Generalization and specification of ML models

Abstract

Certain aspects of the present disclosure provide techniques for determining suitability of a machine learning (ML) model to a cell in a wireless communication network. An example method performed at a user equipment (UE) for wireless communication includes, for a first machine learning (ML) model, storing at least a first type of generalization characteristics supported by the first ML model; and determining the suitability of the first ML model to the cell based on the generalization characteristics of the first type supported by the first ML model and the deployment characteristics of the cell indicated by the network entity of the cell.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to International Application No. PCT / CN2022 / 115317, filed on August 26, 2022, which is assigned to the assignee of this application and is hereby expressly incorporated by reference in its entirety, as if fully set forth below and for all available purposes. Technical Field

[0002] Aspects of the present disclosure relate to wireless communications and, more particularly, to techniques for determining the applicability of a machine learning (ML) model to a cell in a wireless communication network. Background Art

[0003] Wireless communication systems are widely deployed to provide various types of telecommunication content, such as telephony, video, data, messaging, broadcasts, or other similar types of services. These wireless communication systems may employ multiple-access technologies capable of supporting communications with multiple users by sharing available wireless communication system resources with these users.

[0004] Although wireless communication systems have achieved tremendous technical progress over the years, challenges still exist. For example, complex and dynamic environments can still attenuate or block signals between wireless transmitters and wireless receivers. Therefore, it is always desirable to improve the technical performance of wireless communication systems, including, for example: increasing the speed of communication and data carrying capacity, improving the efficiency of using shared communication media, reducing the power used by transmitters and receivers when performing communications, improving the reliability of wireless communication, avoiding redundant transmission and / or reception and related processing, improving the coverage area of ​​wireless communication, increasing the number and types of devices that can access wireless communication systems, increasing the ability of different types of devices to communicate with each other, increasing the number and types of wireless communication media available for use, etc. Therefore, there is a need to further improve wireless communication systems to overcome the above-mentioned technical challenges and other challenges. Summary of the invention

[0005] One aspect provides a method for wireless communication by a user equipment (UE). The method includes storing at least a first type of generalization characteristics supported by a first machine learning (ML) model, and determining applicability of the first ML model to a cell based on the first type of generalization characteristics supported by the first ML model and deployment characteristics of the cell.

[0006] In another aspect, a method for wireless communication by a network entity is provided. The method includes determining a deployment characteristic of a cell served by the network entity, and receiving an indication that at least a first machine learning (ML) model deployed at a user equipment (UE) is applicable to the cell based on at least a first type of generalization characteristic supported by the first ML model and the deployment characteristic.

[0007] Other aspects provide: an apparatus operable, configured, or otherwise adapted to perform any one or more of the foregoing methods and / or those described elsewhere herein; a non-transitory computer-readable medium comprising instructions, which, when executed by a processor of the apparatus, causes the apparatus to perform the foregoing methods and those described elsewhere herein; a computer program product implemented on a computer-readable storage medium, comprising code for performing the foregoing methods and those described elsewhere herein; and / or an apparatus comprising a unit for performing the foregoing methods and those described elsewhere herein. For example, an apparatus may include a processing system, a device having a processing system, or a processing system cooperating on one or more networks.

[0008] For purposes of illustration, the following description and drawings set forth certain features. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The accompanying drawings illustrate certain features of the various aspects described herein and should not be considered as limiting the scope of the disclosure.

[0010] Figure 1 An example wireless communication network is illustrated.

[0011] Figure 2 An example decomposed base station architecture is illustrated.

[0012] Figure 3 Aspects of an example base station and example user equipment are illustrated.

[0013] Figure 4A , 4B 4C and 4D illustrate various example aspects of data structures for a wireless communication network.

[0014] Figure 5 An example beam refinement process in accordance with certain aspects of the present disclosure is shown.

[0015] Figure 6 is a schematic diagram illustrating example operations by which beam management may be performed.

[0016] Figure 7 A general functional framework applied to RAN intelligence with AI capabilities is shown.

[0017] Figure 8 An example call flow diagram is shown in accordance with aspects of the present disclosure.

[0018] Fig. 9 A method for wireless communication is illustrated.

[0019] Fig.10 A method for wireless communication is illustrated.

[0020] Fig.11 A communication device according to aspects of the present disclosure is shown that may include various components configured to perform operations for the techniques disclosed herein. DETAILED DESCRIPTION

[0021] Aspects of the present disclosure provide apparatus, methods, processing systems, and computer-readable media for classifying a machine learning (ML) model used in a wireless communication network.

[0022] Performance Monitoring and Reporting Feedback (e.g., based on performance monitoring) is very important for the correct operation of artificial intelligence (AI) and machine learning (ML) based algorithms (herein referred to as ML models) deployed in wireless communication networks. Feedback can include the values ​​of common parameters known as key performance indicators (KPIs). Feedback can also include use case specific parameters that need to be monitored, evaluated and reported from time to time for appropriate actions.

[0023] Certain ML models may be suitable or applicable to certain cell deployment scenarios. For example, certain ML models may be classified as applicable across one or more cell deployment scenarios with common generalization characteristics. Various aspects of the present disclosure provide various aspects of ML model generalization indication for such classifications, which may improve the usability and efficacy of ML models in a greater number of cell deployment scenarios. Introduction to wireless communication networks

[0024] The techniques and methods described herein can be used in various wireless communication networks. Although terms commonly associated with 3G, 4G and / or 5G wireless technologies are used herein to describe various aspects, various aspects of the present disclosure may also be applicable to other communication systems and standards not explicitly mentioned herein.

[0025] Figure 1 An example of a wireless communication network 100 is illustrated in which various aspects described herein may be implemented.

[0026] In general, the wireless communication network 100 includes various network entities (alternatively, network elements or network nodes). A network entity is typically a communication device and / or a communication function performed by a communication device (e.g., user equipment (UE), base station (BS), component of a BS, server, etc.). For example, various functions of a network and various devices associated with and interacting with the network can be considered network entities. In addition, the wireless communication network 100 includes: terrestrial aspects, such as ground-based network entities (e.g., BS 102), and non-terrestrial aspects, such as satellites 140 and aircraft 145, which may include airborne network entities (e.g., one or more BSs) capable of communicating with other network elements (e.g., terrestrial BSs) and user equipment.

[0027] In the illustrated example, the wireless communication network 100 includes a BS 102, a UE 104, and one or more core networks, such as an evolved packet core (EPC) 160 and a 5G core (5GC) network 190, which interoperate to provide communication services over various communication links, including wired links and wireless links.

[0028] Figure 1 Various example UEs 104 are illustrated, which may more generally include: a cellular phone, a smart phone, a Session Initiation Protocol (SIP) phone, a laptop, a personal digital assistant (PDA), a satellite radio, a global positioning system, a multimedia device, a video device, a digital audio player, a camera, a game console, a tablet computer, a smart device, a wearable device, a vehicle, an electric meter, a gas pump, a large or small kitchen appliance, a healthcare device, an implant, a sensor / actuator, a display, an Internet of Things (IoT) device, an Always-On (AON) device, an edge processing device, or other similar devices. UE 104 may also be more generally referred to as a mobile device, a wireless device, a wireless communication device, a station, a mobile station, a user station, a mobile user station, a mobile unit, a user 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 the like.

[0029] BS 102 wirelessly communicates with UE 104 (e.g., sends signals to or receives signals from UE 104) via communication link 120. Communication link 120 between BS 102 and UE 104 may include uplink (UL) (also referred to as a reverse link) transmissions from UE 104 to BS 102 and / or downlink (DL) (also referred to as a forward link) transmissions from BS 102 to UE 104. Communication link 120 may use multiple-input multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming, and / or transmit diversity in various aspects.

[0030] BS 102 may generally include: 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 and reception point, etc. Each BS 102 may provide communication coverage for a corresponding geographic coverage area 110, which may sometimes be referred to as a cell and may overlap in some cases (e.g., a small cell 102' may have a coverage area 110' that overlaps with the coverage area 110 of a macro cell). For example, a BS may provide communication coverage for a macro cell (covering a relatively large geographic area), a pico cell (covering a relatively small geographic area, such as a stadium), a femto cell (relatively small geographic area (e.g., a home)), and / or other types of cells.

[0031] Although BS102 is illustrated as a single communication device in various aspects, BS102 can be implemented in various configurations. For example, one or more components of the base station can be decomposed, including a central unit (CU), one or more distributed units (DU), one or more radio units (RU), a near real-time (near RT) RAN intelligent controller (RIC) or a non-real-time (non-RT) RIC, just to name a few examples. In another example, various aspects of the base station can be virtualized. More generally, a base station (e.g., BS102) can include components located at a single physical location or components located at various physical locations. In an example where the base station includes components located at various physical locations, the various components can each perform a function so that the various components collectively implement functions similar to those of a base station located at a single physical location. In some aspects, a base station including components located at various physical locations can be referred to as a decomposed radio access network architecture (such as an open RAN (O-RAN) or virtualized RAN (VRAN) architecture). Figure 2 An exemplary decomposed base station architecture is illustrated and described.

[0032] Different BSs 102 within the wireless communication network 100 may also be configured to support different radio access technologies, such as 3G, 4G, and / or 5G. For example, a BS 102 configured for 4G LTE (collectively referred to as the Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN)) may interface with the EPC 160 via a first backhaul link 132 (e.g., an S1 interface). A BS 102 configured for 5G (e.g., 5G NR or Next Generation RAN (NG-RAN)) may interface with the 5GC 190 via a second backhaul link 184. The BSs 102 may communicate with each other directly or indirectly (e.g., via the EPC 160 or the 5GC 190) over a third backhaul link 134 (e.g., an X2 interface), which may be wired or wireless.

[0033] The wireless communication network 100 may subdivide the electromagnetic spectrum into various categories, frequency bands, channels, or other characteristics. In some aspects, this subdivision is provided based on wavelength and frequency, where frequency may also be referred to as a carrier, subcarrier, frequency channel, tone, or subband. For example, 3GPP currently defines Frequency Range 1 (FR1) as including 410 MHz–7125 MHz, which is often (interchangeably) referred to as “Sub-6 GHz”. Similarly, 3GPP currently defines Frequency Range 2 (FR2) as including 24,250 MHz–52,600 MHz, which is sometimes (interchangeably) referred to as “millimeter wave” (“mmW” or “mmWave”). A base station (e.g., a mmWave base station such as BS 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.

[0034] The communication link 120 between the BS 102 and, for example, the UE 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 may be aggregated in various aspects. The carriers may be adjacent to each other or may not be adjacent to each other. The allocation of carriers may be asymmetric for DL ​​and UL (e.g., more or fewer carriers may be allocated for DL ​​than for UL).

[0035] Communications using higher frequency bands may have higher path loss and shorter range than communications using lower frequencies. Figure 1180) can utilize beamforming 182 with UE 104 to improve path loss and range. For example, BS 180 and UE 104 can each include multiple antennas (such as, antenna elements, antenna panels, and / or antenna arrays) to facilitate beamforming. In some cases, BS 180 can send beamformed signals to UE 104 in one or more transmit directions 182'. UE 104 can receive beamformed signals from BS 180 in one or more receive directions 182". UE 104 can also send beamformed signals to BS 180 in one or more transmit directions 182". BS 180 may also receive beamformed signals from UE 104 in one or more receive directions 182″. BS 180 and UE 104 may then perform beam training to determine an optimal receive direction and an optimal transmit direction for each of BS 180 and UE 104. Notably, the transmit direction and the receive direction for BS 180 may be the same or may be different. Similarly, the transmit direction and the receive direction for UE 104 may be the same or may be different.

[0036] The wireless communication network 100 also includes a Wi-Fi AP 150 that communicates with Wi-Fi stations (STAs) 152 via communication links 154 in an unlicensed spectrum such as 2.4 GHz and / or 5 GHz.

[0037] Certain UEs 104 may communicate with each other using a device-to-device (D2D) communication link 158. The D2D communication 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).

[0038] The 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 illustrated example. The MME 162 may communicate with a home subscriber server (HSS) 174. The MME 162 is a control node that handles signaling between the UE 104 and the EPC 160. In general, the MME 162 provides bearer and connection management.

[0039] Typically, user Internet Protocol (IP) packets are forwarded through the Serving Gateway 166, which itself is connected to the PDN Gateway 172. The PDN Gateway 172 provides UE IP address allocation and other functions. The 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.

[0040] The BM-SC 170 may provide functionality for MBMS user service provisioning and delivery. The BM-SC 170 may serve as an entry point for content provider MBMS transmissions, 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. The MBMS Gateway 168 may be used to distribute MBMS services to 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.

[0041] 5GC 190 may include various functional components, including: access and mobility management function (AMF) 192, other AMFs 193, session management function (SMF) 194, and user plane function (UPF) 195. AMF 192 may communicate with unified data management (UDM) 196.

[0042] AMF 192 is a control node that handles signaling between UE 104 and 5GC 190. AMF 192 provides, for example, Quality of Service (QoS) flow and session management.

[0043] Internet Protocol (IP) packets are forwarded through UPF 195, which is connected to IP Services 197 and provides UE IP address allocation and other functions for 5GC 190. IP Services 197 may include, for example, the Internet, Intranet, IMS, PS streaming services, and / or other IP services.

[0044] In various aspects, a network entity or network node may be implemented as an aggregated base station, a decomposed 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.

[0045] Figure 2The architecture of an example decomposed base station 200 is illustrated. The decomposed 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 decomposed 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. The CU 210 may communicate with one or more distributed units (DUs) 230 via respective mid-haul links, such as an F1 interface. The DU 230 may communicate with one or more radio units (RUs) 240 via respective fronthaul links. The RU 240 may communicate with respective UEs 104 via one or more radio frequency (RF) access links. In some embodiments, a UE 104 may be served simultaneously by multiple RUs 240.

[0046] Each unit, such as CU 210, DU 230, RU 240, and near-RT RIC 225, non-RT RIC 215, and SMO framework 205, may include one or more interfaces, or be coupled to one or more interfaces, which are configured to receive or send signals, data, or information (collectively referred to as signals) via a wired or wireless transmission medium. Each unit, or an associated processor or controller that provides instructions to the communication interface of the unit, may be configured to communicate with one or more other units via a transmission medium. For example, a unit may include a wired interface, which is configured to receive or send signals from or to one or more other units via a wired transmission medium. Additionally or alternatively, a unit may include a wireless interface, which may include a receiver, a transmitter, or a transceiver (such as a radio frequency (RF) transceiver) that is configured to receive or send signals from or to one or more other units via a wireless transmission medium, or both.

[0047] In some aspects, CU 210 may host one or more higher-layer control functions. Such control functions may include: radio resource control (RRC), packet data convergence protocol (PDCP), service data adaptation protocol (SDAP), and the like. Each control function may be implemented with an interface configured to communicate signals with other control functions accommodated by CU 210. CU 210 may be configured to process user plane functions (e.g., central unit-user plane (CU-UP)), control plane functions (e.g., central unit-control plane (CU-CP)), or a combination thereof. In some embodiments, CU210 may be logically divided into one or more CU-UP units and one or more CU-CP units. When implemented in an O-RAN configuration, the CU-UP unit may communicate bidirectionally with the CU-CP unit via an interface such as an E1 interface. CU 210 may be implemented to communicate with DU 230 when necessary for network control and signaling.

[0048] DU 230 may correspond to a logical unit that includes one or more base station functions to control the operation of one or more RU 240. In some aspects, DU 230 may accommodate one or more of the following based at least in part on a functional partitioning, such as that defined by the Third Generation Partnership Project (3GPP): 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, etc.). In some aspects, DU 230 may further accommodate one or more low PHY layers. Each layer (or module) may be implemented with an interface that is configured to communicate signals with other layers (and modules) accommodated by DU 230 or with control functions accommodated by CU 210.

[0049] The lower layer functions may be implemented by one or more RUs 240. In some deployments, a RU 240 controlled by a DU 230 may correspond to a logical node that accommodates 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, etc.) or both based at least in part on functional partitioning, such as lower layer functional partitioning. In such an architecture, the RU 240 may be implemented to handle over-the-air (OTA) communications with one or more UEs 104. In some embodiments, real-time and non-real-time aspects of control and user plane communications with the RU 240 may be controlled by the corresponding DU 230. In some scenarios, this configuration may enable the DU 230 and the CU 210 to be implemented in a cloud-based RAN architecture (such as a vRAN architecture).

[0050] The SMO framework 205 may be configured to support RAN deployment and provisioning of non-virtualized network elements 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 operation 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 lifecycle management (such as for instantiating virtualized network elements) via a cloud computing platform interface (such as an O2 interface). Such virtualized network elements may include, but are not limited to, CU 210, DU 230, RU 240, and near-RT RIC 225. In some embodiments, the SMO framework 205 may communicate with hardware aspects of the 4G RAN (such as an open eNB (O-eNB) 211) via the O1 interface. In addition, in some embodiments, the SMO framework 205 may communicate directly with one or more RU240 via the O1 interface. The SMO framework 205 may also include a non-RT RIC 215 that is configured to support the functionality of the SMO framework 205 .

[0051] The non-RT RIC 215 may be configured to include logic functions that enable policy-based guidance of non-real-time control and optimization of RAN elements and resources, artificial intelligence / machine learning (AI / ML) workflows including model training and updating, or applications / features in the near-RT RIC 225. The non-RT RIC 215 may be coupled to or in communication with the near-RT RIC 225 (e.g., via an A1 interface). The near-RT RIC 225 may be configured to include logic functions that enable near real-time control and optimization of RAN elements and resources via data collection and actions on an interface connecting one or more CUs 210, one or more DUs 230, or both, and an O-eNB with the near-RT RIC 225 (e.g., via an E2 interface).

[0052] In some embodiments, in order to generate an AI / ML model to be deployed in the near-RT RIC 225, the non-RT RIC 215 may receive parameters or external enrichment information from an external server. Such information may be used by the near-RT RIC 225 and may be received from a non-network data source or from a network function at the SMO framework 205 or the non-RT RIC 215. In some examples, the non-RT RIC 215 or the near-RT RIC 225 may be configured to adjust RAN behavior or performance. For example, the non-RT RIC 215 may monitor long-term trends and patterns of performance and perform corrective actions using the AI / ML model through the SMO framework 205 (such as via reconfiguration of O1) or via the creation of a RAN management policy (such as an A1 policy).

[0053] Figure 3 Aspects of an example BS 102 and UE 104 are illustrated.

[0054] In general, BS 102 includes various processors (e.g., 320, 330, 338, and 340), antennas 334a-t (collectively 334), transceivers 332a-t (collectively 332) including modulators and demodulators, and other aspects that enable wireless transmission of data (e.g., data source 312) and wireless reception of data (e.g., data sink 339). For example, BS 102 can send and receive data between BS 102 and UE 104. BS 102 includes a controller / processor 340 that can be configured to implement various functions related to wireless communications as described herein.

[0055] In general, the UE 104 includes various processors (e.g., 358, 364, 366, and 380), antennas 352a-r (collectively 352), transceivers 354a-r (collectively 354) including modulators and demodulators, and other aspects that enable wireless transmission of data (e.g., retrieved from a data source 362) and wireless reception of data (e.g., provided to a data sink 360). The UE 104 includes a controller / processor 380 that can be configured to implement various functions related to wireless communications as described herein.

[0056] Regarding example downlink transmissions, BS 102 includes a transmit processor 320 that can receive data from a data source 312 and control information from a controller / processor 340. The control information can be for a physical broadcast channel (PBCH), a physical control format indicator channel (PCFICH), a physical HARQ indicator channel (PHICH), a physical downlink control channel (PDCCH), a group common PDCCH (GC PDCCH), etc. In some examples, the data can be for a physical downlink shared channel (PDSCH).

[0057] The transmit processor 320 may process (e.g., encode and symbol map) data and control information to obtain data symbols and control symbols, respectively. The transmit processor 320 may also generate reference symbols, e.g., for a primary synchronization signal (PSS), a secondary synchronization signal (SSS), a PBCH demodulation reference signal (DMRS), and a channel state information reference signal (CSI-RS).

[0058] The transmit (TX) multiple-input multiple-output (MIMO) processor 330 may perform spatial processing (e.g., precoding) on ​​the data symbols, control symbols, and / or reference symbols, if applicable, and may provide an output symbol stream to a modulator (MOD) in a transceiver 332a-332t. Each modulator in a transceiver 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 up-convert) the output sample stream to obtain a downlink signal. The downlink signals from the modulators in the transceivers 332a-332t may be transmitted via antennas 334a-334t, respectively.

[0059] To receive downlink transmissions, UE 104 includes antennas 352a-352r that can receive downlink signals from BS 102 and can provide received signals to demodulators (DEMODs) in transceivers 354a-354r, respectively. Each demodulator in transceivers 354a-354r can condition (e.g., filter, amplify, downconvert, and digitize) a corresponding received signal to obtain input samples. Each demodulator can further process the input samples to obtain received symbols.

[0060] An RX MIMO detector 356 may obtain received symbols from all demodulators in the transceivers 354a-354r, perform MIMO detection on the received symbols if applicable, and provide detected symbols. A 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.

[0061] With respect to example uplink transmissions, the UE 104 also includes a transmit processor 364 that can receive and process data from a data source 362 (e.g., for a PUSCH) and control information from a controller / processor 380 (e.g., for a physical uplink control channel (PUCCH)). The transmit processor 364 can also generate reference symbols for reference signals (e.g., for a sounding reference signal (SRS)). The symbols from the transmit processor 364 can be precoded by a TX MIMO processor 366 (if applicable), further processed by a modulator in the transceiver 354a-354r (e.g., for SC-FDM), and transmitted to the BS 102.

[0062] At BS 102, uplink signals from UE 104 may be received by antennas 334a-t, processed by demodulators in transceivers 332a-332t, detected by RX MIMO detector 336 (if applicable), and further processed by receive processor 338 to obtain decoded data and control information sent by UE 104. Receive processor 338 may provide decoded data to data sink 339 and decoded control information to controller / processor 340.

[0063] Memories 342 and 382 may store data and program codes for BS 102 and UE 104, respectively.

[0064] A scheduler 344 may schedule UEs for data transmission on the downlink and / or uplink.

[0065] In various aspects, the 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 for outputting data, such as outputting data from a data source 312, a scheduler 344, a memory 342, a transmit processor 320, a controller / processor 340, a TX MIMO processor 330, a transceiver 332a-t, an antenna 334a-t, and / or other aspects described herein. Similarly, "receiving" may refer to various mechanisms for obtaining data, such as obtaining data from an antenna 334a-t, a transceiver 332a-t, an RX MIMO detector 336, a controller / processor 340, a receive processor 338, a scheduler 344, a memory 342, and / or other aspects described herein.

[0066] In various aspects, the UE 104 may also 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 for outputting data, such as outputting data from a data source 362, a memory 382, ​​a transmit processor 364, a controller / processor 380, a TX MIMO processor 366, a transceiver 354a-r, an antenna 352a-r, and / or other aspects described herein. Similarly, "receiving" may refer to various mechanisms for obtaining data, such as obtaining data from an antenna 352a-r, a transceiver 354a-r, a RX MIMO detector 356, a controller / processor 380, a receive processor 358, a memory 382, ​​and / or other aspects described herein.

[0067] In some aspects, the processor may be configured to perform various operations, such as those associated with the methods described herein, and to send (output) data to or receive (obtain) data from another interface that is configured to send or receive data, respectively.

[0068] Figure 4A , 4B , 4C and 4D illustrate a wireless communication network (such as Figure 1 Various aspects of the data structure of the wireless communication network 100).

[0069] Specifically, Figure 4A is a diagram 400 illustrating an example of a first subframe within a 5G (e.g., 5G NR) frame structure, Figure 4B is a schematic diagram 430 showing an example of DL channels within a 5G subframe, Figure 4C is a schematic diagram 450 showing an example of a second subframe within a 5G frame structure, and Figure 4D is a schematic diagram 480 showing an example of UL channels within a 5G subframe.

[0070] Wireless communication systems may use orthogonal frequency division multiplexing (OFDM) with a cyclic prefix (CP) on both the uplink and downlink. Such systems may also support half-duplex operation using time division duplex (TDD). OFDM and single carrier frequency division multiplexing (SC-FDM) can be used to maximize the system bandwidth (e.g., Figure 4B and 4D ) is divided into multiple orthogonal subcarriers. Each subcarrier can be modulated with data. Modulation symbols can be sent in the frequency domain using OFDM and / or in the time domain using SC-FDM.

[0071] The wireless communication frame structure may be frequency division duplex (FDD), where, for a particular set of subcarriers, subframes within the subcarrier set are dedicated to one of DL or UL. The wireless communication frame structure may also be time division duplex (TDD), where, for a particular set of subcarriers, subframes within the subcarrier set are dedicated to both DL and UL.

[0072] exist Figure 4A and 4C In the wireless communication frame structure, TDD is used, where D is DL, U is UL, and X is flexible for use between DL / UL. The UE can be configured with a time slot format through a received time slot format indicator (SFI) (dynamically received through DL control information (DCI), or semi-statically / statically received through radio resource control (RRC) signaling). In the illustrated example, the 10ms frame is divided into 10 1ms subframes of equal size. Each subframe may include one or more time slots. In some examples, each time slot may include 7 or 14 symbols, depending on the time slot format. A subframe may also include a small slot, which typically has fewer symbols than a whole time slot. Other wireless communication technologies may have different frame structures and / or different channels.

[0073] In certain aspects, the number of slots within a subframe is based on the slot configuration and the numerology. For example, for slot configuration 0, different numerologies (μ) 0 to 5 allow 1, 2, 4, 8, 16, and 32 slots per subframe, respectively. For slot configuration 1, different numerologies 0 to 2 allow 2, 4, and 8 slots per subframe, respectively. Thus, 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 2. μ × 15kHz, where μ is the digital scheme 0 to 5. Thus, digital scheme μ=0 has a subcarrier spacing of 15kHz, and digital scheme μ=5 has a subcarrier spacing of 480kHz. The symbol length / duration is inversely related to the subcarrier spacing. Figure 4A , 4B4C and 4D provide examples of slot configuration 0 with 14 symbols per slot and a digital scheme 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.

[0074] like Figure 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 a physical RB (PRB)), which extends, for example, 12 consecutive subcarriers. The resource grid is divided into a plurality of resource elements (REs). The number of bits carried by each RE depends on the modulation scheme.

[0075] like Figure 4A As shown, some REs carry data for UEs (e.g. Figure 1 and 3 The RS may include a demodulation RS (DMRS) and / or a channel state information reference signal (CSI-RS) for channel estimation at the UE. The RS may also include a beam measurement RS (BRS), a beam refinement RS (BRRS), and / or a phase tracking RS (PT-RS).

[0076] Figure 4B Examples of various DL channels within a subframe of a frame are shown.The Physical Downlink Control Channel (PDCCH) carries DCI within one or more Control Channel Elements (CCEs), each CCE comprising, for example, nine RE Groups (REGs), each REG comprising, for example, four consecutive REs in an OFDM symbol.

[0077] The primary synchronization signal (PSS) may be in symbol 2 of a particular subframe in a frame. The PSS is transmitted by a UE (e.g., Figure 1 and 3 104) is used to determine subframe / symbol timing and physical layer identification.

[0078] A Secondary Synchronization Signal (SSS) may be located within symbol 4 of a specific subframe of a frame. The SSS is used by the UE to determine the physical layer cell identity group number and radio frame timing.

[0079] Based on the physical layer identifier and the physical layer cell identifier group number, the UE can determine the physical cell identifier (PCI). Based on the PCI, the UE can determine the location of the above-mentioned DMRS. The physical broadcast channel (PBCH) carrying the master information block (MIB) can be logically grouped with the PSS and SSS to form a synchronization signal (SS) / PBCH block. The MIB provides the number of RBs in the system bandwidth and the system frame number (SFN). The physical downlink shared channel (PDSCH) carries user data, broadcast system information (such as system information blocks (SIBs)) not sent via the PBCH, and / or paging messages.

[0080] like Figure 4C As shown in , some of the REs carry DMRS (indicated as R for a specific configuration, but other DMRS configurations are also possible) for channel estimation at the base station. The UE can send DMRS for PUCCH and DMRS for PUSCH. The PUSCH DMRS can be sent, for example, in the first or first two symbols of the PUSCH. Depending on whether a short PUCCH or a long PUCCH is sent and depending on the specific PUCCH format used, the PUCCH DMRS can be sent in different configurations. UE 104 can send a sounding reference signal (SRS). For example, the SRS can be sent in the last symbol of the subframe. The SRS can have a comb structure, and the UE can send the SRS on one of the combs. The SRS can be used by the base station for channel quality estimation to achieve frequency-dependent scheduling of the UL.

[0081] Figure 4D Examples of various UL channels within a subframe of a frame are shown. The PUCCH may be located as indicated in a configuration. The PUCCH carries uplink control information (UCI), such as scheduling requests, channel quality indicator (CQI), precoding matrix indicator (PMI), 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. Example beam refinement process

[0082] In mmWave systems, beamforming may be important to overcome high path loss. As described herein, beamforming may refer to establishing a link between a BS and a UE, where the two devices form beams corresponding to each other. Both the BS and the UE find at least one appropriate beam to form a communication link. The BS beam and the UE beam form a so-called beam pair link (BPL). As an example, on the DL, the BS may use a transmit beam, and the UE may receive a transmission using a receive beam corresponding to the transmit beam. The combination of a transmit beam and a corresponding receive beam may be a BPL.

[0083] As part of beam management, the beams used by the BS and UE must be refined from time to time due to changes in channel conditions, for example, due to movement of the UE or other objects. In addition, the performance of the BPL may be affected by fading due to Doppler spread. Since the channel state changes over time, the BPL should be updated or refined periodically. Therefore, it may be beneficial if the BS and UE monitor the beams and the new BPL.

[0084] At least one BPL must be established for network access. As mentioned above, new BPLs may need to be discovered later for different purposes. The network may decide to use a different BPL for a different channel, or for communicating with a different BS (e.g., a Transmit Receiving Point (TRP)), or as a backup BPL in the event of a failure of an existing BPL.

[0085] The UE usually monitors the quality of the BPL, and the network may refine the BPL from time to time.

[0086] Figure 5 An example 500 for BPL discovery and refinement is shown. In 5G-NR, P1, P2, and P3 procedures are used for BPL discovery and refinement. The network uses the P1 procedure to enable discovery of new BPLs. In the P1 procedure, Figure 5 As shown, the BS transmits different symbols of the reference signal, and each beam is formed in a different spatial direction so as to reach several (eg, most or all) relevant locations of the cell. In other words, the BS uses different transmit beams to transmit beams in different directions over time.

[0087] In order to successfully receive at least one symbol of this "P1-signal", the UE must find a suitable receive beam. It searches using the available receive beams and applies a different UE beam during each occurrence of the periodic P1 signal.

[0088] Once the UE successfully receives a symbol of the P1 signal, it has discovered a BPL. The UE may not want to wait until it finds the best UE receive beam, as this may delay further action. The UE can measure the Reference Signal Received Power (RSRP) and report the symbol index together with the RSRP to the BS. Such a report usually contains the discovery of one or more BPLs.

[0089] In one example, the UE may determine a received signal with a high RSRP. The UE may not know which beam the BS is using to transmit; however, the UE may report to the BS the time when it observed a signal with a high RSRP. The BS may receive the report and may determine which BS beam the BS is using at that given time.

[0090] The BS may then provide P2 and P3 procedures to refine a single BPL. The P2 procedure refines the BS beam of the BPL. For example, the BS may transmit several symbols of the reference signal using a different BS beam that is spatially close to the BS beam of the BPL (the BS performs the sweep using neighboring beams around the selected beam). In P2, the UE keeps its beam constant. Therefore, although the UE uses the same beam as in the BPL (e.g. Figure 5 ), but the BS beams for P2 differ from those for P1 in that they may be spaced closer together or more focused. The UE may measure the RSRP for various BS beams and indicate the best one to the BS.

[0091] The P3 process refines the UE beam for BPL (see Figure 5 The UE uses different receive beams for scanning while the BS beam remains constant (UE sweeps using adjacent beams). The UE can measure the RSRP of each beam and identify the best UE beam. Afterwards, the UE can use this best UE beam for BPL and report the RSRP to the BS.

[0092] Over time, the BS and UE establish several BPLs. When the BS sends a specific channel or signal, it lets the UE know which BPL will be involved so that the UE can tune to the direction of the correct UE receive beam before the start of the signal. In this way, the UE can use the correct receive beam to receive each sample of the signal or channel. In one example, the BS can indicate which BPL is involved for a scheduled signal (e.g., SRS, CSI-RS) or channel (e.g., PDSCH, PDCCH, PUSCH, PUCCH). In NR, this information can be called a quasi-co-site (QCL) indication.

[0093] If the properties of the channel over which the symbols on one antenna port are transmitted can be inferred from the channel over which the symbols on the other antenna port are transmitted, then the two antenna ports are quasi co-located (QCL). QCL supports at least beam management functions, frequency / timing offset estimation functions, and radio resource management (RRM) functions.

[0094] The BS may use the BPL that the UE has received, established and / or used in the past. The transmit beam for the signal to be sent and the previously received signal both point in the same direction and / or are QCL. The UE may require a QCL indication (before the signal to be received) so that the UE can use the correct receive beam for each signal or channel. When the BPL of the signal or channel changes, some QCL indications may be needed from time to time, and some QCL indications are needed for each scheduled instance. The QCL indication may be sent in the downlink control information (DCI), which may be part of the PDCCH channel. Because DCI is needed to control this information, it may be desirable that the number of bits required to indicate the QCL is not too large. The QCL may be sent in a medium access control-control unit (MAC-CE) or a radio resource control (RRC) message.

[0095] According to an example, whenever a UE reports a BS beam that it has received with sufficient RSRP, and the BS decides to use that BPL in the future, the BS assigns it a BPL tag. Thus, two BPLs with different BS beams may be associated with different BPL tags. BPLs based on the same BS beam may be associated with the same BPL tag. Thus, according to this example, the tag is a function of the BS beam of the BPL.

[0096] As mentioned above, wireless systems, such as millimeter wave (mmW) systems, have brought gigabit speeds to cellular networks due to the availability of large amounts of bandwidth. However, the unique challenges of severe path loss faced by such wireless systems require new technologies such as hybrid beamforming (analog and digital) that did not exist in 3G and 4G systems. Hybrid beamforming can enhance the link budget / signal-to-noise ratio (SNR) that can be utilized during the RACH.

[0097] In such a system, a Node B (NB) and a User Equipment (UE) can communicate via active beamforming transmit beams. An active beam can be considered as a pair of transmit (Tx) beams and receive (Rx) beams between the NB and the UE, which carry data and control channels such as PDSCH, PDCCH, PUSCH, and PUCCH. As described above, the transmit beam used by the NB for downlink transmissions and the corresponding receive beam used by the UE can be referred to as a beam pair link (BPL). Similarly, the transmit beam used by the UE for uplink transmissions and the corresponding receive beam used by the NB can also be referred to as a BPL.

[0098] Since the direction of the reference signal is unknown to the UE, the UE can evaluate several beams to obtain the best Rx beam for a given NB Tx beam. However, if the UE has to "sweep through" all of its Rx beams to perform measurements (e.g., to determine the best Rx beam for a given NB Tx beam), the UE may incur significant delays in the measurements and battery life impacts. In addition, having to sweep through all Rx beams is highly resource inefficient. Therefore, aspects of the present disclosure provide techniques for assisting UEs in performing measurements of serving cells and neighboring cells when using Rx beamforming. Example Beam Management

[0099] In wireless communications, various procedures may be performed for beam management. Figure 6 600 is a diagram illustrating example operations in which beam management may be performed. In initial access 602, the network may sweep several beams, for example via synchronization signal blocks (SSBs), as described herein for Figure 4B Further described. The network may configure a random access channel (RACH) resource associated with a beamformed SSB to the UE to facilitate initial access via the RACH resource. In certain aspects, the SSB may have a wider beam shape compared to other reference signals such as a channel state information reference signal (CSI-RS). The UE may use SSB detection to identify a RACH opportunity (RO) for sending a RACH preamble (e.g., as part of a contention-based random access (CBRA) procedure).

[0100] In connected mode 604, the network and the UE may perform hierarchical beam refinement, which includes beam selection (e.g., a process referred to as P1), beam refinement of the transmitter (e.g., a process referred to as P2), and beam refinement of the receiver (e.g., a process referred to as P3). In beam selection (P1), for example, the network may sweep the beams and the UE may report the beam with the best channel characteristics. In beam refinement of the transmitter (P2), the network may sweep the narrower beams and the UE may report the beam with the best channel characteristics among the narrow beams. In beam refinement of the receiver (P3), the network may repeatedly use the same beam for transmission, and the UE may refine the spatial reception parameters (e.g., spatial filters) used to receive signals from the network via the beam. In certain aspects, the network and the UE may perform complementary processes (e.g., U1, U2, and U3) for uplink beam management.

[0101] In certain situations where a beam failure occurs (e.g., due to beam misalignment and / or blocking), the UE may perform a beam failure recovery (BFR) procedure 606, which may allow the UE to return to connected mode 604 without performing a radio link failure procedure 608. For example, the UE may be configured with candidate beams for beam failure recovery. In response to detecting a beam failure, the UE may request the network to perform beam failure recovery via one of the candidate beams (e.g., one of the candidate beams having a reference signal received power (RSRP) above a certain threshold). In certain situations where a radio link failure (RLF) occurs, the UE may perform an RLF procedure 608 (e.g., a RACH procedure) to recover from the radio link failure. Example framework for AI / ML in radio access networks

[0102] Figure 7 An example of an AI / ML functional framework 700 for RAN intelligence is illustrated in which various aspects described herein may be implemented.

[0103] The AI / ML functional framework includes: data collection function 702, model training function 704, model inference function 706 and actor function 708, which interoperate to provide a platform for collaboratively applying AI / ML to various processes in the RAN.

[0104] The data collection function 702 typically provides input data to the model training function 704 and the model inference function 706. AI / ML algorithm-specific data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) may not be performed in the data collection function 702.

[0105] Examples of input data to the data collection function 702 (or other functions) may include: measurements from UEs or different network entities, feedback from actor functions, and outputs from AI / ML models. In some cases, analysis of data required at the model training function 704 and the model inference function 706 may be performed at the data collection function 702. As shown, the data collection function 702 may deliver training data to the model training function 704 and deliver inference data to the model inference function 706.

[0106] The model training function 704 can perform AI / ML model training, validation, and testing, which can generate model performance metrics as part of the model testing process. If necessary, the model training function 704 can also be responsible for data preparation (e.g., data preprocessing and cleaning, formatting, and conversion) based on the training data delivered by the data collection function 702.

[0107] The model training function 704 may provide model deployment / update data to the model inference function 706. The model deployment / update data may be used to initially deploy a trained, validated, and tested AI / ML model to the model inference function 706, or to deliver an updated model to the model inference function 706.

[0108] As shown, the model inference function 706 can provide AI / ML model inference output (e.g., prediction or decision) to the actor function 708, and sometimes also provide model performance feedback to the model training function 704. The model inference function 706 can also sometimes be responsible for data preparation (e.g., data preprocessing and cleaning, formatting and transformation) based on the inference data delivered by the data collection function 702.

[0109] Inference output of the AI / ML model may be produced by the model inference function 706. The specific details of this output may be specific in terms of the use case. Model performance feedback may sometimes be used to monitor the performance of the AI / ML model. In some cases, for example, if certain information derived from the model inference function 706 is suitable for improving the AI / ML model trained in the model training function 704, the model performance feedback may be delivered to the model training function 704.

[0110] The model inference function 706 may signal the outputs of the model to nodes that have requested them (e.g., via subscriptions), or to nodes that take actions based on the outputs of the model inference function 708. The AI / ML models used in the model inference function 706 may require initial training, validation, and testing by the model training function 704 prior to deployment. The model training function 704 and the model inference function 706 may be able to request specific information for training or executing the AI / ML algorithm and avoid receiving unnecessary information. The nature of such information may depend on the use case and the AI / ML algorithm.

[0111] The actor function 708 may receive output from the model inference function 706, which may trigger or perform corresponding actions. The actor function 708 may trigger actions for other entities or for itself. The feedback generated by the actor function 708 may provide information for deriving training data, inferring data, or for monitoring the performance of the AI / ML model. As described above, the input data for the data collection function 702 may include this feedback from the actor function 708. Feedback from the actor function 708 or other network entities (e.g., via the data collection function 702) may also be used at the model inference function 706.

[0112] The AI / ML functional framework 700 can be deployed in various RAN intelligence-based use cases, which may include CSI feedback enhancement, enhanced beam management (BM), positioning and location determination (Pos-Loc) accuracy enhancement, and various other use cases. Aspects related to ML model generalization and regularization

[0113] Aspects of the present disclosure provide apparatus, methods, processing systems, and computer-readable media for classifying a machine learning (ML) model used in a wireless communication network.

[0114] As used herein, a generalized characteristic generally refers to a characteristic or parameter that can be shared between one or more cells or network configurations. For example, a generalized characteristic can be classified based on base station location information, cell range, propagation characteristics, antenna configuration, or one or more parameters. Certain generalized characteristics, such as delay spread range, path loss range, UE speed range, multiple sectors in a cell, or an area map in a cell, may be applicable to a specific cell (e.g., indicated by a global cell ID), a cell group (e.g., indicated by a cell ID list, a RAN notification area ID, a paging area ID, or a model cell group ID), and / or to all cells or a subset of cells (e.g., having the same network ID). As used herein, the applicability of a generalized characteristic may, for example, refer to the compatibility, availability, relevance, or appropriateness of the generalized characteristic relative to a cell, a cell group, a cell subset, and / or a cell-specific model (e.g., a model specific to a cell or a cell subset).

[0115] In some cases, the applicability may be determined based on the category / group / type of the cell (e.g., a dense urban type or an indoor factory type) and / or deployment characteristics (e.g., similarities in deployment characteristics between different cells). Deployment characteristics may refer to: the location / position of the deployment, characteristics of the deployment location (e.g., congestion conditions, traffic conditions, weather conditions, and / or other characteristics). For example, an ML model suitable for a cell deployment corresponding to a factory may not be suitable for a cell deployment corresponding to an urban area. Because a factory deployment may represent a relatively controlled environment, the input to the ML model may be optimized for relatively low variability (e.g., a narrow range). On the other hand, an urban area may represent an environment that is difficult to control. Therefore, a different ML model with an input optimized for higher variability (and a larger range) may be more suitable for cell deployments in urban areas.

[0116] As used herein, a machine learning (ML) model generally refers to a computing system that can learn from data and can make predictions or decisions. For example, an ML model can refer to a mathematical algorithm that is trained on a data set to make predictions or decisions based on input data. An ML model can be considered to be a type of artificial intelligence (AI) that enables a computer to learn and improve its performance over time without being explicitly programmed to do so. An ML model can learn from the data provided to it and identify patterns or relationships in the data that can be used to make predictions for new data. ML models can be used in a wide range of applications, including image and speech recognition, natural language processing, and predictive analysis. There are several types of machine learning models, including supervised learning, unsupervised learning, and reinforcement learning.

[0117] Certain ML models may be suitable or applicable to certain cell deployment scenarios. For example, certain ML models may be classified as applicable across one or more cell deployment scenarios with common generalization characteristics. Various aspects of the present disclosure provide various aspects of such classified ML model generalization indications.

[0118] The classification of ML models may be beneficial for developing new 5G systems, for example, to operate in various frequency bands (e.g., up to 100 GHz). The classification can help achieve accurate radio propagation models. Channel models can include typical deployment scenarios for urban microcells (UMi) and urban macrocells (UMa). Certain (e.g., baseline) models can be used to incorporate path loss, shadow fading, line of sight probability, penetration, and blocking models for typical scenarios.

[0119] In some cases, certain generalization types may correspond to categories of ML models (as described above). The following generalization types may be considered (eg, considered to have common generalization cell deployment characteristics).

[0120] The first type (Type 1) may correspond to a heterogeneous inter-site cell deployment (e.g., with UMi and Uma). For this first type, the AI / ML model may be expected to perform on a (previously) unseen deployment type (e.g., the model may be trained on a dense city and tested on UMi).

[0121] The second type (Type 2) may correspond to homogeneous inter-site cell deployments, and the AI / ML model may be expected to perform on unseen sites of the same deployment type (e.g., trained on dense cities and tested on new placements / locations / scenarios of the dense city type). For example, a model classified by this type may be trained in the city center of one city and deployed in the city center of another city. On the other hand, a model that performs well in some city center areas may not perform well in rural areas.

[0122] The third type (Type 3) may correspond to intra-site cell deployments, and the AI / ML model may be expected to execute on unseen variables within the same site (e.g., unseen UE positions, trajectories within velocity and whereabouts, changes in moving objects in the environment).

[0123] The fourth type (Type 4) may correspond to a cross configuration, with the performance of the AI / ML model in the various configuration types described above (i.e., unseen beam configurations).

[0124] The classification techniques proposed in this paper can generally be applied to any type of ML model for various purposes (e.g., for channel state feedback, beam prediction, or other purposes).

[0125] You can refer to Figure 8 800 to understand the general concepts for classifying an ML model and determining its suitability.

[0126] As shown, for one or more machine learning (ML) models, the UE may store a type of generalized features supported by each ML model. In some cases, the network entity may configure the UE using the ML model.

[0127] In some cases, a network entity (e.g., a gNB) may indicate (e.g., via broadcast signaling / SIB) deployment characteristics of a cell (e.g., a cell served by the network entity). The UE may determine the applicability of the ML model to the cell based on the generalization characteristics of the corresponding type supported by the ML model and the cell deployment characteristics.

[0128] In some cases, the network (or standard) may specify a standardized description of model generalization properties, e.g., for the various types described above (types 1-4). While this classification may be specific to generating simulated data from a channel model, a similar approach may be applicable to other (e.g., real-world) models.

[0129] According to certain aspects, for each type used for generalization, there can be several subtypes. If necessary, subtypes can provide the ability to indicate fine-grained capabilities (e.g., properties of certain types or values ​​of certain parameters).

[0130] For example, for the first type (Type 1: heterogeneous sites) described above, there may be various subtypes. For example, these subtypes may include: Rural macrotype 1, 2, ..., N1, urban macrotype 1, 2, ..., N2, urban microtype 1, 2, ..., N3... In some cases, the categories defining each subtype may be defined (e.g., in a standard specification) based on characteristics such as: gNB location information, range of cells, propagation characteristics (e.g., delay spread, angular spread of channels, etc.), antenna configuration, or some combination of parameters determined by explicit indication from the network. For example, an urban macro subtype and an urban micro subtype may be defined based on similar gNB location information (associated with an urban area), but different sizes of cell coverage (e.g., an urban macro type is associated with a larger coverage area than an urban micro type).

[0131] In some cases, the UE may indicate (e.g., to the network) that a given ML model may be applicable to any cell belonging to a certain group or category (e.g., which in some cases may correspond to a generalized type), such as: Rural macro types 4 and 5 (and urban macro types 1, 2, 3, etc.); Indoor Hotspot Type 1 and Indoor Factory Type 2; or Indoor plants Type 2 and Type 3.

[0132] Certain types (e.g. Type 2: Homogeneous inter-site) may apply to cell specific models, but also to groups of cells with similar deployment characteristics. For example, within the classification of UMi Type 2 cells, the ML model may be applicable (only) to a smaller sub-category of cells. According to a first option, such an ML model may be applicable to a specific cell indicated by a global cell ID. According to a second option, the ML model may be applicable to a smaller group of cells indicated by a cell ID or a RAN notification area ID or a paging area ID or a list of newly defined IDs for the ML model cell group. According to a third option, for a dedicated network, the ML model may be applicable to all cells (or a subset of cells) with the same network ID.

[0133] There are also various possible subtypes for type 3 classification where the AI / ML model can be expected to perform for unseen variables within the same location (e.g., unseen UE positions, trajectories within speed and whereabouts, changes in moving objects in the environment). Such subtypes for type 3 generalized characteristics can be applicable within a cell. Examples of such cell generalized characteristics applicable within a cell may include at least one of the following: a delay spread range, a path loss range, a UE speed range, multiple sectors in a cell, or an area map in a cell.

[0134] For certain types of classification (e.g., Type 4: Cross Configuration), the AI / ML model may be expected to perform in various configurations (e.g., using beam configurations that have not been seen before). Therefore, there are various subtypes for Type 4. For example, if the network indicates the RS configuration ID when collecting the training data set, the subtype may be based on the range of RS configuration IDs supported by the model. Another subtype may indicate whether the model supports RS configurations that have not been seen before.

[0135] There are various signaling options for indicating model generalization parameters. For example, an indication of model generalization parameters may be provided as the type (and subtype) of supported generalizations and may be stored with the model and may be registered with the network together with the model. For example, the UE may register the supported model generalizations by providing (e.g., via a UE capability report) a description of the model generalization or model generalization parameters and the associated ML model to the network. This registration may help the network determine what ML models the UE supports. In some cases, when the UE uses a proprietary AI / ML model, the UE may provide a generalized description (e.g., which may include generalized characteristics) to the network. In some cases, the network may indicate the generalized description to the UE (e.g., when the network configures the AI / ML model to the UE). Such generalization information may be indicated to the UE in broadcast signaling (e.g., via a system information block (SIB)) or via unicast signaling (e.g., via RRC, MAC-CE, DCI) or group (multicast) signaling for a group of UEs (e.g., via RRC, MAC-CE, or DCI). In some cases, the UE may acknowledge receipt of the generalized information to the network (eg, in a response to a message conveying the generalized information).

[0136] In some cases, multiple generalized classifications (e.g., generalized types) may be indicated for each model. For example, there may be a type 2 indication for a cell group, a type 3 indication for a specific parameter within a cell, and / or a type 4 RS configuration (e.g., RS configuration 1, 2, 3...).

[0137] In some cases, the UE may detect / determine an out-of-model-coverage-scenario. In this case, for example, when the input data is outside the configured generalization description of the configured AI / ML model, the UE may report "out-of-model-coverage" to the network. There are various options for how the UE reports "out-of-coverage" or the actions it performs when detecting and reporting "out-of-coverage".

[0138] For example, according to one option, the UE may request the network to deactivate the ML model (for the ML model configured by the network). In some cases, the UE may request the network to (re)activate the ML model. According to a second option, the UE may request a model switch by indicating the currently observed channel characteristics. According to a third option, the UE may continue to use the best matching model with reduced performance. According to a fourth option, the UE may enable a data collection function to further optimize the network. Example Operations of User Equipment

[0139] Fig. 9 A method for use by a UE such as, Figure 1 and Figure 3 An example of method 900 for performing wireless communications with a UE 104).

[0140] Method 900 begins at step 905 by storing at least a first type of generalization feature supported by a first machine learning (ML) model. In some cases, the operation of this step involves referring to Fig.11 The circuit for storing and / or the code for storing described herein may be referred to as Fig.11 The described circuits for storing and / or codes for storing are executed.

[0141] Method 900 then proceeds to step 910 to determine the applicability of the first ML model to the cell based on the generalization characteristics of the first type supported by the first ML model and the deployment characteristics of the cell. In some cases, the operation of this step involves referring to Fig.11 The circuit for determining and / or the code for determining described herein may be referred to as Fig.11 The described circuits for determining and / or codes for determining are performed.

[0142] In one aspect, method 900 or any aspect related thereto may be performed by an apparatus such as Fig.11 The communication device 1100 includes various components operable, configured or adapted to perform the method 900. The communication device 1100 will be described in more detail below.

[0143] Notice, Fig. 9 This is but one example of one method, and other methods including fewer, additional, or alternative steps are also possible consistent with the present disclosure. Example Operations of Network Entities

[0144] Fig.10 Components for use by a network entity (such as base station 102 or a decomposed base station, as described above for Figure 1 , 2 and 3) as described in an example of a method 1000 for conducting wireless communication.

[0145] Method 1000 starts at step 1005 by determining deployment characteristics of cells served by a network entity. Deployment characteristics may refer to: deployment location / position, characteristics of the deployment location (e.g., congestion conditions, traffic conditions, weather conditions, and / or other characteristics). For example, the determination may be based on network configuration of a network provider and / or information (information about deployment characteristics) obtained by the network entity (e.g., via a discovery process). In some cases, the network entity may send an indication about the deployment characteristics to the UE. In some cases, the operation of this step involves reference to Fig.11 The circuit for determining and / or the code for determining described herein may be referred to as Fig.11 The described circuits for determining and / or codes for determining are performed.

[0146] Method 1000 then proceeds to step 1010, receiving an indication that at least a first machine learning (ML) model deployed at the UE is applicable to the cell based on at least a first type of generalization characteristic supported by the first ML model and the deployment characteristic. In some cases, the operation of this step involves reference to Fig.11 The circuit for receiving and / or the code for receiving described herein may be referred to as Fig.11 The described circuits for receiving and / or codes for receiving are performed.

[0147] In one aspect, method 1000 or any aspect related thereto may be performed by an apparatus such as Fig.11 The communication device 1100 includes various components operable, configured or adapted to perform the method 1000. The communication device 1100 will be described in more detail below.

[0148] Notice, Fig.10 This is but one example of one method, and other methods including fewer, additional, or alternative steps are also possible consistent with the present disclosure. Example Wireless Communication Device

[0149] Fig.11 A communication device 1100 is shown, which may include devices configured to perform operations of the techniques disclosed herein (such as, for example, Fig. 9 and / or Fig.10 The operations shown in FIG. 4A and FIG. 4B may be used to illustrate various components (eg, corresponding to means-plus-function components) of the invention.

[0150] Fig.11 Various aspects of an example communication device 1100 are illustrated. In some aspects, the communication device 1100 is a user equipment (such as UE 104), or a network entity (such as base station 102, or a component of a decomposed base station, as described above with respect to Figure 1 , 2and 3).

[0151] The communication device 1100 includes a processing system 1105 coupled to a transceiver 1155 (e.g., a transmitter and / or a receiver). The transceiver 1155 is configured to transmit and receive signals for the communication device 1100, such as various signals as described herein, via an antenna 1160. The processing system 1105 may be configured to perform processing functions for the communication device 1100, including processing signals received by the communication device 1100 and / or signals to be transmitted by the communication device 1100.

[0152] The processing system 1105 may include one or more processors 1110. In various aspects, the one or more processors 1110 may represent a processor such as a processor 1110 for processing a program. Figure 3 One or more of the receive processor 358, transmit processor 364, TX MIMO processor 366, and / or controller / processor 380 described herein. The one or more processors 1110 are coupled to the computer-readable medium / memory 1130 via the bus 1150. In certain aspects, the computer-readable medium / memory 1130 is configured to store instructions (e.g., computer-executable code) that, when executed by the one or more processors 1110, cause the one or more processors 1100 to perform operations directed to the computer-readable medium / memory 1130. Fig. 9 and / or Fig.10 The method 800 described herein, or any aspect related thereto. Note that reference to a processor performing a function of the communication device 1100 may include one or more processors 1110 performing that function of the communication device 1100 .

[0153] In the illustrated example, the computer readable medium / memory 1130 stores code (e.g., executable instructions), such as code for storing 1135, code for determining 1140, code for sending 1145, and code for receiving 1146. The processing of the code may cause the communication device 1100 to perform operations for Fig. 9 and / or Fig.10 or any aspects thereof.

[0154] The one or more processors 1110 include circuits configured to implement (e.g., execute) code stored in the computer-readable medium / memory 1130, including circuits such as: circuits for storing 1115, circuits for determining 1120, circuits for sending 1125, and circuits for receiving 1126, which may be configured to cause the communication device 1100 to perform operations for Fig. 9 and / or Fig.10 or any aspects thereof.

[0155] The various components of the communication device 1100 may provide for performing Fig. 9 and / or Fig.10 The described method 800 or any aspect thereof may include means for storing, determining, sending and / or receiving. For example, the means for storing, determining, sending and / or receiving may include Figure 3 The transceiver 354 and / or antenna 352 of the UE 104 shown in FIG. The means for receiving or obtaining may include Figure 3 1. The transceiver 354 and / or antenna(s) 352 of the UE 104 are shown in FIG.

[0156] In some cases, Fig. 9 and / or Fig.10 The operations shown in, as well as other operations described herein for performing ML model generalization and specification, may be implemented by one or more unit-plus-functional components. For example, in some cases, such operations may be implemented by a unit for ML model generalization and specification. Sample Clauses

[0157] Examples of implementations are described in the following numbered clauses:

[0158] Clause 1: A method for wireless communications at a user equipment (UE), comprising: storing at least a first type of generalized characteristics supported by a first machine learning (ML) model; and determining applicability of the first ML model to a cell based on the first type of generalized characteristics supported by the first ML model and deployment characteristics of the cell.

[0159] Clause 2: The method of clause 1, further comprising: receiving an indication of the deployment characteristic from a network entity of the cell.

[0160] Clause 3: The method of any of clauses 1-2, wherein: the at least first type of generalization characteristics includes different types of generalization characteristics supported by different ML models including the first ML model. For example, the at least first type of generalization characteristics may include multiple types of generalization characteristics, wherein each type of generalization characteristic is supported by at least one ML model in a plurality of ML models, the plurality of ML models including the first ML model.

[0161] Clause 4: The method of clause 3, wherein at least one type of the different types of generalized properties (or multiple types of generalized properties) has one or more subtypes.

[0162] Clause 5: The method of clause 4, wherein: each of the one or more subtypes is defined by one or more classifications of generalized characteristics, and the one or more classifications are based on at least one of the following: base station location information, cell range, propagation characteristics, antenna configuration, or a combination of parameters indicated by a network entity.

[0163] Clause 6: The method of clause 4 or clause 5, further comprising: indicating that at least one of the different ML models is applicable to one or more cells deployed according to the one or more subtypes.

[0164] Clause 7: The method of any of clauses 3 to 6, wherein: different ML models are applied to cell-specific models or groups of cells with similar deployment characteristics.

[0165] Clause 8: The method of clause 7, wherein: at least one ML model type applies to at least one of: a specific cell indicated by a global cell ID; a group of cells indicated by a cell ID, a RAN notification area ID, a paging area ID, or a list of IDs for ML model cell groups; or all cells or a subset of cells having the same network ID.

[0166] Clause 9: The method of any one of clauses 3 to 8, wherein: the different types of generalized characteristics include at least one of the following: a delay spread range, a path loss range, a UE speed range, a plurality of sectors in a cell, or an area map in a cell.

[0167] Clause 10: The method of any of clauses 3 to 9, wherein the different ML models support a range of reference signal (RS) configuration IDs.

[0168] Clause 11: The method of any one of clauses 1-10, further comprising: providing a generalized description of the first ML model to a network entity.

[0169] Clause 12: The method of any of clauses 1-11, further comprising: receiving an indication of the first type of generalization characteristics supported by the first ML model from a network entity.

[0170] Clause 13: The method of clause 12, wherein the indication of the first type of generalized properties supported by the first ML model is received via at least one of broadcast signaling or multicast signaling.

[0171] Clause 14: The method of clause 12, further comprising: sending signaling to confirm receipt of the indication of the first type of generalized features supported by the first ML model.

[0172] Clause 15: The method of any of clauses 1-14, wherein the UE stores at least a second type of generalized features supported by the first ML model.

[0173] Clause 16: The method of any of clauses 1-15, further comprising: reporting model coverage out to a network entity when data to be input to the first ML model is outside the first type of generalization characteristics supported by the first ML model; and performing one or more actions in response to detecting the model coverage out.

[0174] Clause 17: The method of clause 16, wherein the one or more actions include at least one of: requesting the network entity to deactivate the first ML model; requesting a switch from the first ML model to a second ML model by indicating currently observed channel characteristics; using a best matching ML model type with reduced performance; or turning on a data collection function.

[0175] Clause 18: A method for wireless communications at a network entity, comprising: determining deployment characteristics of a cell served by the network entity, and receiving an indication that at least a first machine learning (ML) model deployed at a user equipment (UE) is applicable to the cell based on at least a first type of generalization characteristics supported by the first ML model and the deployment characteristics.

[0176] Clause 19: The method of clause 18, further comprising: sending an indication regarding said deployment characteristics.

[0177] Clause 20: The method of any of clauses 18-19, wherein: the at least first type of generalization properties comprises different types of generalization properties supported by different ML models including the first ML model.

[0178] Clause 21: The method of clause 20, wherein at least one type of the different types of generalized properties has one or more subtypes.

[0179] Clause 22: The method of clause 21, wherein: each of the one or more subtypes is defined by one or more classifications of generalized characteristics, and the one or more classifications are based on at least one of the following: base station location information, cell range, propagation characteristics, antenna configuration, or a combination of parameters indicated by a network entity.

[0180] Clause 23: The method of clause 20 or clause 21, wherein: different ML models are applied to cell-specific models or groups of cells with similar deployment characteristics.

[0181] Clause 24: The method of any of clauses 18-23, further comprising: receiving signaling for registering the first type of generalized characteristics supported by the first ML model with the network entity.

[0182] Clause 25: The method of any of clauses 18-24, further comprising: sending an indication of a type of generalization properties supported by the first ML model.

[0183] Clause 26: The method of clause 25, wherein the indication of the types of generalized features supported by the first ML model is sent via at least one of broadcast signaling or multicast signaling.

[0184] Clause 27: The method of clause 25 or clause 26, further comprising: receiving signaling confirming that the UE received the indication of the type of generalized characteristics supported by the first ML model.

[0185] Clause 28: The method of any one of clauses 18 to 27 further includes: receiving a report about model out-of-coverage detected by the UE, indicating that the data to be input to the first ML model is outside the generalization characteristics of the type supported by the first ML model; and together with or after receiving the report about model out-of-coverage, receiving at least one of a request to deactivate the first ML model or a request for ML model type switching.

[0186] Clause 29: The method of any of clauses 1-16, wherein the first ML model is deployed at the UE.

[0187] Clause 30: An apparatus comprising: a memory including executable instructions; and a processor configured to execute the executable instructions and cause the apparatus to perform the method according to any of clauses 1-29.

[0188] Clause 31: An apparatus comprising means for performing the method according to any of clauses 1-29.

[0189] Clause 32: A non-transitory computer-readable medium comprising executable instructions that, when executed by a processor of an apparatus, cause the apparatus to perform the method according to any of clauses 1-29.

[0190] Clause 33: A computer program product embodied on a computer readable storage medium, comprising code for performing the method according to any of clauses 1-29. Other considerations

[0191] The foregoing description is provided to enable any person skilled in the art to implement the various aspects described herein. The examples discussed herein are not limited to the scope, applicability or aspects set forth in the claims. Various modifications to these aspects will be apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects. For example, without departing from the scope of the present disclosure, the functions and arrangements of the elements discussed may be changed. Each example may ignore, replace or add each process or component as appropriate. For example, the described method may be performed in a different order than described, and various steps may be added, omitted or combined. In addition, the features described for some examples may be combined in some other examples. For example, any number of aspects set forth herein may be used to implement a device or practice method. In addition, the scope of the present disclosure is intended to cover such a device or method implemented using other structures, functions, or structures and functions that are supplemented or substituted for the various aspects of the present 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 the claims.

[0192] The various illustrative logic blocks, modules, and circuits described in conjunction with the present disclosure may be implemented or performed using 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 an alternative, the processor may be any commercially available processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, for example, 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.

[0193] 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, ab, ac, bc, and abc, as well as any combination with multiples of the same element (e.g., aa, aaa, aab, aac, abb, acc, bb, bbb, bbc, cc, and ccc or any other ordering of a, b, and c).

[0194] As used herein, the term "determine" encompasses a wide variety of actions. For example, "determine" may include calculating, computing, processing, deriving, investigating, searching (e.g., searching in a table, a database, or another data structure), ascertaining, and the like. Also, "determine" may include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory), and the like. Furthermore, "determine" may include resolving, selecting, choosing, establishing, and the like.

[0195] The method disclosed herein includes one or more actions for implementing the method. The actions of the method can be interchangeable with each other without departing from the scope of the claims. That is, unless a specific order of actions is specified, the order and / or use of specific actions can be modified without departing from the scope of the claims. In addition, the various operations of the method described above can be performed by any appropriate unit that can perform the corresponding function. The unit may include various hardware and / or software components and / or modules, including but not limited to circuits, application specific integrated circuits (ASICs) or processors.

[0196] The attached claims are not intended to be limited to the various aspects shown herein, but are given the full scope consistent with the text of the claims. Within the claims, unless explicitly stated as such, reference to an element in the singular form is not intended to mean "one and only one", but "one or more". Unless otherwise specifically stated, the term "some" refers to one or more. No element of a claim is to be interpreted according to 35 U.S.C. § 112 (f) unless the element is explicitly stated using the phrase "unit for...". All structural and functional equivalents of the elements of the various aspects described throughout the present disclosure that are known or will be known later to a person of ordinary skill in the art are expressly incorporated herein by reference and are intended to be covered by the claims. In addition, nothing disclosed herein is intended to be dedicated to the public, regardless of whether such disclosure is explicitly stated in the claims.

Claims

1. An apparatus for wireless communication by a user equipment (UE), comprising: A memory including executable instructions, and one or more processors, wherein the one or more processors are configured to execute the executable instructions and cause the UE to perform the following operations: storing at least a first type of generalization features supported by a first machine learning (ML) model; and Applicability of the first ML model to the cell is determined based on a first type of generalization characteristics supported by the first ML model and deployment characteristics of the cell.

2. The device according to claim 1, wherein: The one or more processors are further configured to execute the executable instructions and cause the UE to perform the following operations: receive an indication of the deployment characteristic from a network entity of the cell.

3. The device according to claim 1, wherein: The at least first type of generalization properties includes different types of generalization properties supported by different ML models including the first ML model.

4. The device according to claim 3, wherein: At least one type of the different types of generalized properties has one or more subtypes.

5. The device according to claim 4, wherein: Each of the one or more subtypes is defined by one or more classifications of generalized properties, and The one or more classifications are based on at least one of: base station location information, a range of a cell, propagation characteristics, an antenna configuration, or a combination of parameters indicated by a network entity.

6. The device according to claim 4, wherein: The one or more processors are further configured to execute the executable instructions and cause the UE to perform the following operations: indicate that at least one ML model among the different ML models is applicable to one or more cells deployed according to the one or more subtypes.

7. The device according to claim 3, wherein: Different ML models are applied to groups of cells with similar deployment characteristics.

8. The device according to claim 7, wherein: At least one ML model type applies to at least one of the following: A specific cell indicated by a global cell ID; a cell group indicated by a cell ID, a RAN notification area ID, a paging area ID, or a list of IDs for ML model cell groups; or All cells or a subset of cells with the same network ID.

9. The device according to claim 3, wherein: The different types of generalized characteristics include at least one of: a delay spread range, a path loss range, a UE speed range, a number of sectors in a cell, or an area map in a cell.

10. The device according to claim 3, wherein: The different ML models support a range of reference signal (RS) configuration IDs.

11. The device according to claim 1, wherein: The one or more processors are further configured to execute the executable instructions and cause the UE to perform the following operations: provide a generalized description of the first ML model to a network entity.

12. The device according to claim 1, wherein: The one or more processors are further configured to execute the executable instructions and cause the UE to receive an indication of the first type of generalized characteristics supported by the first ML model from a network entity.

13. The device according to claim 12, wherein: The indication of the first type of generalized properties supported by the first ML model is received via at least one of broadcast signaling or multicast signaling.

14. The device according to claim 12, wherein: The one or more processors are further configured to execute the executable instructions and cause the UE to perform the following operations: send signaling to confirm receipt of the indication of the first type of generalized characteristics supported by the first ML model.

15. The device according to claim 1, wherein: The deployment characteristics of the cell include at least one of the following: a deployment location, or one or more characteristics of a deployment location.

16. The device according to claim 1, wherein The one or more processors are further configured to execute the executable instructions and cause the UE to perform the following operations: reporting model out-of-coverage to a network entity when data to be input to the first ML model is outside of the first type of generalized properties supported by the first ML model; and One or more actions are performed in response to detecting the model coverage out.

17. The device according to claim 16, wherein: The one or more actions include at least one of the following: requesting the network entity to deactivate the first ML model; requesting a switch from the first ML model to a second ML model by indicating currently observed channel characteristics; Use the best matching ML model type with reduced performance; or Turn on data collection.

18. An apparatus for wireless communication by a network entity, comprising: a memory including executable instructions, and One or more processors, wherein the one or more processors are configured to execute the executable instructions and cause the network entity to perform the following operations: determining deployment characteristics of cells served by the network entity; and An indication is received that at least a first machine learning (ML) model deployed at a user equipment (UE) is applicable to the cell based on at least a first type of generalization characteristic supported by the first ML model and the deployment characteristic.

19. The device according to claim 18, wherein The one or more processors are further configured to execute the executable instructions and cause the network entity to perform the following operations: send an indication of the deployment characteristic.

20. The apparatus of claim 18, wherein: The at least first type of generalization properties includes different types of generalization properties supported by different ML models including the first ML model.

21. The device according to claim 20, wherein: At least one type of the different types of generalized properties has one or more subtypes.

22. The apparatus of claim 21, wherein: Each of the one or more subtypes is defined by one or more classifications of generalized properties, and The one or more classifications are based on at least one of: base station location information, a range of a cell, propagation characteristics, an antenna configuration, or a combination of parameters indicated by a network entity.

23. The apparatus of claim 20, wherein: Different ML models are suitable for cell-specific models or groups of cells with similar deployment characteristics.

24. The device according to claim 18, wherein The one or more processors are further configured to execute the executable instructions and cause the network entity to perform the following operations: receive signaling for registering the first type of generalized characteristics supported by the first ML model with the network entity.

25. The apparatus of claim 18, wherein: The one or more processors are further configured to execute the executable instructions and cause the network entity to perform the following operations: send an indication of the first type of generalization characteristics supported by the first ML model.

26. The device according to claim 25, wherein The indication of the first type of generalized properties supported by the first ML model is sent via at least one of broadcast signaling or multicast signaling.

27. The device according to claim 25, wherein: The one or more processors are further configured to execute the executable instructions and cause the network entity to perform the following operations: receive signaling for confirming that the UE received the indication of the first type of generalized characteristics supported by the first ML model.

28. The apparatus of claim 18, wherein: The one or more processors are further configured to execute the executable instructions and cause the network entity to perform the following operations: receiving a report of model out-of-coverage detected by the UE, indicating that data to be input to the first ML model is outside of the first type of generalized characteristics supported by the first ML model; as well as Together with or after receiving the report of model out of coverage, at least one of a request to deactivate the first ML model or a request for ML model type switching is received.

29. A method for wireless communication at a user equipment (UE), comprising: storing at least a first type of generalization features supported by a first machine learning (ML) model; as well as The applicability of the first ML model to the cell is determined based on generalization characteristics of the first type supported by the first ML model and deployment characteristics of the cell.

30. A method for wireless communication at a network entity, comprising: determining deployment characteristics of cells served by the network entity; as well as An indication is received that at least a first machine learning (ML) model deployed at a user equipment (UE) is applicable to the cell based on at least a first type of generalization characteristic supported by the first ML model and the deployment characteristic.

Images