included transparently together with the standardized data message) , and no standardized visibility (e.g. only UE proprietary information can be included transparently) .
[0161] The OTA exchange of the parameter set, dataset, and / or reference model may reduce inter-vendor collaboration, however, such exchange may still involve high cost, significant Uu interface overhead, and may not be practical. Some solutions reduce this overhead by offline delivery of the parameter set, dataset, and / or reference model, however, these approaches may require inter-vendor collaboration. Accordingly, enhancements are still needed for OTA signaling for CSF information AI / ML model development and exchange.
[0162] Aspects of the present disclosure provide techniques for signaling various aspects associated with data collection for training, transferring, and deploying two-sided models. The signaling aspects described herein may help to alleviate and reduce inter-vender collaboration complexity for model training, transfer, and / or deployment.
[0163] According to certain aspects, a UE may signal information to the network (e.g., to a BS) of a location of a parameter set, a training dataset, and / or a reference model. The information may include a location to send model information and / or an identifier for retrieving the model information. The model information may include a parameter set, a training dataset, and / or a reference model for an AI / ML model, such as for a two-sided CSF information AI / ML model.
[0164] In some aspects, the information indicates where the parameter set, the training dataset, and / or the reference model should be sent for offline training at the UE-side. For example, the location may be a uniform resource locator (URL) or an Internet Protocol (IP) address. In some aspects, the location is the location of a UE-side server (e.g., an OTT server) .
[0165] In some aspects, the information is an identifier that can be used for retrieving the parameter set, the training dataset, and / or the reference model. For example, the identifier may be a pairing ID, an associated ID, or a vendor-specific ID. A pairing ID may be a unique identifier used to link or authenticate two devices so they can communicate or work together. An associated ID may be unique identifier that links or relates one entity to another within a system. A vendor-specific ID may be a unique identifier assigned by a vendor or manufacturer to distinguish its products, devices, or services from those of other vendors.
[0166] In some aspects, the network (e.g., a cell, base station (BS) , or a gNB) may configure the UE to provide the information (e.g., the location and / or the identifier) . For example, the network may configure the UE for providing the information while configuring the UE for training data collection. In some aspects, the network may configure the UE for providing the information once the network entity has a parameter set, a training dataset, or a reference model for sharing.
[0167] According to certain aspects, the network shares the parameter set, training dataset, and / or reference model with the UE-side. For example, the network may provide the parameter set, training dataset, and / or reference model to the UE-side by sending the parameter set, training dataset, and / or reference model to the location (e.g., the URL or IP address) signaled by the UE.
[0168] According to certain aspects, the OAM entity shares the parameter set, training dataset, and / or reference model with the UE-side. In some aspects, the base station forwards the location (e.g., the URL or IP address) signaled by the UE to the OAM entity. The OAM entity may provide the parameter set, training dataset, and / or reference model to the UE-side by sending the parameter set, training dataset, and / or reference model to the location forward by the base station signaled by the UE.
[0169] In some aspects, a network data analytics function (NWDAF) requests the parameter set, training dataset, and / or reference model from the OAM using the identifier. The NWDAF is a centralized data collection and analysis system, part of the 5G core network, that provides real-time operational intelligence for the 5G network. The NWDAF collects and analyzes data (e.g., such as statistics, metrics, and events) from various network functions (NFs) , UEs, and operations systems including the OAM, within the core network, cloud networks, and edge networks. The NWDAF may use ML models to process the collected data for forecasting network behavior, user experience, and potential issues, such as for load prediction for network slices, quality of service (QoS) monitoring and prediction, UE behavior analysis, anomaly detection in network traffic, and other uses.
[0170] In some aspects, the base station uses the identifier (e.g., the pairing ID or the associated ID) for tagging the parameter set, training dataset, and / or reference model. In some aspects, the base station uses a pairing ID and / or associated ID for tagging the parameter set, training dataset, and / or reference model without the identifier being signaled from the UE. In some aspects, the UE signals a UE-specific identifier to the network (e.g., to the base station) .
[0171] In some aspects, cell IDs and / or base station IDs may be signaled in addition to the UE-signaled ID (e.g., the pairing ID, associated ID, vendor-specific ID, UE-specific ID) for offline training.
[0172] Techniques presented herein using OTA signaling for data collection may be understood with reference to FIGs. 17-23, which depict call flow diagrams illustrating operations 1700-2300, respectively, in accordance with certain aspects of the present disclosure.
[0173] In some aspects, the UE shown in FIGs. 17-23 may be an example of a UE 104 depicted and described with respect to FIGs. 1 and 3. In some aspects, the network entity shown in FIGs. 17-23 may be an example of a BS 102 (e.g., a gNB) depicted and described with respect to FIGs. 1 and 3 or a disaggregated base station depicted and described with respect to FIG. 2. In some aspects, a UE-side server shown in FIGs. 17-23 may be an example of a computing system (s) (e.g., server (s) ) associated with the UE (e.g., an OTT server) .
[0174] FIG. 17 is a call flow diagram illustrating example operations 1700 for providing model information. As shown in FIG. 17, at operation 1710, the UE 1704 is provisioned with location information of the UE-side server 1708. The UE-side server 1708 performs model training. At operation 1712, the network entity 1702 (e.g., a base station) requests the UE 1704 to provide information for parameter set, training dataset, and / or reference model sharing. In some aspects, at the operation 1712, the network entity 1702 requests the IP address or URL of the UE-side server 1708. In some aspects, the network entity 1702 requests a UE-specific identifier and / or a vendor-specific identifier. At operation 1714, the UE 1704 provides the requested information to the network entity 1702.
[0175] In some aspects, additionally or alternatively, at operation 1718, the OAM 1706 determines that information is required for model sharing. At operation 1720, the OAM 1706 sends a request to the network entity 1702 for the information for parameter set, training dataset, and / or reference model sharing. In this case, at operation 1722, the network entity 1702 request the information from the UE 1704 in response to the request received from the OAM 1706. At operation 1724, the UE 1704 provides the requested information to the network entity 1702 and, at operation 1726, the network entity 1702 forwards the requested information to the OAM 1706.
[0176] According to certain aspects, the configuration for the model sharing information can be provided or requested together with the training data collection, or once the network entity 1702 has a parameter set, a training dataset, and / or a reference model for sharing.
[0177] FIG. 18 is a call flow diagram illustrating example operations 1800 for signaling a parameter set, training data, and / or reference model from the NW-side to the UE-side. As shown in FIG. 18, at operation 1810, a network entity 1802 (e.g., a base station) shares (e.g., reports) model information (e.g., a parameter set, a training dataset, and / or a reference model) to the UE-side server 1808. Additionally or alternatively, at operation 1812, the OAM 1806 may share the model information to the UE-side server 1808. In some aspects, the location (e.g., the URL or the IP address) of the UE-side 1808 is first obtained by the network entity 1802 and / or the OAM 1806 according to the operations 1700 illustrated in FIG. 17. The network entity 1802 and / or the OAM 1806 may provide additional information to the UE-side server 1808, such as one or more identifiers associated with the model information. For example, the additional information may include an associated ID, a pairing ID, a relevant cell ID, and / or a base station ID. At operation 1814, the UE-side server 1808 trains the decoder model based on the model information and, at operation 1816, trains the encoder model based on the collected training data. For example, the collected training data may be obtained by the UE-side server 1808 according to any of the options described with respect to the Tables 1500 and 1600. After training the models, the UE 1804 and UE-side server 1808 deploy the encoder and decoder models parameter set at operation 1818. The models may be used for inference and / or monitoring.
[0178] FIG. 19 is a call flow diagram illustrating example operations 1900 for retrieving a parameter set, training data, and / or reference model from the NW-side. As shown in FIG. 19, at operation 1912, the NWDAF 1910 requests the OAM 1906 to provide model information (e.g., parameter set, training dataset, and / or reference model) . In some aspects, the NWDAF 1910 requests the model information by providing the identifier (e.g., the associated ID, pairing ID, UE-vendor specific ID) signaled by the UE to the OAM 1906, such as according to the operations 1700 and / or 1800 illustrated in FIG. 17 and FIG. 18, respectively. The NWDAF 1910 may be (pre-) configured with the identifier for which the model information is requested. At operation 1914, the OAM 1906 signals the requested model information to the NWDAF 1910. The OAM 1906 may signal additional information with the requested model information, such as relevant cell IDs and / or base station IDs. At operation 1916, the NWDAF 1910 trains the decoder model based on the model information and the additional information. The UE side may trained a private decoder inside the NWDAF 1910 based on the model information, for example by a model training logical function (MTLF) . The MTLF is a logical function in the NWDAF which trains ML models and exposes new training services.
[0179] FIG. 20 is a call flow diagram illustrating example operations 2000 for retrieving UE-side training data for encoder training. As shown in FIG. 20, at operation 2012, the UE 2004 performs UE-side training data collection and reporting with the OAM, network function, or server 2008. Additionally or alternatively, the UE 2004 performs UE-side training data collection and reporting with the NWDAF 2010 at operation 2014. At operation 2016, NWDAF 2010 requests training data from the OAM 2008, NF, or server, using the associated ID, pairing ID, UE-vendor-specific ID, or other ID signaled from the UE 2004. At operation 2018, the OAM, NF, or server 2008 provides the training data to the NWDAF 2010 in response to the request. The OAM, NF, or server 2008 may provide the additional information with the training data, such as the relevant cell ID and / or base station ID. At operation 2020, the NWDAF 2010 trains the encoder based on the UE-side training data and, at operation 2022, deploys the encoder and (private) decoder model and / or parameter set, which may be used for inference and / or monitoring.
[0180] FIG. 21 is a call flow diagram illustrating example operations 2100 for signaling a parameter set, training data, and / or reference model to the UE-side. As shown in FIG. 21, at operation 2112, the NWDAF 2110 receives the location (e.g., URL or IP address) of the UE-side sever 2108 either directly from the UE 2104 or forwarded from the OAM 2106, for example, according to the operations 1700 in FIG. 17. At 2114, the NWDAF 2110 request model information (e.g., parameter set, training dataset, and / or reference model) form the OAM 2106 using an identifier (e.g., associated ID, pairing ID, UE-vendor-specific ID) . At operation 2116, the OAM 2106 provides the requested model information to the NWDAF 2110. The OAM 2106 may also provide the additional information at the operation 2116, such the relevant cell ID or base station ID. At operation 2118, the NWDAF 2110 exposes the model information to the UE-side server 2108. At operation 2120, the UE-side training data collection and reporting is performed. At operation 2122, the UE-side server 2108 trains the private decoder model based on the model information and additional information. At operation 2124, the UE-side server 2108 trains the encoder based on the UE-side training data collection. At operation 2126, the encoder and private decoder model and / or parameter set are deployed, and may be used for inference and / or monitoring.
[0181] An application may perform registration with a RAN intelligent controller (RIC) in an O-RAN.
[0182] An eXtended application (xAPP) is an application that runs on a near-RT RIC and enables functions such as optimization and automation of the RAN in near-RT. The near-RT RIC is a component that operations in the near-RT domain, hosts xAPPs, and is responsible for optimizing and controlling RAN functions on short time scales, such as radio resource allocation and interference management.
[0183] A RAN application (rAPP) may performs E2 setup and subscription with a non-RT RIC. An rAPP runs on a non-RT RIC and performs functions that do not involve real-time action to enhance RAN efficiency over longer time scales, such as performance monitoring and data analytics. A non-RT RIC operates in the non-RT domain, dealing with longer time scales. The non-RT RIC may host rAPPs, and handle tasks that involve analysis over longer periods, such as policy management, training of AI models, and advanced network optimizations.
[0184] E2 is the interface between the (near-RT or non-RT) RIC and E2 nodes, such as base stations. During the E2 setup, the RIC establishes a connection with the E2 node to allow the RIC to begin managing and controlling RAN functions. RIC subscription allows applications (xAPPs or rAPPs) to subscribe to specific events or data streams from E2 nodes via the E2 interface, so that the applications can receive event notifications, performance metrics, or specific information (e.g., traffic data, QoS, etc. ) from the E2 node.
[0185] FIG. 22 is a call flow diagram illustrating example operations 2200 for configuring a base station in an O-RAN to report a parameter set, training data, and / or reference model. As shown in FIG. 22 at operation 2206, the APP 2204 may send the RIC 2202 a registration request including the ID (s) (e.g., pairing ID, associated ID, UE-vendor-specific ID, cell ID, base station ID, infra-vendor ID) for retrieving model information. The near-RT RIC can configure a base station to provide the model information (e.g., parameter set, training dataset, and / or reference model) based on the identifier (s) . At operation 2208, the RIC 2202 sends a registration response.
[0186] FIG. 23 is another call flow diagram illustrating example operations 2300 for configuring a base station to report a parameter set, training data, and / or reference model. As shown in FIG. 23, at operation 2306, the APP 2304 may retrieve the model information using a shared data layer (SDL) application programming interface (API) . In some aspects, the APP may subscribe to a platform data and fetch the model information from the platform data. In some aspects, the APP 2304 retrieves the model information using a new API.
[0187] According to certain aspects, a link between the UE and the UE-side server may be secured. In one example, the link between the UE and UE-side server using an authentication and key management for applications (AKMA) procedure. In one example, after the UE is authenticated, a primary key is generated, and one more application-specific keys can be derived from the primary key. To communicate with the UE-side server, the UE may send a request, including a key identifier, to a function. The function may then generate distribute session keys for communication between the UE and the UE-side server.
[0188] In some aspects, the UE communicates with the NWDAF via an intermediate entity that relays communications between the UE and the NWDAF. In one example, the intermediate entity is a Data Collection and Analytics Function (DCAF) . In some aspects, the link between the UE and the intermediate entity is secured.
[0189] In some aspects, a link between the NWDAF and the UE-side server is secured. In some aspects, a link between the OAM and the UE-side server is secured. Example Operations
[0190] FIG. 24 shows an example of a method 2400 of wireless communication by a user equipment (UE) , such as a UE 104 of FIGS. 1 and 3.
[0191] Method 2400 begins at step 2405 with transmitting signaling to a network entity, the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity to a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 28.
[0192] In some aspects, information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination therefore; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0193] In some aspects, the server is configured to train one or more CSF information encoder models, one or more CSF information private decoder models, one or more parameter sets associated with one or more CSF information encoder models, one or more parameter sets associated with one or more CSF information decoder models, or any combination thereof.
[0194] In some aspects, the transmitting the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information is based on configuration received from the network entity.
[0195] In some aspects, the transmitting the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information is based on a configuration received from the network entity when the network entity has one or more parameter sets, one or more training datasets, one or more reference models, or a combination thereof, to share with the server for training the one or more UE ML models.
[0196] In some aspects, the network entity comprises: a serving cell; a gNB; an Operations, Administration, and Maintenance (OAM) entity; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0197] In some aspects, the server comprises: a UE vendor-specific training server within a Mobile Network Operator (MNO) network of the network entity; a training server outside the MNO network; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0198] In some aspects, the method 2400 further includes receiving, from the server, at least one of: a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with the CSF information encoder model, a trained parameter set associated with the CSF information decoder model, or any combination thereof. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 28.
[0199] In one aspect, method 2400, or any aspect related to it, may be performed by an apparatus, such as communications device 2800 of FIG. 28, which includes various components operable, configured, or adapted to perform the method 2400. Communications device 2800 is described below in further detail.
[0200] Note that FIG. 24 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.
[0201] FIG. 25 shows an example of a method 2500 of wireless communication by a network entity, such as a BS 102 of FIGS. 1 and 3, or a disaggregated base station as discussed with respect to FIG. 2.
[0202] Method 2500 begins at step 2505 with receiving signaling from a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity and a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 29.
[0203] Method 2500 then proceeds to step 2510 with transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, to the server based on the received information from the UE. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 29.
[0204] In some aspects, the network entity comprises: a serving cell; or a gNB.
[0205] In some aspects, the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0206] In some aspects, the method 2500 further includes tagging the one or more parameter sets, the one or more training datasets, the one or more reference model, or the combination of thereof, with one or more of the associated IDs or pairing IDs. In some cases, the operations of this step refer to, or may be performed by, circuitry for tagging and / or code for tagging as described with reference to FIG. 29.
[0207] In some aspects, the method 2500 further includes transmitting signaling configuring the UE to provide the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 29.
[0208] In some aspects, the transmitting the signaling configuring the UE to provide the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information is in response to the network entity having one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, to share to the server.
[0209] In some aspects, the method 2500 further includes forwarding the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information, to an Operations, Administration, and Maintenance (OAM) entity. In some cases, the operations of this step refer to, or may be performed by, circuitry for forwarding and / or code for forwarding as described with reference to FIG. 29.
[0210] In some aspects, the method 2500 further includes transmitting a request to the UE for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information, wherein the information is received from the UE in response to the request. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 29.
[0211] In some aspects, the method 2500 further includes receiving a request, from an Operations, Administration, and Maintenance (OAM) entity, for the information for sharing the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof to the server that is configured to train the one or more UE ML models for CSF information, wherein the transmitting the request to the UE is in response to receiving the request from the OAM entity. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 29.
[0212] In some aspects, the method 2500 further includes transmitting signaling of one or more identifiers (IDs) associated with the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the server. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 29.
[0213] In some aspects, the one or more IDs comprises: one or more pairing IDs, one or more associated IDs, one or more UE vendor-specific IDs, one or more cell IDs, or one or more base station IDs associated with the one or more training datasets, the one or more reference models, or the combination thereof.
[0214] In some aspects, the server comprises: a UE vendor specific training server within a Mobile Network Operator (MNO) network of the network entity; a training server outside the MNO network; Network Data Analytics Function (NWDAF) entity; or Model Training Logical Function (MTLF) entity at the NWDAF.
[0215] In one aspect, method 2500, or any aspect related to it, may be performed by an apparatus, such as communications device 2900 of FIG. 29, which includes various components operable, configured, or adapted to perform the method 2500. Communications device 2900 is described below in further detail.
[0216] Note that FIG. 25 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.
[0217] FIG. 26 shows an example of a method 2600 of wireless communication by a training server, such as an OTT.
[0218] Method 2600 begins at step 2605 with transmitting signaling to a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination thereof, with the training server. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 30.
[0219] Method 2600 then proceeds to step 2610 with receiving channel state feedback (CSF) information training data. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 30.
[0220] Method 2600 then proceeds to step 2615 with receiving the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, from a network entity. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 30.
[0221] Method 2600 then proceeds to step 2620 with training, based on the CSF information training data and on the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with a CSF information encoder model, or a trained parameter set associated with a CSF information decoder model, or any combination thereof. In some cases, the operations of this step refer to, or may be performed by, circuitry for training and / or code for training as described with reference to FIG. 30.
[0222] Method 2600 then proceeds to step 2625 with providing, to the UE, the trained CSF information encoder model, the trained CSF information decoder model, the trained parameter set associated with the CSF information encoder model, or the trained parameter set associated with the CSF information decoder model, or the combination thereof. In some cases, the operations of this step refer to, or may be performed by, circuitry for providing and / or code for providing as described with reference to FIG. 30.
[0223] In some aspects, information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the training server comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0224] In some aspects, the method 2600 further includes receiving a request from a Network Data Analytics Function (NWDAF) for the CSF information training data, wherein the request includes at least one of: the one or more Associated IDs, the one or more pairing IDs, or the one or more UE vendor-specific IDs. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 30.
[0225] In some aspects, the method 2600 further includes transmitting the CSF information training data to the NWDAF in response to the request. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 30.
[0226] In some aspects, the method 2600 further includes transmitting, to the NWDAF, one or more cell IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, or one or more base station IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 30.
[0227] In some aspects, the network entity comprises: a serving cell; a gNB; an Operations, Administration, and Maintenance (OAM) entity; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0228] In some aspects, the server comprises: a UE vendor-specific training server within a Mobile Network Operator (MNO) network of the network entity; a training server outside the MNO network; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0229] In some aspects, the method 2600 further includes obtaining a configuration of one or more identifiers (IDs) associated with the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof. In some cases, the operations of this step refer to, or may be performed by, circuitry for obtaining and / or code for obtaining as described with reference to FIG. 30.
[0230] In some aspects, the one or more IDs comprise one or more base station (BS) IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof.
[0231] In one aspect, method 2600, or any aspect related to it, may be performed by an apparatus, such as communications device 3000 of FIG. 30, which includes various components operable, configured, or adapted to perform the method 2600. Communications device 3000 is described below in further detail.
[0232] Note that FIG. 26 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.
[0233] FIG. 27 shows an example of a method 2700 of wireless communication by an operations, administration, and maintenance (OAM) entity.
[0234] Method 2700 begins at step 2705 with receiving signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, with a server that is configured to train one or more user equipment (UE) machine learning (ML) models for channel state feedback (CSF) information. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 31.
[0235] Method 2700 then proceeds to step 2710 with transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the server based on the received information from the UE. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 31.
[0236] In some aspects, the receiving the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server comprises receiving the signaling from a user equipment (UE) .
[0237] In some aspects, the method 2700 further includes sending a request to a network entity for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, with the server, wherein the receiving the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, with the server is in response to the request. In some cases, the operations of this step refer to, or may be performed by, circuitry for sending and / or code for sending as described with reference to FIG. 31.
[0238] In some aspects, information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server that is configured to train one or more UE ML models for CSF information comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0239] In some aspects, the method 2700 further includes receiving a request from a Network Data Analytics Function (NWDAF) of a core network for the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, wherein the request includes the one or more Associated IDs or the one or more pairing IDs. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 31.
[0240] In some aspects, the method 2700 further includes transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the NWDAF in response to the request from the NWDAF. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 31.
[0241] In some aspects, the method 2700 further includes receiving CSF information training data from a user equipment (UE) . In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 31.
[0242] In some aspects, the method 2700 further includes receiving a request from a Network Data Analytics Function (NWDAF) for the CSF information training data, wherein the request includes the one or more Associated IDs or the one or more pairing IDs. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 31.
[0243] In some aspects, the method 2700 further includes transmitting the CSF information training data to the NWDAF in response to the request from the NWDAF. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 31.
[0244] In some aspects, the method 2700 further includes transmitting the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server, to a Network Data Analytics Function (NWDAF) . In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 31.
[0245] In some aspects, the method 2700 further includes receiving request from the NWDAF for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server, wherein the transmitting the information for sharing the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the server, to the NWDAF is in response to the request from the NWDAF. In some cases, the operations of this step refer to, or may be performed by, circuitry for receiving and / or code for receiving as described with reference to FIG. 31.
[0246] In some aspects, the method 2700 further includes transmitting one or more cell IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the NWDAF. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 31.
[0247] In some aspects, the method 2700 further includes transmitting one or more base station IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the NWDAF. In some cases, the operations of this step refer to, or may be performed by, circuitry for transmitting and / or code for transmitting as described with reference to FIG. 31.
[0248] In one aspect, method 2700, or any aspect related to it, may be performed by an apparatus, such as communications device 3100 of FIG. 31, which includes various components operable, configured, or adapted to perform the method 2700. Communications device 3100 is described below in further detail.
[0249] Note that FIG. 27 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure. Example Communications Device (s)
[0250] FIG. 28 depicts aspects of an example communications device 2800. In some aspects, communications device 2800 is a user equipment, such as UE 104 described above with respect to FIGS. 1 and 3.
[0251] The communications device 2800 includes a processing system 2805 coupled to the transceiver 2845 (e.g., a transmitter and / or a receiver) . The transceiver 2845 is configured to transmit and receive signals for the communications device 2800 via the antenna 2850, such as the various signals as described herein. The processing system 2805 may be configured to perform processing functions for the communications device 2800, including processing signals received and / or to be transmitted by the communications device 2800.
[0252] The processing system 2805 includes one or more processors 2810. In various aspects, the one or more processors 2810 may be representative of one or more of receive processor 358, transmit processor 364, TX MIMO processor 366, and / or controller / processor 380, as described with respect to FIG. 3. The one or more processors 2810 are coupled to a computer-readable medium / memory 2825 via a bus 2840. In certain aspects, the computer-readable medium / memory 2825 is configured to store instructions (e.g., computer-executable code) that when executed by the one or more processors 2810, cause the one or more processors 2810 to perform the method 2400 described with respect to FIG. 24, or any aspect related to it. Note that reference to a processor performing a function of communications device 2800 may include one or more processors 2810 performing that function of communications device 2800.
[0253] In the depicted example, computer-readable medium / memory 2825 stores code (e.g., executable instructions) , such as code for transmitting 2830 and code for receiving 2835. Processing of the code for transmitting 2830 and code for receiving 2835 may cause the communications device 2800 to perform the method 2400 described with respect to FIG. 24, or any aspect related to it.
[0254] The one or more processors 2810 include circuitry configured to implement (e.g., execute) the code stored in the computer-readable medium / memory 2825, including circuitry such as circuitry for transmitting 2815 and circuitry for receiving 2820. Processing with circuitry for transmitting 2815 and circuitry for receiving 2820 may cause the communications device 2800 to perform the method 2400 described with respect to FIG. 24, or any aspect related to it.
[0255] Various components of the communications device 2800 may provide means for performing the method 2400 described with respect to FIG. 24, or any aspect related to it. For example, means for transmitting, sending or outputting for transmission may include transceivers 354 and / or antenna (s) 352 of the UE 104 illustrated in FIG. 3 and / or the transceiver 2845 and the antenna 2850 of the communications device 2800 in FIG. 28. Means for receiving or obtaining may include transceivers 354 and / or antenna (s) 352 of the UE 104 illustrated in FIG. 3 and / or the transceiver 2845 and the antenna 2850 of the communications device 2800 in FIG. 28.
[0256] FIG. 29 depicts aspects of an example communications device 2900. In some aspects, communications device 2900 is a network entity, such as BS 102 of FIGS. 1 and 3, or a disaggregated base station as discussed with respect to FIG. 2.
[0257] The communications device 2900 includes a processing system 2905 coupled to the transceiver 2965 (e.g., a transmitter and / or a receiver) and / or a network interface 2975. The transceiver 2965 is configured to transmit and receive signals for the communications device 2900 via the antenna 2970, such as the various signals as described herein. The network interface 2975 is configured to obtain and send signals for the communications device 2900 via communication link (s) , such as a backhaul link, midhaul link, and / or fronthaul link as described herein, such as with respect to FIG. 2. The processing system 2905 may be configured to perform processing functions for the communications device 2900, including processing signals received and / or to be transmitted by the communications device 2900.
[0258] The processing system 2905 includes one or more processors 2910. In various aspects, one or more processors 2910 may be representative of one or more of receive processor 338, transmit processor 320, TX MIMO processor 330, and / or controller / processor 340, as described with respect to FIG. 3. The one or more processors 2910 are coupled to a computer-readable medium / memory 2935 via a bus 2960. In certain aspects, the computer-readable medium / memory 2935 is configured to store instructions (e.g., computer-executable code) that when executed by the one or more processors 2910, cause the one or more processors 2910 to perform the method 2500 described with respect to FIG. 25, or any aspect related to it. Note that reference to a processor of communications device 2900 performing a function may include one or more processors 2910 of communications device 2900 performing that function.
[0259] In the depicted example, the computer-readable medium / memory 2935 stores code (e.g., executable instructions) , such as code for receiving 2940, code for transmitting 2945, code for tagging 2950, and code for forwarding 2955. Processing of the code for receiving 2940, code for transmitting 2945, code for tagging 2950, and code for forwarding 2955 may cause the communications device 2900 to perform the method 2500 described with respect to FIG. 25, or any aspect related to it.
[0260] The one or more processors 2910 include circuitry configured to implement (e.g., execute) the code stored in the computer-readable medium / memory 2935, including circuitry such as circuitry for receiving 2915, circuitry for transmitting 2920, circuitry for tagging 2925, and circuitry for forwarding 2930. Processing with circuitry for receiving 2915, circuitry for transmitting 2920, circuitry for tagging 2925, and circuitry for forwarding 2930 may cause the communications device 2900 to perform the method 2500 described with respect to FIG. 25, or any aspect related to it.
[0261] Various components of the communications device 2900 may provide means for performing the method 2500 described with respect to FIG. 25, or any aspect related to it. Means for transmitting, sending or outputting for transmission may include transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3 and / or the transceiver 2965 and the antenna 2970 of the communications device 2900 in FIG. 29. Means for receiving or obtaining may include transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3 and / or the transceiver 2965 and the antenna 2970 of the communications device 2900 in FIG. 29.
[0262] FIG. 30 depicts aspects of an example communications device 3000. In some aspects, communications device 3000 is a training entity, such a UE-side OTT server.
[0263] The communications device 3000 includes a processing system 3005 coupled to the transceiver 3075 (e.g., a transmitter and / or a receiver) and / or a network interface 3085. The transceiver 3075 is configured to transmit and receive signals for the communications device 3000 via the antenna 3080, such as the various signals as described herein. The network interface 3085 is configured to obtain and send signals for the communications device 3000 via communication link (s) , such as a backhaul link, midhaul link, and / or fronthaul link as described herein, such as with respect to FIG. 2. The processing system 3005 may be configured to perform processing functions for the communications device 3000, including processing signals received and / or to be transmitted by the communications device 3000.
[0264] The processing system 3005 includes one or more processors 3010. In various aspects, one or more processors 3010 may be representative of one or more of receive processor 338, transmit processor 320, TX MIMO processor 330, and / or controller / processor 340, as described with respect to FIG. 3. The one or more processors 3010 are coupled to a computer-readable medium / memory 3040 via a bus 3070. In certain aspects, the computer-readable medium / memory 3040 is configured to store instructions (e.g., computer-executable code) that when executed by the one or more processors 3010, cause the one or more processors 3010 to perform the method 2600 described with respect to FIG. 26, or any aspect related to it. Note that reference to a processor of communications device 3000 performing a function may include one or more processors 3010 of communications device 3000 performing that function.
[0265] In the depicted example, the computer-readable medium / memory 3040 stores code (e.g., executable instructions) , such as code for transmitting 3045, code for receiving 3050, code for training 3055, code for providing 3060, and code for obtaining 3065. Processing of the code for transmitting 3045, code for receiving 3050, code for training 3055, code for providing 3060, and code for obtaining 3065 may cause the communications device 3000 to perform the method 2600 described with respect to FIG. 26, or any aspect related to it.
[0266] The one or more processors 3010 include circuitry configured to implement (e.g., execute) the code stored in the computer-readable medium / memory 3040, including circuitry such as circuitry for transmitting 3015, circuitry for receiving 3020, circuitry for training 3025, circuitry for providing 3030, and circuitry for obtaining 3035. Processing with circuitry for transmitting 3015, circuitry for receiving 3020, circuitry for training 3025, circuitry for providing 3030, and circuitry for obtaining 3035 may cause the communications device 3000 to perform the method 2600 described with respect to FIG. 26, or any aspect related to it.
[0267] Various components of the communications device 3000 may provide means for performing the method 2600 described with respect to FIG. 26, or any aspect related to it. Means for transmitting, sending or outputting for transmission may include transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3 and / or the transceiver 3075 and the antenna 3080 of the communications device 3000 in FIG. 30. Means for receiving or obtaining may include transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3 and / or the transceiver 3075 and the antenna 3080 of the communications device 3000 in FIG. 30.
[0268] FIG. 31 depicts aspects of an example communications device 3100. In some aspects, communications device 3100 is an OAM entity.
[0269] The communications device 3100 includes a processing system 3105 coupled to the transceiver 3155 (e.g., a transmitter and / or a receiver) and / or a network interface 3165. The transceiver 3155 is configured to transmit and receive signals for the communications device 3100 via the antenna 3160, such as the various signals as described herein. The network interface 3165 is configured to obtain and send signals for the communications device 3100 via communication link (s) , such as a backhaul link, midhaul link, and / or fronthaul link as described herein, such as with respect to FIG. 2. The processing system 3105 may be configured to perform processing functions for the communications device 3100, including processing signals received and / or to be transmitted by the communications device 3100.
[0270] The processing system 3105 includes one or more processors 3110. In various aspects, one or more processors 3110 may be representative of one or more of receive processor 338, transmit processor 320, TX MIMO processor 330, and / or controller / processor 340, as described with respect to FIG. 3. The one or more processors 3110 are coupled to a computer-readable medium / memory 3130 via a bus 3150. In certain aspects, the computer-readable medium / memory 3130 is configured to store instructions (e.g., computer-executable code) that when executed by the one or more processors 3110, cause the one or more processors 3110 to perform the method 2700 described with respect to FIG. 27, or any aspect related to it. Note that reference to a processor of communications device 3100 performing a function may include one or more processors 3110 of communications device 3100 performing that function.
[0271] In the depicted example, the computer-readable medium / memory 3130 stores code (e.g., executable instructions) , such as code for receiving 3135, code for transmitting 3140, and code for sending 3145. Processing of the code for receiving 3135, code for transmitting 3140, and code for sending 3145 may cause the communications device 3100 to perform the method 2700 described with respect to FIG. 27, or any aspect related to it.
[0272] The one or more processors 3110 include circuitry configured to implement (e.g., execute) the code stored in the computer-readable medium / memory 3130, including circuitry such as circuitry for receiving 3115, circuitry for transmitting 3120, and circuitry for sending 3125. Processing with circuitry for receiving 3115, circuitry for transmitting 3120, and circuitry for sending 3125 may cause the communications device 3100 to perform the method 2700 described with respect to FIG. 27, or any aspect related to it.
[0273] Various components of the communications device 3100 may provide means for performing the method 2700 described with respect to FIG. 27, or any aspect related to it. Means for transmitting, sending or outputting for transmission may include transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3 and / or the transceiver 3155 and the antenna 3160 of the communications device 3100 in FIG. 31. Means for receiving or obtaining may include transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3 and / or the transceiver 3155 and the antenna 3160 of the communications device 3100 in FIG. 31. Example Clauses
[0274] Implementation examples are described in the following numbered clauses:
[0275] Clause 1: A method for wireless communication by a user equipment (UE) , comprising: transmitting signaling to a network entity, the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity and a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information.
[0276] Clause 2: The method of Clause 1, wherein information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination therefore; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0277] Clause 3: The method of any combination of Clauses 1-2, wherein the server is configured to train one or more CSF information encoder models, one or more CSF information private decoder models, one or more parameter sets associated with one or more CSF information encoder models, one or more parameter sets associated with one or more CSF information decoder models, or any combination thereof.
[0278] Clause 4: The method of any combination of Clauses 1-3, wherein the transmitting the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information is based on configuration received from the network entity.
[0279] Clause 5: The method of any combination of Clauses 1-4, wherein the transmitting the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information is based on a configuration received from the network entity when the network entity has one or more parameter sets, one or more training datasets, one or more reference models, or a combination thereof, to share with the server for training the one or more UE ML models.
[0280] Clause 6: The method of any combination of Clauses 1-5, wherein the network entity comprises: a serving cell; a gNB; an Operations, Administration, and Maintenance (OAM) entity; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0281] Clause 7: The method of any combination of Clauses 1-6, wherein the server comprises: a UE vendor-specific training server within a Mobile Network Operator (MNO) network of the network entity; a training server outside the MNO network; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0282] Clause 8: The method of any combination of Clauses 1-7, further comprising receiving, from the server, at least one of: a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with the CSF information encoder model, a trained parameter set associated with the CSF information decoder model, or any combination thereof.
[0283] Clause 9: A method for wireless communication by a network entity, comprising: receiving signaling from a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity to a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information; and transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, to the server based on the received information from the UE.
[0284] Clause 10: The method of Clause 9, wherein the network entity comprises: a serving cell; or a gNB.
[0285] Clause 11: The method of any combination of Clauses 9-10, wherein the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0286] Clause 12: The method of Clause 11, further comprising tagging the one or more parameter sets, the one or more training datasets, the one or more reference model, or the combination of thereof, with one or more of the associated IDs or pairing IDs.
[0287] Clause 13: The method of Clause 12, further comprising transmitting signaling configuring the UE to provide the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information.
[0288] Clause 14: The method of Clause 13, wherein the transmitting the signaling configuring the UE to provide the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information is in response to the network entity having one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, to share to the server.
[0289] Clause 15: The method of any combination of Clauses 9-14, further comprising forwarding the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information, to an Operations, Administration, and Maintenance (OAM) entity.
[0290] Clause 16: The method of any combination of Clauses 9-15, further comprising transmitting a request to the UE for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information, wherein the information is received from the UE in response to the request.
[0291] Clause 17: The method of Clause 16, further comprising receiving a request, from an Operations, Administration, and Maintenance (OAM) entity, for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information, wherein the transmitting the request to the UE is in response to receiving the request from the OAM entity.
[0292] Clause 18: The method of any combination of Clauses 9-17, further comprising transmitting signaling of one or more identifiers (IDs) associated with the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the server.
[0293] Clause 19: The method of Clause 18, wherein the one or more IDs comprises: one or more pairing IDs, one or more associated IDs, one or more UE vendor-specific IDs, one or more cell IDs, or one or more base station IDs associated with the one or more training datasets, the one or more reference models, or the combination thereof.
[0294] Clause 20: The method of any combination of Clauses 9-19, wherein the server comprises: a UE vendor specific training server within a Mobile Network Operator (MNO) network of the network entity; a training server outside the MNO network; Network Data Analytics Function (NWDAF) entity; or Model Training Logical Function (MTLF) entity at the NWDAF.
[0295] Clause 21: A method for wireless communication by a training server, comprising: transmitting signaling to a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination thereof, with the training server; receiving channel state feedback (CSF) information training data; receiving the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, from a network entity; training, based on the CSF information training data and on the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with a CSF information encoder model, or a trained parameter set associated with a CSF information decoder model, or any combination thereof; and providing, to the UE, the trained CSF information encoder model, the trained CSF information decoder model, the trained parameter set associated with the CSF information encoder model, or the trained parameter set associated with the CSF information decoder model, or the combination thereof.
[0296] Clause 22: The method of Clause 21, wherein information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the training server comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0297] Clause 23: The method of Clause 22, further comprising: receiving a request from a Network Data Analytics Function (NWDAF) for the CSF information training data, wherein the request includes at least one of: the one or more Associated IDs, the one or more pairing IDs, or the one or more UE vendor-specific IDs; and transmitting the CSF information training data to the NWDAF in response to the request.
[0298] Clause 24: The method of Clause 23, further comprising transmitting, to the NWDAF, one or more cell IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, or one or more base station IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof.
[0299] Clause 25: The method of any combination of Clauses 21-24, wherein the network entity comprises: a serving cell; a gNB; an Operations, Administration, and Maintenance (OAM) entity; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0300] Clause 26: The method of any combination of Clauses 21-25, wherein the training server comprises: a UE vendor-specific training server within a Mobile Network Operator (MNO) network of the network entity; a training server outside the MNO network; a Network Data Analytics Function (NWDAF) entity; or a Model Training Logical Function (MTLF) entity at the NWDAF.
[0301] Clause 27: The method of any combination of Clauses 21-26, further comprising obtaining a configuration of one or more identifiers (IDs) associated with the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof.
[0302] Clause 28: The method of Clause 27, wherein the one or more IDs comprise one or more base station (BS) IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof.
[0303] Clause 29: A method for wireless communication by an operations, administration, and maintenance (OAM) entity, comprising: receiving signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, with a server that is configured to train one or more user equipment (UE) machine learning (ML) models for channel state feedback (CSF) information; and transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server based on the received information from the UE.
[0304] Clause 30: The method of Clause 29, wherein the receiving the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server comprises receiving the signaling from a user equipment (UE) .
[0305] Clause 31: The method of any combination of Clauses 29-30, further comprising sending a request to a network entity for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, with the server, wherein the receiving the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, with the server is in response to the request.
[0306] Clause 32: The method of any combination of Clauses 29-31, wherein information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server that is configured to train one or more UE ML models for CSF information comprises: a uniform resource location (URL) of the server; an Internet Protocol (IP) address of the server; one or more UE vendor-specific identifiers (IDs) ; one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; or any combination thereof.
[0307] Clause 33: The method of Clause 32, further comprising: receiving a request from a Network Data Analytics Function (NWDAF) of a core network for the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, wherein the request includes the one or more Associated IDs or the one or more pairing IDs; and transmitting the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the NWDAF in response to the request from the NWDAF.
[0308] Clause 34: The method of Clause 32, further comprising: receiving CSF information training data from a user equipment (UE) ; receiving a request from a Network Data Analytics Function (NWDAF) for the CSF information training data, wherein the request includes the one or more Associated IDs or the one or more pairing IDs; and transmitting the CSF information training data to the NWDAF in response to the request from the NWDAF.
[0309] Clause 35: The method of any combination of Clauses 29-34, further comprising transmitting the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server, to a Network Data Analytics Function (NWDAF) .
[0310] Clause 36: The method of Clause 35, further comprising receiving request from the NWDAF for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server, wherein the transmitting the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, with the server, to the NWDAF is in response to the request from the NWDAF.
[0311] Clause 37: The method of Clause 35, further comprising transmitting one or more cell IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the NWDAF.
[0312] Clause 38: The method of Clause 35, further comprising transmitting one or more base station IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the NWDAF.
[0313] Clause 39: An apparatus, comprising: at least one memory comprising executable instructions; and at least one processor configured to execute the executable instructions and cause the apparatus to perform a method in accordance with any combination of Clauses 1-38.
[0314] Clause 40: An apparatus, comprising means for performing a method in accordance with any combination of Clauses 1-38.
[0315] Clause 41: A non-transitory computer-readable medium comprising executable instructions that, when executed by at least one processor of an apparatus, cause the apparatus to perform a method in accordance with any combination of Clauses 1-38.
[0316] Clause 42: A computer program product embodied on a computer-readable storage medium comprising code for performing a method in accordance with any combination of Clauses 1-38. Additional Considerations
[0317] The preceding description is provided to enable any person skilled in the art to practice the various aspects described herein. The examples discussed herein are not limiting of the scope, applicability, or aspects set forth in the claims. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects. For example, changes may be made in the function and arrangement of elements discussed without departing from the scope of the disclosure. Various examples may omit, substitute, or add various procedures or components as appropriate. For instance, the methods described may be performed in an order different from that described, and various actions may be added, omitted, or combined. Also, features described with respect to some examples may be combined in some other examples. For example, an apparatus may be implemented or a method may be practiced using any number of the aspects set forth herein. In addition, the scope of the disclosure is intended to cover such an apparatus or method that is practiced using other structure, functionality, or structure and functionality in addition to, or other than, the various aspects of the disclosure set forth herein. It should be understood that any aspect of the disclosure disclosed herein may be embodied by one or more elements of a claim.
[0318] The various illustrative logical blocks, modules and circuits described in connection with the present disclosure may be implemented or performed with a general purpose processor, a graphics processing unit (GPU) , a neural processing unit (NPU) , a digital signal processor (DSP) , an ASIC, a field programmable gate array (FPGA) or other programmable logic device (PLD) , discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any commercially available processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, a system on a chip (SoC) , or any other such configuration.
[0319] As used herein, “aprocessor, ” “at least one processor” or “one or more processors” generally refers to a single processor configured to perform one or multiple operations or multiple processors configured to collectively perform one or more operations. In the case of multiple processors, performance of the one or more operations could be divided amongst different processors, though one processor may perform multiple operations, and multiple processors could collectively perform a single operation. Similarly, “amemory, ” “at least one memory” or “one or more memories” generally refers to a single memory configured to store data and / or instructions, multiple memories configured to collectively store data and / or instructions.
[0320] In some cases, rather than actually transmitting a signal, an apparatus (e.g., a wireless node or device) may have an interface to output the signal for transmission. For example, a processor may output a signal, via a bus interface, to a radio frequency (RF) front end for transmission. Accordingly, a means for outputting may include such an interface as an alternative (or in addition) to a transmitter or transceiver. Similarly, rather than actually receiving a signal, an apparatus (e.g., a wireless node or device) may have an interface to obtain a signal from another device. For example, a processor may obtain (or receive) a signal, via a bus interface, from an RF front end for reception. Accordingly, a means for obtaining may include such an interface as an alternative (or in addition) to a receiver or transceiver.
[0321] While the present disclosure may describe certain operations as being performed by one type of wireless node, the same or similar operations may also be performed by another type of wireless node. For example, operations performed by a user equipment (UE) may also (or instead) be performed by a network entity (e.g., a base station or unit of a disaggregated base station) . Similarly, operations performed by a network entity may also (or instead) be performed by a UE.
[0322] Further, while the present disclosure may describe certain types of communications between different types of wireless nodes (e.g., between a network entity and a UE) , the same or similar types of communications may occur between same types of wireless nodes (e.g., between network entities or between UEs, in a peer-to-peer scenario) . Further, communications may occur in reverse order than described.
[0323] Means for obtaining, means for performing, means for determining, and means for providing may comprise one or more processors, such as one or more of the processors described above with reference to FIG. 25.
[0324] As used herein, a phrase referring to “at least one of” a list of items refers to any combination of those items, including single members. As an example, “at least one of: a, b, or c” is intended to cover a, b, c, a-b, a-c, b-c, and a-b-c, as well as any combination with multiples of the same element (e.g., a-a, a-a-a, a-a-b, a-a-c, a-b-b, a-c-c, b-b, b-b-b, b-b-c, c-c, and c-c-c or any other ordering of a, b, and c) .
[0325] As used herein, the term “determining” encompasses a wide variety of actions. For example, “determining” may include calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, a database or another data structure) , ascertaining and the like. Also, “determining” may include receiving (e.g., receiving information) , accessing (e.g., accessing data in a memory) and the like. Also, “determining” may include resolving, selecting, choosing, establishing and the like.
[0326] The methods disclosed herein comprise one or more actions for achieving the methods. The method actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of actions is specified, the order and / or use of specific actions may be modified without departing from the scope of the claims. Further, the various operations of methods described above may be performed by any suitable means capable of performing the corresponding functions. The means may include various hardware and / or software component (s) and / or module (s) , including, but not limited to a circuit, an application specific integrated circuit (ASIC) , or processor. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, or functions, whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.
[0327] The following claims are not intended to be limited to the aspects shown herein, but are to be accorded the full scope consistent with the language of the claims. Within a claim, reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more. ” Unless specifically stated otherwise, the term “some” refers to one or more. No claim element is to be construed under the provisions of 35 U. S. C. §112 (f) unless the element is expressly recited using the phrase “means for” . All structural and functional equivalents to the elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims.
Claims
1.A user equipment for wireless communication, the user equipment comprising:at least one memory storing computer-executable code; andat least one processor configured to execute the computer-executable code and cause the user equipment to transmit signaling to a network entity, the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, from the network entity to a server that is configured to train one or more UE machine learning (ML) models for channel state feedback (CSF) information.2.The user equipment of claim 1, wherein information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, from the network entity and the server that is configured to train the one or more UE ML models for CSF information comprises:a uniform resource location (URL) of the server;an Internet Protocol (IP) address of the server;one or more UE vendor-specific identifiers (IDs) ;one or more associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination therefore;one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; orany combination thereof.3.The user equipment of claim 1, wherein the server is configured to train one or more CSF information encoder models, one or more CSF information private decoder models, one or more parameter sets associated with one or more CSF information encoder models, one or more parameter sets associated with one or more CSF information decoder models, or any combination thereof.4.The user equipment of claim 1, wherein the at least one processor is configured to cause the user equipment to transmit the signaling including the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, from the network entity and the server that is configured to train the one or more UE ML models for CSF information based on configuration received from the network entity.5.The user equipment of claim 4, wherein the configuration is received from the network entity when the network entity has one or more parameter sets, one or more training datasets, one or more reference models, or a combination thereof, to share with the server for training the one or more UE ML models.6.The user equipment of claim 1, wherein the network entity comprises:a serving cell;a gNode B (gNB) ;an Operations, Administration, and Maintenance (OAM) entity;a Network Data Analytics Function (NWDAF) entity; ora Model Training Logical Function (MTLF) entity at the NWDAF.7.The user equipment of claim 1, wherein the server comprises:a UE vendor-specific training server within a Mobile Network Operator (MNO) network of the network entity;a training server outside the MNO network;a Network Data Analytics Function (NWDAF) entity; ora Model Training Logical Function (MTLF) entity at the NWDAF.8.The user equipment of claim 1, wherein the at least one processor is further configured to cause the user equipment to receive, from the server, at least one of: a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with the CSF information encoder model, a trained parameter set associated with the CSF information decoder model, or any combination thereof.9.A network entity, comprising:at least one memory storing computer-executable code; andat least one processor configured to execute the computer-executable code and cause the network entity to:receive signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, between the network entity and a server that is configured to train one or more user equipment (UE) machine learning (ML) models for channel state feedback (CSF) information; andtransmit the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, to the server based on the received information.10.The network entity of claim 9, wherein the network entity comprises:a serving cell;a gNode B (gNB) ,an Operations, Administration, and Maintenance (OAM) entity;a Network Data Analytics Function (NWDAF) entity; ora Model Training Logical Function (MTLF) entity at the NWDAF.11.The network entity of claim 9, wherein the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information comprises:a uniform resource location (URL) of the server;an Internet Protocol (IP) address of the server;one or more UE vendor-specific identifiers (IDs) ;one or more Associated IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination;one or more pairing IDs associated with the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination thereof; orany combination thereof.12.The network entity of claim 11, wherein the at least one processor is further configured to cause the network entity to tag the one or more parameter sets, the one or more training datasets, the one or more reference model, or the combination of thereof, with one or more of the associated IDs or pairing IDs.13.The network entity of claim 12, wherein the at least one processor is further configured to cause the network entity to transmit signaling configuring the UE to provide the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information.14.The network entity of claim 13, wherein the at least one processor is configured to cause the network entity to transit the signaling configuring the UE to provide the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, between the network entity and the server that is configured to train the one or more UE ML models for CSF information in response to the network entity having one or more parameter sets, one or more training datasets, one or more reference models, or any combination of thereof, to share to the server.15.The network entity of claim 9, wherein the at least one processor is further configured to cause the network entity to forward the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof to the server that is configured to train the one or more UE ML models for CSF information to an Operations, Administration, and Maintenance (OAM) entity.16.The network entity of claim 9, wherein the at least one processor is further configured to cause the network entity to transmit a request to the UE for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, with the server that is configured to train the one or more UE ML models for CSF information, wherein the information is received from the UE in response to the request.17.The network entity of claim 16, wherein the at least one processor is further configured to cause the network entity to receive a request, from an Operations, Administration, and Maintenance (OAM) entity, for the information for communicating the one or more parameter sets, the one or more training datasets, the one or more reference models, or the combination of thereof, with the server that is configured to train the one or more UE ML models for CSF information, wherein the at least one processor is configured to cause the network entity to transmit the request to the UE in response to receiving the request from the OAM entity.18.The network entity of claim 9, wherein the at least one processor is further configured to cause the network entity to transmit signalling of one or more identifiers (IDs) associated with the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, to the server.19.The network entity of claim 18, wherein the one or more IDs comprises: one or more pairing IDs, one or more associated IDs, one or more UE vendor-specific IDs, one or more cell IDs, or one or more base station IDs associated with the one or more training datasets, the one or more reference models, or the combination thereof.20.A training server, comprising:at least one memory storing computer-executable code; andat least one processor configured to execute the computer-executable code and cause the training server to:transmit signaling to a user equipment (UE) , the signaling including information for communicating one or more parameter sets, one or more training datasets, one or more reference models, or any combination thereof, from to the training server ;receive channel state feedback (CSF) information training data;receive the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, from a network entity;train, based on the CSF information training data and on the one or more parameters sets, the one or more training datasets, the one or more reference models, or the combination thereof, a trained CSF information encoder model, a trained CSF information decoder model, a trained parameter set associated with a CSF information encoder model, or a trained parameter set associated with a CSF information decoder model, or any combination thereof; andprovide, to the UE, the trained CSF information encoder model, the trained CSF information decoder model, the trained parameter set associated with the CSF information encoder model, or the trained parameter set associated with the CSF information decoder model, or the combination thereof.
Citation Information
Patent Citations
Method, device and computer readable medium for communication
CN117897923A
Method and apparatus for multiple-input multiple-output channel state information feedback
CN117957792A
Enhancements for Distributed Machine Learning Models in Wireless Communication Systems
US20240152752A1
Data collection procedure and model training
WO2023206380A1