Model pairing for backward compatible models
By managing ML model pairings and associations, the method addresses compatibility and efficiency issues in wireless communications, enhancing performance and reliability through optimized training and reduced signaling overhead.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- QUALCOMM INC
- Filing Date
- 2024-12-27
- Publication Date
- 2026-07-02
AI Technical Summary
Existing wireless communications systems face challenges in managing complex and dynamic environments, including signal attenuation and blockage, leading to inefficiencies in signal transmission and reception, power consumption, and compatibility issues with updated machine learning models.
The method involves obtaining signaling for machine learning model pairings and associations, allowing for efficient deployment and management of ML models across wireless nodes, ensuring backward compatibility and optimizing training processes for user equipment and network entities.
This approach enhances wireless communication performance by reducing signaling overhead, improving accuracy in channel state estimation, and ensuring seamless integration of updated models without unnecessary updates, thereby increasing efficiency and reliability.
Smart Images

Figure CN2024143023_02072026_PF_FP_ABST
Abstract
Description
MODEL PAIRING FOR BACKWARD COMPATIBLE MODELSINTRODUCTION
[0001] Aspects of the present disclosure relate to wireless communications, and more particularly, to managing deployment of machine learning (ML) models.
[0002] 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.
[0003] 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
[0004] One aspect provides a method for wireless communication at an apparatus (e.g., a user equipment (UE) or UE training server) . The method includes obtaining first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at the apparatus and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; and performing one or more actions based on the association.
[0005] Another aspect provides a method for wireless communication at an apparatus (e.g., a network entity) . The method includes generating first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at a wireless node and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; and outputting the first signaling.
[0006] 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.
[0007] The following description and the appended figures set forth certain features for purposes of illustration.BRIEF DESCRIPTION OF DRAWINGS
[0008] 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.
[0009] FIG. 1 depicts an example wireless communications network.
[0010] FIG. 2 depicts an example disaggregated base station architecture.
[0011] FIG. 3 depicts aspects of an example base station and an example user equipment.
[0012] FIGS. 4A, 4B, 4C, and 4D depict various example aspects of data structures for a wireless communications network.
[0013] FIG. 5 illustrates a general functional framework applied for AI-enabled RAN intelligence.
[0014] FIG. 6 depicts an example of an ML-based CSI feedback mechanism.
[0015] FIG. 7 depicts an example of encoder input and decoder output for the ML-based CSI feedback mechanism of FIG. 6.
[0016] FIG. 8 depicts an example deployment of an ML-based CSI feedback mechanism.
[0017] FIG. 9 depicts an example deployment of an ML-based CSI feedback mechanism.
[0018] FIG. 10 depicts an example deployment of an ML-based CSI feedback mechanism.
[0019] FIG. 11 depicts example pairing of ML models, in accordance with aspects of the present disclosure.
[0020] FIG. 12 depicts example deployment and updating of ML model pairings, in accordance with aspects of the present disclosure.
[0021] FIG. 13 depicts example deployment and updating of ML model pairings, in accordance with aspects of the present disclosure.
[0022] FIG. 14 depicts a method for wireless communications.
[0023] FIG. 15 depicts a method for wireless communications.
[0024] FIG. 16 depicts aspects of an example communications device.DETAILED DESCRIPTION
[0025] Aspects of the present disclosure relate to wireless communications, and more particularly, to managing deployment of machine learning (ML) models.
[0026] 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 reporting feedback so that performance can be improved. For example, performance can be improved by measuring the channel state and reporting feedback. Channel state information (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, which 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.
[0027] Artificial Intelligence (AI) or Machine learning (ML) represents an opportunity to potentially improve techniques for measuring channel state and reporting feedback. For example, machine learning 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 ML models to improve accuracy of channel estimation) . In some cases, ML-based CSI feedback may replace codebook-based processing by a CSI encoder and decoder (at least the decoder is needed) . In such cases, the encoder may be analogous to the PMI searching algorithm, while the decoder may be analogous to the PMI codebook used to translate the CSI reporting bits to a PMI code word.
[0028] Different types of training may be employed prior to deployment of such a CSI encoder and decoder. According to a first type, joint training of an encoder and decoder pair is performed at a single entity, such as a network-side training entity. According to a second type, training of an encoder and decoder pair may be performed jointly across multiple entities, such as a network-side training entity and a UE side training entity. According to a third type, training of an encoder and decoder of a pair may be performed separately (e.g., and / or sequentially) at the different entities. In the case of separate or sequential training, the training entity which performs training of (e.g., their part of) the model first may share model information to a different (e.g., other side –network-side or UE-side) training entity to facilitate the training of the other part of the model at the other side.
[0029] In certain scenarios, one or both sides involved in training may update one or both of the models (for encoding / decoding) . In such cases, the entities may need to share some information regarding the updated model. For example, if the network trains a new model, it may share information regarding the new model (e.g., updated either encoder parameters or encoder response) . The UE side may then train a compatible model which performs well with this model. The new model may be given a new pairing ID.
[0030] When the network updates (e.g., fine-tunes) an existing model in this manner, if the model is not compatible with existing UE side models, this model may essentially be equivalent to a new model. So, new training (e.g., offline-engineering) at the UE side may be performed, and a new pairing ID may be given. On the other hand, when the network fine-tunes an existing model that is backward compatible with existing UE models, then some UEs may not need to go through offline-engineering (and can continue to use their old models) . Other UEs (e.g., new UEs) may need to train new models via offline-engineering.
[0031] One particular challenge is how to achieve backwards compatibility and accommodate both types of UEs. While a new ID could be used for updated models, this approach comes with certain associated risks. For example one risk is that model updates may be performed too often and, without considering the difference compared to an existing model, UEs may decide to not update their model. On the alternative, all UEs may update their models, even when not necessary.
[0032] Aspects of the present disclosure provide mechanisms that may allow a UE to determine when to performing training to update UE-side models and / or when to continue using existing UE-side models, when a NW-side model is updated. As a result, the techniques presented herein may help facilitate ML model deployment and improve overall performance. Introduction to Wireless Communications Networks
[0033] 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.
[0034] FIG. 1 depicts an example of a wireless communications network 100, in which aspects described herein may be implemented.
[0035] 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.
[0036] 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.
[0037] 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.
[0038] 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.
[0039] BSs 102 may generally include: a NodeB, 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.
[0040] 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.
[0041] 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 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. 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. 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) .
[0042] 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.
[0043] 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.
[0044] 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) .
[0045] 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.
[0046] 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.
[0047] 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.
[0048] 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.
[0049] 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.
[0050] 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.
[0051] Wireless communication network 100 includes a machine learning component 199, which may perform the operations described herein related to machine learning timelines and / or machine learning concurrent processing. Wireless network 100 further includes a machine learning component 198, which may perform the operations described herein related to machine learning timelines and / or machine learning concurrent processing.
[0052] 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.
[0053] 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.
[0054] 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.
[0055] 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.
[0056] 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 3rd Generation Partnership Project (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.
[0057] 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.
[0058] 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.
[0059] 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.
[0060] 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) .
[0061] FIG. 3 depicts aspects of an example BS 102 and a UE 104.
[0062] 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.
[0063] Base station 102 includes controller / processor 340, which may be configured to implement various functions related to wireless communications. In the depicted example, controller / processor 340 includes machine learning component 241, which may be representative of the machine learning component 199 of FIG. 1. Notably, while depicted as an aspect of controller / processor 340, the machine learning component 341 may be implemented additionally or alternatively in various other aspects of base station 102 in other implementations.
[0064] Generally, UE 104 includes various processors (e.g., 358, 364, 366, and 380) , antennas 352a-r (collectively 352) , transceivers 354a-r (collectively 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.
[0065] User equipment 104 includes controller / processor 380, which may be configured to implement various functions related to wireless communications. In the depicted example, controller / processor 380 includes machine learning component 381, which may be representative of the machine learning component 198 of FIG. 1. Notably, while depicted as an aspect of controller / processor 380, the machine learning component 381 may be implemented additionally or alternatively in various other aspects of user equipment 104 in other implementations.
[0066] 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 HARQ 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.
[0067] 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) .
[0068] 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.
[0069] 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.
[0070] 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.
[0071] 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 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 SC-FDM) , and transmitted to BS 102.
[0072] 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.
[0073] Memories 342 and 382 may store data and program codes for BS 102 and UE 104, respectively.
[0074] Scheduler 344 may schedule UEs for data transmission on the downlink and / or uplink.
[0075] 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, RX MIMO detector 336, controller / processor 340, receive processor 338, scheduler 344, memory 342, and / or other aspects described herein.
[0076] 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.
[0077] In some aspects, a processor 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.
[0078] 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.
[0079] 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.
[0080] 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 single-carrier frequency division multiplexing (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.
[0081] 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.
[0082] 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.
[0083] 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.
[0084] 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.
[0085] 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) .
[0086] 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.
[0087] 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.
[0088] 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.
[0089] 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.
[0090] 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.
[0091] 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 HARQ ACK / 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. Introduction to mmWave Wireless Communications
[0092] In wireless communications, an electromagnetic spectrum is often subdivided into various classes, bands, channels, or other features. The subdivision is often 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.
[0093] In 5G NR two initial operating bands have been identified as frequency range designations FR1 (410 MHz –7125 MHz) and FR2 (24,250 MHz –71,000 MHz) . 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. It should be understood that although a portion of FR1 is greater than 6 GHz, FR1 is often referred to (interchangeably) as a “Sub-6 GHz” band in various documents and articles. A similar nomenclature issue sometimes occurs with regard to FR2, which is often referred to (interchangeably) as a “millimeter wave” band in documents and articles, despite being different from the extremely high frequency (EHF) band (30 GHz –300 GHz) which is identified by the International Telecommunications Union (ITU) as a “millimeter wave” band.
[0094] The frequencies between FR1 and FR2 are often referred to as mid-band frequencies. Recent 5G NR studies have identified an operating band for these mid-band frequencies as frequency range designation FR3 (7.125 GHz –24.25 GHz) . Frequency bands falling within FR3 may inherit FR1 characteristics and / or FR2 characteristics, and thus may effectively extend features of FR1 and / or FR2 into mid-band frequencies. In addition, higher frequency bands are currently being explored to extend 5G NR operation beyond 52.6 GHz. For example, three higher operating bands have been identified as frequency range designations FR4a or FR4-1 (52.6 GHz –71 GHz) , FR4 (52.6 GHz –114.25 GHz) , and FR5 (114.25 GHz –300 GHz) . Each of these higher frequency bands falls within the EHF band.
[0095] With the above aspects in mind, unless specifically stated otherwise, it should be understood that the term “sub-6 GHz” or the like if used herein may broadly represent frequencies that may be less than 6 GHz, may be within FR1, or may include mid-band frequencies. Further, unless specifically stated otherwise, it should be understood that the term “millimeter wave” or the like if used herein may broadly represent frequencies that may include mid-band frequencies, may be within FR2, FR4, FR4-aor FR4-1, and / or FR5, or may be within the EHF band.
[0096] Communications using mmWave / near mmWave radio frequency band (e.g., 3 GHz –300 GHz) may have higher path loss and a shorter range compared to lower frequency communications. As described above with respect to FIG. 1, a base station (e.g., 180) configured to communicate using mmWave / near mmWave radio frequency bands may utilize beamforming (e.g., 182) with a UE (e.g., 104) to improve path loss and range.
[0097] Further, as described herein, a UE may estimate a channel or generate channel state information in mmWave bands and / or other frequency bands using machine learning model (s) . Overview of AI / ML Functional Framework for RAN intelligence
[0098] The AI / ML functional framework 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.
[0099] 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.
[0100] 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, and output from an AI / ML model. In some cases, analysis of data needed at the model training function 504 and 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.
[0101] 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, formatting, and transformation) based on the training data delivered by the data collection function 502, if required.
[0102] The model training function 504 may provide model deployment / update data to the model inference 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.
[0103] 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, at times. The model inference function 506 may also be responsible for data preparation (e.g., data pre-processing and cleaning, formatting, and transformation) based on inference data delivered by the data collection function 502, at times.
[0104] 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.
[0105] 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.
[0106] 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, 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) may also be used at the model inference function 506.
[0107] 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.
[0108] In certain aspects, a UE or a BS may perform ML-based beam prediction using continuous measured or reported L1-RSRP in time domain. In some cases, a pre-trained deep neural network (DNN) model may be used for such ML-based predictive beam management.
[0109] Traditionally, beam qualities and failures are identified through measurement reports carried by relevant downlink (DL) and uplink (UL) reference signals (e.g., SSB, CSI-RS, RSRP) , which increase beam selection latency and beam management overhead, while at the same beam selection accuracy may be limited due to restrictions on power and overhead that may cause poor system performance.
[0110] Instead, the AI / ML based predictive beam management may reduce the amount of reference signal transmissions used to predict non-measured beam qualities and future possibility of beam blockage / failure. In predictive beam management, beam prediction may be a highly non-linear problem, which may be efficiently solved by the pre-trained DNN model that may predict future beam qualities, for example, based on a UE moving speed and trajectory that is difficult to be modeled through statistical processing methods. Overview of ML Model based Techniques
[0111] 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 and measured by another device in order to estimate channel state and to provide channel state information (CSI) feedback that is useful for optimizing wireless communications between the two devices.
[0112] However, owing 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 multiple-input multiple-output (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. One method of mitigating such issues is to implement machine learning models that may more accurately, and more efficiently, perform various functions related to channel state estimation and feedback.
[0113] For example, wireless communication systems 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 machine learning-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.
[0114] As another example, a CSI reporting configuration may rely on a precoding matrix indicator (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 machine learning-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 user equipment side 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 side is trained to recover the channel or the precoding matrix using the reported bits.
[0115] Thus, generally speaking, machine learning 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.
[0116] There are various options for what type of information to feedback as channel state feedback (CSF) 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.
[0117] For CSI feedback, a CSI report configuration may include a codebook. The codebook is used as a precoding matrix indicator (PMI) dictionary, from which the UE may select and report the best PMI codewords, using a sequence of bits to report the PMI.
[0118] As illustrated in the example block diagram 600 of FIG. 6, in some cases, AI -based CSI feedback may replace the codebook by a CSI encoder 602 (at the UE) and a decoder 604 at the network (e.g., gNB) . In some cases, the encoder may not be needed and just the decoder may be used. In general, the encoder is analogous to the PMI searching algorithm in current systems, while the decoder is generally analogous to the PMI codebook, which is used to translate the CSI reporting bits to a PMI code word.
[0119] As illustrated in the table 700 of FIG. 7, input to the decoder include a downlink channel matrix (H) , downlink precoders (V) , and 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 UE based on its demodulation filter.
[0120] 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. This may apply to 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.
[0121] ML based CSI schemes involving low-density CSI-RS patterns are examples of scenarios where information may be exchanged between a UE and gNB. For example, due to an association between low-density CSI-RS patterns and a UE channel estimation (CHEST) ML model. In some cases, a CHEST ML model may need to be changed. For example, when the environment changes, the CSI-RS pattern may be changed. In such cases, the UE may need assistance information from the gNB for training, adaptation, and the like.
[0122] 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 pattern could be a uniform RB pattern, a random RB pattern, or a learned RB pattern (e.g., some form of pattern may be used specifically for certain scenarios) .
[0123] Non-orthogonal cover codes may be used to multiplex N_t 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 channel matrix
[0124] 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 pattern could be a uniform RB pattern, a random RB pattern, or a learned RB pattern (e.g., some form of pattern may be used specifically for certain scenarios) .
[0125] Non-orthogonal cover codes may be used to multiplex N_t 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 channel matrix
[0126] 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.
[0127] The UE may update an ML model based on the first CSI-RS pattern. For example, as noted above, during training the UE may apply a loss function by comparing output of the ML model, generated based on measurements taken from CSI-RS transmitted according to the first CSI-RS pattern, 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) .
[0128] For example 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. For example, 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. Example Paired Model Training Collaboration
[0129] As noted above, different types of training may be employed prior to deployment of paired ML models used at a UE and network entity. In this context, a pair of ML models (a “pairing” ) refers to ML models that are compatible (e.g., have been trained to work well together) . As an example, a pairing may include a CSI encoder (deployable at a UE) and a CSI decoder (deployable at a network entity) .
[0130] Call flow diagram 800 of FIG. 8 illustrates an example of training of an ML model pairing at a single entity. As illustrated, entities involved in jointly training ML model pairings may include a UE side training entity 802 (e.g., a server) or a network side training entity 804 (e.g., a server) . In the illustrated example, the network side training entity 804 jointly trains a pairing of an encoder 806 (e.g., an ML-based CSI encoder) and a decoder 808 (e.g., an ML-based CSI encoder) , which may be referred to as an E-D pair. After the training process, the encoder and decoder may be transferred and deployed at the UE side and NW side, respectively, for inference.
[0131] In some cases, there may be collaboration between vendors (inter-vendor collaboration) outside the training process. Such collaboration may involve model structure and input format alignment before the training process, model delivery / transfer to the other side (in which the training did not occur) , model compilation and testing after the training process. While the example shows the E-D pair trained at the network side training entity 804, the E-D pair can also be trained at UE side training entity 802.
[0132] As illustrated in call flow diagram 900 of FIG. 9, a UE side training entity 902 and network side training entity 904 may jointly train an ML model pairing (encoder 906 and decoder 908) . With such joint training, there may be close inter-vendor collaboration during the training process. Such collaboration may involve, for example, an application programming interface (API) to exchange activation and gradient information. Even with this type, there may be various options for training. According to a first option, the encoder 906 and decoder 908 may be trained / updated concurrently in a distributed manner. According to a second option, the encoder 906 may be trained with a fixed (frozen parameters) decoder 908 or vice-versa.
[0133] As illustrated in call flow diagram 1000 of FIG. 10, a UE side training entity 1002 and network side training entity 1004 may separately train encoder 1006 and 1008, respectively. Separate training may involve relatively lighter inter-vendor collaboration outside the training process, such as exchanging datasets of (latent vector, precoder) pairs, where the datasets may account for payload quantization. While the illustrated example shows training starting at the network side training entity 1004, separate training could also first start at the UE side training entity 1002.
[0134] As noted above, there are various options for (and levels of) inter-vendor collaboration. A first option involves a fully standardized (reference model) model. According to this option, each (UE and NW) side may further develop their actual model, while maintaining compatibility with the standardized model. In this case, the development at the two sides can be independent, with little or no coordination.
[0135] Other options may involve a fully standardize dataset used for model training, sharing of model parameters, and / or model sharing from the network side to the UE side. In the case of model / model parameter sharing, information may be transferred to the UE side (training) server to facilitate UE side offline engineering. In the context of the present disclosure, offline engineering may refer to potential training / re-training, development / re-development of a different model, and / or offline testing. In some cases, entities may share encoder parameters, decoder parameters, or both. In some cases, information may be transferred to a UE device for inference directly. In some cases, an entire model may be transferred or parameters for a standardized model structure may be transferred.
[0136] One option may involve dataset sharing (e.g., from the network side to the UE side) , in order to facilitate UE side offline engineering. In such cases, the network may share encoder input / output, decoder input / output, or both.
[0137] In some cases, offline-engineering performed at two sides (independently without collaboration) may involve a fully specified reference encoder trained on synthetic data. In such cases, the UE side can implement the reference encoder directly, or can train an equivalent one using synthetic data and / or field data. Similarly, the network side may train their decoder to be compatible to the specified encoder (e.g., to perform well with the specified encoder) .
[0138] In some cases, offline-engineering may involve a fully specified reference decoder trained on synthetic data. In such cases, the UE side may train their encoder to be compatible to the specified decoder (to perform well with the specified decoder) . Similarly, the network side may implement the reference decoder directly or train an equivalent decoder using synthetic data and / or field data.
[0139] In some cases, offline-engineering may involve a fully specified reference encoder-decoder pair trained on synthetic data. In such cases, both sides can implement the respective model directly or train a compatible model. For example, the UE side can train an encoder compatible to the specified decoder, while the network side can train a decoder compatible to the specified encoder.
[0140] In some cases (involving dataset sharing) , the network side may share their target CSI and associated CSI feedback generated by their reference encoder. In such cases, the network side data can be data aggregated from multiple UE vendors. In cases involving model parameter sharing, the network side may share reference encoder parameters and their target CSI.
[0141] In certain cases, no specific model pairing or pairing identification (ID) may be needed. Instead, multiple models may be specified considering different usage or computation constraints. In some cases, the network may configure the ID of a specified encoder and / or decoder for inference. Upon receiving the configuration, the UE side may adopt the corresponding encoder, equivalent encoder, or compatible encoder developed via offline-engineering.
[0142] In certain cases, however, model pairing and a pairing identification mechanism may be needed. For example, when sharing a dataset or encoder parameter, a pairing ID may be needed to label the shared dataset / encoder parameter (e.g., to identify the corresponding encoder and decoder) . In such cases, if there are multiple models, different pairing IDs may be assigned. After training, a UE may acknowledge the applicability of those pairing IDs. For an inference phase, a corresponding ID may be used (e.g., configured for inference) . Model Pairing for Backward Compatible Models
[0143] In general, when a network trains a new model, the network shares information of the new model (e.g., model parameters and / or model response) . The UE side may have to train a compatible model which performs well with this (newly trained) model. This new model may be assigned a new pairing ID.
[0144] When NW updates (e.g., fine-tunes) an existing model, if the model is not compatible with existing UE side models, this model should be treated as being equivalent to a new model. In other words, new offline-engineering should be performed at UE side and a new pairing ID should be assigned.
[0145] On the other hand, when the network fine-tunes an existing model, and this model is backward compatible with existing UE models, then some UEs (e.g., existing UEs that have previously trained the compatible models) may not need to go through offline-engineering (they can stay with their old model) . A new or updated model may be backward compatible with an existing UE-side model, for example, if it was trained with a model similar to the existing UE-side model (e.g., with similar model parameters) or with a similar dataset as used to train the existing UE-side model. New UEs (that have not already trained compatible models) , on the other hand, may need to train new models via offline-engineering.
[0146] One design consideration is whether the backward compatible model should be assigned a new pairing ID. This may be understood by considering some different example scenarios.
[0147] In a first scenario, a new UE vendor or a new UE device (e.g., a new version of a phone) comes into market, the network may collect data from these new UEs. Then the network may train a backward compatible model (e.g., this model is compatible with the encoders used by older UEs (e.g., previous versions) . In this scenario, old / existing UEs would not need (to train) a new model, but the new UEs would need to train a UE-side model to be compatible to the new NW side model.
[0148] In a second scenario, a fully standardized model may be designed (e.g., and / or shared) for relatively low (e.g., minimum) performance. Some infrastructure vendors, however, may want to achieve better than the minimum performance. As a result, such vendors may update the fully standardized model (e.g., using field data) and use it for actual inference in real world scenarios, while still staying compatible with UEs who only maintain the standardized model (or equivalent model) . So, for some UEs that just want to ensure minimum performance, those UEs would not have to update their model (e.g., encoder) , while some other UEs can further develop their model to be compatible with the enhanced model (and achieve better performance) .
[0149] While new pairing IDs may be assigned in each of the scenarios described above, always assigning new pairing IDs may not be feasible due to scalability issues. For example, in some cases, if new pairing IDs are always used and the UEs do not know how different the new model is, the UEs may decide to not update their models (potentially missing out on performance improvements) . As an alternative, UEs may decide to always update their models, even though the updates may not be necessary (e.g., they have already trained a model the new network side model is backward compatible with) . The additional training (offsite engineering) may take substantial time and resources, while resulting in little or no improvement in performance.
[0150] Aspects of the present disclosure, however, provide a mechanism that may help determine when it is appropriate to perform additional training at the UE (or UE training server side) when the network indicates a model update. For example, aspects of the present disclosure propose utilizing an association between one or more models of a first pairing (e.g., of an encoder and decoder) with one or more models of a second pairing. Given the association, a UE may make an informed decision on when to perform additional model training and / or when to use an existing model.
[0151] As will be described in greater detail below, such an association may allow a UE (or UE side training server) to determine when to update UE-side models (e.g., perform training) and / or when the UE can continue using existing models, when a NW-side model is updated, the signaling mechanisms proposed herein may help facilitate ML model deployment and improve overall performance. The signaling mechanisms may include an explicit indication (e.g., that a pairing ID is associated / related to a “source” or “root” model ID for a model deemed related) or an implicit indication (e.g., via a two-part pairing ID) .
[0152] The model pairing association proposed herein may be understood with reference to diagram 1100 of FIG. 11.
[0153] The example assumes a first pairing of an encoder 1106-1 (UE side model E1) and a decoder 1108-1 (network side model D1) has been previously trained. As illustrated, this first pairing may be identified by a first two-part pairing ID 1.0 or via an indication of a first pairing ID (ID1) that is sourced by itself (e.g., by first pairing ID (ID1)) .
[0154] The example further assumes that of the network side model D1 is updated (fine-tuned) to generate a new decoder 1108-2 (network side model D2) compatible with a different encoder 1106-2 (UE-side model E2) . As indicated at 1110, this new model D2 is still compatible with the existing UE side model E1, which is compatible with (e.g., was trained to perform well with) the original NW-side model D1.
[0155] As noted above, the association between this new model D2 and the existing model (D1) compatible with the existing UE side model (E1) may be indicated according to various alternatives.
[0156] According to a first alternative, the new model may be assigned a new ID and the association may be indicated via an explicit configuration of a root or source model for the new ID (e.g., new ID2 may be sourced by ID1 as shown in FIG. 11) .
[0157] An example of this first alternative is depicted in call flow diagram 1200 of FIG. 12. As illustrated, there are two types of occasions when pairing IDs may be provided. The first occasion is for the pairing procedure during a training phase. The second occasion is for inference configuration.
[0158] A network side entity may perform data collection from a first UE (UE1) or a first set of UEs (e.g., a set of existing UEs on the market) . As indicated at 1210, a first network side model (decoder D1) is trained via offline-engineering, this decoder D1 may be developed together with a reference encoder (denoted E1-ref) . A first model pairing ID, ID1, is assigned along with an explicit indication the pairing ID is sourced by ID1.
[0159] As indicated at 1212, this model pairing ID and the source may be indicating to UE1, with parameters or input / output data for the reference encoder E1_ref.
[0160] If the model pairing ID and the source indicates a compatible encoder has not yet been trained for UE1, UE-side offline engineering may be performed (e.g. at a UE training server) to train a UE-side model (encoder E1) . In other words, if the UE has not developed an encoder using the information shared along with this pairing ID and its source, UE-side offline engineering may be performed. Specifically, based on the shared input / output data of E1_ref or the parameters of E1_ref or the model of E1_ref, UE may first train a reference decoder (D1_ref) , as indicated at 1220. Then, the UE may develop or train their encoder E1 based on the reference decoder D1_ref. After the UE-side offline-engineering, the UE-side may indicate to the NWside that the pairing ID1 is applicable at the UE. If the model pairing ID and the ID source indicates a compatible encoder has already been trained for UE1, UE-side training may not need to be performed (and UE1 can use a previously trained model) . In other words, if the UE has developed an encoder using the information shared along with this pairing ID and its source, UE-side offline engineering may not be needed.
[0161] The network may subsequently perform data collection from a second UE (UE2) , which could be a new UE after an initial deployment or pairing process. As indicated at 1230, the network side may update the network-side model D1 to obtain a second network-side model D2 via offline-engineering along with a second reference encoder E2_ref.
[0162] A second model pairing ID, ID2, is assigned along with an explicit indication the pairing ID is also sourced by ID1. This second model pairing ID and the source ID1 may be indicating to UE2, as indicated at 1232, with parameters or input / output data for the second reference encoder E2_ref. If the model pairing ID and the source indicates a compatible encoder has not yet been trained for UE2, UE-side offline engineering may be performed to train a UE-side model (encoder E2) . Specifically, based on the shared input / output data of E2_ref or the parameters of E2_ref or the model of E2_ref, UE may first train a reference decoder (D2_ref) , as indicated at 1240. Then, the UE may develop or train their encoder E2 based on the reference decoder D2_ref. After the UE side offline-engineering, the UE side may indicate to the NW side that the pairing ID2 is applicable at the UE. If the model pairing ID and the source indicates a compatible encoder has already been trained for UE2, UE-side training may not need to be performed (and UE2 can use a previously trained model) .
[0163] After training, the network may configure the UEs for inference. As illustrated, the inference configuration for UE1 and UE2 may configure the second pairing ID, ID2 (e.g., along with its source ID, ID1) , for inference. Based on the second pairing ID, ID2, and the indication that ID2 is sourced by ID1, UE1 may choose to use the previously trained encoder E1 for inference even though the network is using D2.
[0164] According to a second alternative, the new model may be assigned a two-part (hierarchical) pairing ID consisting of a multiple ID levels. For example, in a pairing ID in the format of x. y, the first parameter x may be a first level ID and the second parameter y may be the second level ID.
[0165] This format may be interpreted to indicate that the new model is compatible to any model with a pairing ID x. z paired with the same first level ID (same value of x) and a lower second level ID (e.g., where z<y) . In other words, the association between models of the first and second pairings (or lack thereof) is implicitly conveyed via the 2nd level of the hierarchical pairing ID. For example, if two models or pairs are associated with a different first level ID, then they are not associated with each other.
[0166] An example of this second alternative, using a hierarchical pairing ID, is depicted in call flow diagram 1300 of FIG. 13.
[0167] After a first network side model (decoder D1) is trained via offline-engineering based on a reference encoder (E1-ref) , a hierarchical model pairing ID, ID1.0 (e.g., x. y where x=1, y=0) is assigned. As indicated at 1312, for UE-side training, this hierarchical model pairing ID (ID1.0) are indicated to UE1, with parameters or input / output data for the reference encoder E1_ref.
[0168] After the network side updates the network-side model D1 to obtain a second network-side model D2 via offline-engineering based on a second reference encoder E2_ref, a second hierarchical model pairing ID, ID1.1, is assigned. As indicated at 1332, this second hierarchical model pairing ID1.1 is indicated to UE2, with parameters or input / output data for the reference encoder E2_ref.
[0169] After training, the network may configure the UEs for inference. As illustrated, the inference configuration for UE1 and UE2 may configure the second pairing ID, ID1.1, for inference. Based on the second pairing ID, ID1.1, UE1 may choose to use the previously trained encoder E1 for inference even though the network is using D2 (because the first pairing ID (ID1.0) has a same first level ID (x=1) as the second pairing ID (ID1.1) , and a lower second level ID (y=0 < 1) .
[0170] What actions are performed at the UE sides depends on when a pairing ID is received, the value of the pairing ID, and whether a UE side model, compatible with a corresponding network side model, has already been trained.
[0171] For example, a UE may receive an ID y sourced by ID z per the first alternative depicted in FIG. 11. Upon receiving the pairing ID for training, if the UE-side already has already developed a model paired by ID z, then the UE-side may decide that it does not need to train a new model (using the shared reference encoder parameter or input / output) at least for a period of time (e.g., until receiving a new pairing ID indicating the UE has not already developed a compatible UE-side model) . In this case, for the inference phase, the UE may accept (and use) either ID y or ID z as its inference configuration. Otherwise, if the UE-side has not already developed a model paired by ID z, then the UE may choose to develop a new model (e.g., with offline engineering) identified by ID y.
[0172] Similarly, a UE may receive an ID x. y per the second alternative depicted in FIG. 12. If the UE-side already develops a model paired with ID x. z with z<y, then UE-side does not need to train a new model using the shared reference encoder parameter or input / output. In this case, for the inference phase, the UE may accept (and use) either ID x.y or ID x. z as its inference configuration. Otherwise, if the UE-side has not already developed a compatible model (e.g., if z>y) the UE may choose to develop a new model.
[0173] Other alternatives are also possible for conveying association between models of first and second model pairings. For example, some type of mapping (e.g., a list or table) of associated models may be conveyed. In such a case, when a UE receives signaling of a new model pairing (e.g., during a training phase) , it may consult the mapping to determine whether it should update (train) a new model or whether it can use an existing model (compatible with a model of the new pairing) . Example Operations
[0174] FIG. 14 shows an example of a method 1400 of wireless communication at an apparatus, such as a UE 104 of FIGS. 1 and 3.
[0175] Method 1400 begins at step 1405 with obtaining first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at the apparatus and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing. 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. 16.
[0176] Method 1400 then proceeds to step 1410 with performing one or more actions based on the association. In some cases, the operations of this step refer to, or may be performed by, circuitry for performing and / or code for performing as described with reference to FIG. 16.
[0177] In some aspects, the first ML model is usable at the apparatus to encode channel state feedback (CSF) .
[0178] In some aspects, the one or more actions comprise refraining, for a time period, from training the first ML model, if a third ML model associated with the second pairing has been trained by the apparatus.
[0179] In some aspects, the one or more actions comprise training the first ML model using the first parameters, if the apparatus lacks an ML model associated with the second pairing that has been trained by the apparatus.
[0180] In some aspects, the first parameters comprise at least one of: shared reference encoder parameters or training data.
[0181] In some aspects, the first signaling comprises a configuration for inference; and the one or more actions comprise using at least one of the first ML model or third ML model to perform inference.
[0182] In some aspects, the one or more actions comprise using the first ML model or the third ML model based on availability.
[0183] In some aspects, the first parameters comprise a first pairing identifier (ID) associated with the first pairing.
[0184] In some aspects, the first signaling further indicates a first pairing identifier (ID) associated with the first pairing; and the association is indicated via a second pairing ID associated with the second pairing.
[0185] In some aspects, the second pairing ID is associated with a root ID or source ID.
[0186] In some aspects, the first signaling further indicates a first hierarchical pairing identifier (ID) associated with the first pairing, wherein the first hierarchical pairing identifier (ID) has at least a first level ID part and a second level ID part.
[0187] In some aspects, the association is indicated via the second level ID part of the first hierarchical pairing ID if a second hierarchical pairing ID associated with the second pairing has a same value first level ID part as the first hierarchical pairing ID.
[0188] In some aspects, the second hierarchical pairing ID has a value of a second level ID part that is less than a value of the second level ID part of the first hierarchical pairing ID.
[0189] In one aspect, method 1400, or any aspect related to it, may be performed by an apparatus, such as communications device 1600 of FIG. 16, which includes various components operable, configured, or adapted to perform the method 1400. Communications device 1600 is described below in further detail.
[0190] Note that FIG. 14 is just one example of a method, and other methods including fewer, additional, or alternative steps are possible consistent with this disclosure.
[0191] FIG. 15 shows an example of a method 1500 of wireless communication at an apparatus, such as a BS 102 of FIGS. 1 and 3, or a disaggregated base station as discussed with respect to FIG. 2.
[0192] Method 1500 begins at step 1505 with generating first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at a wireless node and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing. In some cases, the operations of this step refer to, or may be performed by, circuitry for generating and / or code for generating as described with reference to FIG. 16.
[0193] Method 1500 then proceeds to step 1510 with outputting the first signaling. In some cases, the operations of this step refer to, or may be performed by, circuitry for outputting and / or code for outputting as described with reference to FIG. 16.
[0194] In some aspects, the first ML model is usable at the wireless node to decode channel state feedback (CSF) .
[0195] In some aspects, the first parameters comprise at least one of: shared reference encoder parameters or training data.
[0196] In some aspects, the first parameters comprise a first pairing identifier (ID) associated with the first pairing.
[0197] In some aspects, the first signaling further indicates a first pairing identifier (ID) associated with the first pairing; and the association is indicated via a second pairing ID associated with the second pairing.
[0198] In some aspects, the second pairing ID is associated with a root ID or source ID.
[0199] In some aspects, the first signaling further indicates a first hierarchical pairing identifier (ID) associated with the first pairing, wherein the first hierarchical pairing identifier (ID) has at least a first level ID part and a second level ID part.
[0200] In some aspects, the association is indicated via the second level part of the first hierarchical pairing ID if a second hierarchical pairing ID associated with the second pairing has a same value first level ID part as the first hierarchical pairing ID.
[0201] In some aspects, the second hierarchical pairing ID has a value of a second level ID part that is less than a value of the second level ID part of the first hierarchical pairing ID.
[0202] In one aspect, method 1500, or any aspect related to it, may be performed by an apparatus, such as communications device 1600 of FIG. 16, which includes various components operable, configured, or adapted to perform the method 1500. Communications device 1600 is described below in further detail.
[0203] Note that FIG. 15 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)
[0204] FIG. 16 depicts aspects of an example communications device 1600. In some aspects, communications device 1600 is a user equipment, such as UE 104 described above with respect to FIGS. 1 and 3. In some aspects, communications device 1600 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.
[0205] The communications device 1600 includes a processing system 1602 coupled to the transceiver 1638 (e.g., a transmitter and / or a receiver) . In some aspects (e.g., when communications device 1600 is a network entity) , processing system 1602 may be coupled to a network interface 1642 that is configured to obtain and send signals for the communications device 1600 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 transceiver 1638 is configured to transmit and receive signals for the communications device 1600 via the antenna 1640, such as the various signals as described herein. The processing system 1602 may be configured to perform processing functions for the communications device 1600, including processing signals received and / or to be transmitted by the communications device 1600.
[0206] The processing system 1602 includes one or more processors 1604. In various aspects, the one or more processors 1604 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. In various aspects, one or more processors 1604 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 1604 are coupled to a computer-readable medium / memory 1620 via a bus 1636. In certain aspects, the computer-readable medium / memory 1620 is configured to store instructions (e.g., computer-executable code) that when executed by the one or more processors 1604, cause the one or more processors 1604 to perform the method 1400 described with respect to FIG. 14, or any aspect related to it; and the method 1500 described with respect to FIG. 15, or any aspect related to it. Note that reference to a processor performing a function of communications device 1600 may include one or more processors 1604 performing that function of communications device 1600.
[0207] In the depicted example, computer-readable medium / memory 1620 stores code (e.g., executable instructions) , such as code for obtaining 1622, code for performing 1624, code for refraining 1626, code for training 1628, code for using 1630, code for generating 1632, and code for outputting 1634. Processing of the code for obtaining 1622, code for performing 1624, code for refraining 1626, code for training 1628, code for using 1630, code for generating 1632, and code for outputting 1634 may cause the communications device 1600 to perform the method 1400 described with respect to FIG. 14, or any aspect related to it; and the method 1500 described with respect to FIG. 15, or any aspect related to it.
[0208] The one or more processors 1604 include circuitry configured to implement (e.g., execute) the code stored in the computer-readable medium / memory 1620, including circuitry for obtaining 1606, circuitry for performing 1608, circuitry for refraining 1610, circuitry for training 1612, circuitry for using 1614, circuitry for generating 1616, and circuitry for outputting 1618. Processing with circuitry for obtaining 1606, circuitry for performing 1608, circuitry for refraining 1610, circuitry for training 1612, circuitry for using 1614, circuitry for generating 1616, and circuitry for outputting 1618 may cause the communications device 1600 to perform the method 1400 described with respect to FIG. 14, or any aspect related to it; and the method 1500 described with respect to FIG. 15, or any aspect related to it.
[0209] Various components of the communications device 1600 may provide means for performing the method 1400 described with respect to FIG. 14, or any aspect related to it; and the method 1500 described with respect to FIG. 15, 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, transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3, and / or the transceiver 1638 and the antenna 1640 of the communications device 1600 in FIG. 16. Means for receiving or obtaining may include transceivers 354 and / or antenna (s) 352 of the UE 104 illustrated in FIG. 3, transceivers 332 and / or antenna (s) 334 of the BS 102 illustrated in FIG. 3, and / or the transceiver 1638 and the antenna 1640 of the communications device 1600 in FIG. 16. Example Clauses
[0210] Implementation examples are described in the following numbered clauses:
[0211] Clause 1: A method for wireless communication at an apparatus (e.g., a user equipment (UE) or UE training server) , comprising: obtaining first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at the apparatus and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; and performing one or more actions based on the association.
[0212] Clause 2: The method of Clause 1, wherein the first ML model is usable at the apparatus to encode channel state feedback (CSF) .
[0213] Clause 3: The method of any one of Clauses 1-2, wherein: the one or more actions comprise refraining, for a time period, from training the first ML model, if a third ML model associated with the second pairing has been trained by the apparatus.
[0214] Clause 4: The method of any one of Clauses 1-3, wherein the one or more actions comprise training the first ML model using the first parameters, if the apparatus lacks an ML model associated with the second pairing that has been trained by the apparatus.
[0215] Clause 5: The method of Clause 4, wherein the first parameters comprise at least one of: shared reference encoder parameters or training data.
[0216] Clause 6: The method of any one of Clauses 1-5, wherein: the first signaling comprises a configuration for inference; and the one or more actions comprise using at least one of the first ML model or third ML model to perform inference.
[0217] Clause 7: The method of Clause 6, wherein the one or more actions comprise using the first ML model or the third ML model based on availability.
[0218] Clause 8: The method of Clause 6, wherein the first parameters comprise a first pairing identifier (ID) associated with the first pairing.
[0219] Clause 9: The method of any one of Clauses 1-8, wherein: the first signaling further indicates a first pairing identifier (ID) associated with the first pairing; and the association is indicated via a second pairing ID associated with the second pairing.
[0220] Clause 10: The method of Clause 9, wherein the second pairing ID is associated with a root ID or source ID.
[0221] Clause 11: The method of any one of Clauses 1-10, wherein the first signaling further indicates a first hierarchical pairing identifier (ID) associated with the first pairing, wherein the first hierarchical pairing identifier (ID) has at least a first level ID part and a second level ID part.
[0222] Clause 12: The method of Clause 11, wherein: the association is indicated via the second level ID part of the first hierarchical pairing ID if a second hierarchical pairing ID associated with the second pairing has a same value first level ID part as the first hierarchical pairing ID.
[0223] Clause 13: The method of Clause 12, wherein the second hierarchical pairing ID has a value of a second level ID part that is less than a value of the second level ID part of the first hierarchical pairing ID.
[0224] Clause 14: A method for wireless communication at an apparatus (e.g., a network entity) , comprising: generating first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at a wireless node and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; and outputting the first signaling.
[0225] Clause 15: The method of Clause 14, wherein the first ML model is usable at the wireless node to decode channel state feedback (CSF) .
[0226] Clause 16: The method of any one of Clauses 14-15, wherein the first parameters comprise at least one of: shared reference encoder parameters or training data.
[0227] Clause 17: The method of any one of Clauses 14-16, wherein the first parameters comprise a first pairing identifier (ID) associated with the first pairing.
[0228] Clause 18: The method of any one of Clauses 14-17, wherein: the first signaling further indicates a first pairing identifier (ID) associated with the first pairing; and the association is indicated via a second pairing ID associated with the second pairing.
[0229] Clause 19: The method of Clause 18, wherein the second pairing ID is associated with a root ID or source ID.
[0230] Clause 20: The method of any one of Clauses 14-19, wherein the first signaling further indicates a first hierarchical pairing identifier (ID) associated with the first pairing, wherein the first hierarchical pairing identifier (ID) has at least a first level ID part and a second level ID part.
[0231] Clause 21: The method of Clause 20, wherein: the association is indicated via the second level part of the first hierarchical pairing ID if a second hierarchical pairing ID associated with the second pairing has a same value first level ID part as the first hierarchical pairing ID.
[0232] Clause 22: The method of Clause 21, wherein the second hierarchical pairing ID has a value of a second level ID part that is less than a value of the second level ID part of the first hierarchical pairing ID.
[0233] Clause 23: 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 1-22.
[0234] Clause 24: An apparatus, comprising means for performing a method in accordance with any combination of Clauses 1-22.
[0235] Clause 25: 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 1-22.
[0236] Clause 26: A computer program product embodied on a computer-readable storage medium comprising code for performing a method in accordance with any combination of Clauses 1-22.
[0237] Clause 27: A wireless node (e.g., a user equipment (UE) or UE side training server) , including: at least one transceiver; at least one memory including executable instructions; and at least one processor configured to execute the executable instructions and cause the wireless node to perform a method in accordance with any combination of Clauses 1-13, wherein the at least one transceiver is configured to receive the first signaling.
[0238] Clause 28: A wireless node (e.g., a network entity) , including: at least one transceiver; at least one memory including executable instructions; and at least one processor configured to execute the executable instructions and cause the wireless node to perform a method in accordance with any combination of Clauses 14-22, wherein the at least one transceiver is configured to transmit the first signaling. Additional Considerations
[0239] 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.
[0240] 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 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.
[0241] Means for obtaining, means for performing, means for refraining, means for training, means for using, means for generating, and means for outputting may comprise one or more processors, such as one or more of the processors described above with reference to FIG. 16.
[0242] 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) .
[0243] 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.
[0244] 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.
[0245] 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.An apparatus for wireless communication, comprising:at least one memory comprising instructions; andone or more processors configured to execute the instructions to cause the apparatus to:obtain first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at the apparatus and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; andperform one or more actions based on the association.2.The apparatus of claim 1, wherein the first ML model is usable at the apparatus to encode channel state feedback (CSF) .3.The apparatus of claim 1, wherein:the one or more actions comprise refraining, for a time period, from training the first ML model, if a third ML model associated with the second pairing has been trained by the apparatus.4.The apparatus of claim 1, whereinthe one or more actions comprise training the first ML model using the first parameters, if the apparatus lacks an ML model associated with the second pairing that has been trained by the apparatus.5.The apparatus of claim 4, wherein the first parameters comprise at least one of:shared reference encoder parameters or training data.6.The apparatus of claim 1, wherein:the first signaling comprises a configuration for inference; andthe one or more actions comprise using at least one of the first ML model or third ML model to perform inference.7.The apparatus of claim 6, wherein the one or more actions comprise using the first ML model or the third ML model based on availability.8.The apparatus of claim 6, wherein the first parameters comprise a first pairing identifier (ID) associated with the first pairing.9.The apparatus of claim 1, wherein:the first signaling further indicates a first pairing identifier (ID) associated with the first pairing; andthe association is indicated via a second pairing ID associated with the second pairing.10.The apparatus of claim 9, wherein the second pairing ID is associated with a root ID or source ID.11.The apparatus of claim 1, wherein the first signaling further indicates a first hierarchical pairing identifier (ID) associated with the first pairing, wherein the first hierarchical pairing identifier (ID) has at least a first level ID part and a second level ID part.12.The apparatus of claim 11, wherein:the association is indicated via the second level ID part of the first hierarchical pairing ID ifa second hierarchical pairing ID associated with the second pairing has a same value first level ID part as the first hierarchical pairing ID.13.The apparatus of claim 12, wherein the second hierarchical pairing ID has a value of a second level ID part that is less than a value of the second level ID part of the first hierarchical pairing ID.14.The apparatus of claim 1, further comprising at least one transceiver configured to receive the first signaling, wherein the apparatus is configured as a user equipment (UE) .15.An apparatus for wireless communication, comprising:at least one memory comprising instructions; andone or more processors configured to execute the instructions to cause the apparatus to:generate first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at a wireless node and a second ML model usable at a wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the apparatus or a fourth ML model usable at the wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; andoutput the first signaling.16.The apparatus of claim 14, wherein the first ML model is usable at the wireless node to decode channel state feedback (CSF) .17.The apparatus of claim 14, wherein the first parameters comprise at least one of:shared reference encoder parameters or training data.18.The apparatus of claim 14, wherein the first parameters comprise a first pairing identifier (ID) associated with the first pairing.19.The apparatus of claim 14, wherein:the first signaling further indicates a first pairing identifier (ID) associated with the first pairing; andthe association is indicated via a second pairing ID associated with the second pairing.20.The apparatus of claim 19, wherein the second pairing ID is associated with a root ID or source ID.21.The apparatus of claim 14, wherein the first signaling further indicates a first hierarchical pairing identifier (ID) associated with the first pairing, wherein the first hierarchical pairing identifier (ID) has at least a first level ID part and a second level ID part.22.The apparatus of claim 21, wherein:the association is indicated via the second level ID part of the first hierarchical pairing ID ifa second hierarchical pairing ID associated with the second pairing has a same value first level ID part as the first hierarchical pairing ID.23.The apparatus of claim 22, wherein the second hierarchical pairing ID has a value of a second level ID part that is less than a value of the second level ID part of the first hierarchical pairing ID.24.The apparatus of claim 15, further comprising at least one transceiver configured to transmit the first signaling, wherein the apparatus is configured as network entity.25.A method for wireless communication at a first wireless node, comprising:obtaining first signaling indicating 1) first parameters associated with a first pairing of a first machine learning (ML) model usable at the first wireless node and a second ML model usable at a second wireless node, and 2) an association between at least one of the first ML model or second ML model and at least one of a third ML model usable at the first wireless node or a fourth ML model usable at the second wireless node, wherein the third ML model and fourth ML model are associated with a second pairing; andperforming one or more actions based on the association.