Machine learning data collection, verification and reporting configuration

Abstract

A device in a wireless network may process information with machine learning associated with a model ID, a machine learning function, or a machine learning case, and report data via a wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning case. A device may provide a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning case; and may receive a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning case.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. patent application serial number 17 / 806,453, filed on June 10, 2022, entitled “MACHINE LEARNING DATA COLLECTION, VALIDATION, AND REPORTING CONFIGURATIONS,” which is expressly incorporated herein by reference in its entirety. Technical Field

[0003] The present disclosure relates generally to communication systems and, more particularly, to wireless communications including machine learning. Background Art

[0004] Wireless communication systems are widely deployed to provide a variety of telecommunication services, such as telephony, video, data, messaging, and broadcast. Typical wireless communication systems may employ multiple access technologies capable of supporting communication with multiple users by sharing available system resources. Examples of such multiple access technologies include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency division multiple access (FDMA) systems, orthogonal frequency division multiple access (OFDMA) systems, single carrier frequency division multiple access (SC-FDMA) systems, and time division synchronous code division multiple access (TD-SCDMA) systems.

[0005] These multiple access technologies have been adopted in various telecommunication standards to provide a common protocol that enables different wireless devices to communicate at city, country, region, and even global levels. An example telecommunication standard is 5G New Radio (NR). 5G NR is part of the continuous mobile broadband evolution promulgated by the Third Generation Partnership Project (3GPP) to meet new requirements associated with latency, reliability, security, scalability (e.g., with the Internet of Things (IoT)) and other requirements. 5GNR includes services associated with enhanced mobile broadband (eMBB), massive machine type communication (mMTC), and ultra-reliable low latency communication (URLLC). Certain aspects of 5G NR may be based on 4G Long Term Evolution (LTE) standards. It is necessary to further improve 5G NR technology. In addition, these improvements may also be applicable to other multiple access technologies and telecommunication standards that employ these technologies. Summary of the invention

[0006] The following presents a summary of one or more aspects in order to provide a basic understanding of these aspects. This summary is not an extensive review of all contemplated aspects. This summary is neither intended to identify the key or important elements of all aspects, nor to describe the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to a more detailed description presented later.

[0007] In one aspect of the present disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus processes information using machine learning associated with a model identifier (ID), a machine learning function, or a machine learning use case; and reports data via wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case.

[0008] In one aspect of the present disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus provides a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and receives a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case.

[0009] To achieve the aforementioned and related purposes, one or more aspects include the features described in full below and particularly pointed out in the claims. The following description and the accompanying drawings set forth in detail some exemplary features of one or more aspects. However, these features are only indicative of only some of the various ways in which the principles of the various aspects can be employed. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a diagram illustrating an example of a wireless communication system and an access network according to various aspects of the present disclosure.

[0011] Figure 2A is a diagram illustrating an example of a first frame according to various aspects of the present disclosure.

[0012] Figure 2B is a diagram illustrating an example of DL channels within a subframe according to various aspects of the present disclosure.

[0013] Figure 2C is a diagram illustrating an example of a second frame according to various aspects of the present disclosure.

[0014] Figure 2D is a diagram illustrating an example of UL channels within a subframe according to various aspects of the present disclosure.

[0015] Figure 3 is a diagram illustrating an example of a base station and a user equipment (UE) in an access network according to various aspects of the present disclosure.

[0016] Figure 4 is a diagram illustrating various aspects of machine learning or artificial intelligence training and inference according to various aspects of the present disclosure.

[0017] Figure 5 Example observations of machine learning models based on different combinations of features and labels according to various aspects of the present disclosure are illustrated.

[0018] Figure 6 An example communication flow including configuration of a measurement object and reporting based on the configuration is illustrated.

[0019] Figure 7 Communication flows including example aspects of data collection and validation for machine learning models according to various aspects of the present disclosure are illustrated.

[0020] Figure 8 Communication flows including example aspects of data collection and validation for machine learning models according to various aspects of the present disclosure are illustrated.

[0021] Fig. 9 Communication flows including example aspects of data collection and validation for machine learning models according to various aspects of the present disclosure are illustrated.

[0022] Fig.10 Communication flows including example aspects of data collection and validation for machine learning models according to various aspects of the present disclosure are illustrated.

[0023] Fig.11 Communication flows including example aspects of data collection and validation for machine learning models according to various aspects of the present disclosure are illustrated.

[0024] Fig.12 Example aspects of data collection, processing, and validation for machine learning models in accordance with various aspects of the present disclosure are illustrated.

[0025] Fig.13 Communication flows including example aspects of data collection and validation for machine learning models according to various aspects of the present disclosure are illustrated.

[0026] Fig.14A and Fig. 14B is a flow chart of a method of wireless communication according to various aspects of the present disclosure.

[0027] Fig.15A and Fig. 15B is a flow chart of a method of wireless communication according to various aspects of the present disclosure.

[0028] Fig.16 is a diagram illustrating an example of a hardware implementation of an example apparatus.

[0029] Fig.17 is a diagram illustrating an example of a hardware implementation of an example network entity.

[0030] Fig.18 is a diagram of an example flow chart illustrating a method of model registration according to various aspects of the present disclosure.

[0031] Fig.19 is a diagram illustrating an example of a hardware implementation of an example model registration. DETAILED DESCRIPTION

[0032] The performance of an artificial intelligence (AI) or machine learning (ML) model may depend on the quality of the data set used in conjunction with the AI / ML model. As an example, a large amount of unnecessary feature space (which may be referred to as input parameters) may lead to overfitting. For example, overfitting may refer to model predictions or inferences on field data that lose accuracy (e.g., low accuracy due to a bias toward the training data) in order to fit the training input data. The selection of feature space (e.g., input parameter set) may help avoid overfitting. Principal component analysis (PCA) techniques may be used to select feature space (e.g., input parameter set) to improve AI / ML performance by avoiding overfitting.

[0033] Similarly, an unbalanced dataset may have a negative impact on AI / ML model performance. For example, if there is a larger number of observations for one class than for another class, the AI / ML model may tend to overfit to the class with more observations. To avoid having unbalanced observations in a dataset, the data may be made independent and identically distributed (iid).

[0034] The various aspects presented herein provide AI / ML data collection, reporting and / or verification aspects that can improve AI / ML performance. The various aspects presented herein provide data collection and reporting (including processing of data for inference or training) to be configured and requested per AI / ML model identifier (ID), per AI / ML use case, or per AI / ML function. The various aspects presented herein also provide data verification to determine whether the collected data meets the included standards for training and inference based on AI / ML models. The various aspects presented herein can help reduce the use of air resources by processing and reporting verified data and skipping the processing and / or reporting of data that does not meet the verification criteria. The various aspects presented herein can also help reduce inference and online training delays by reducing data processing delays.

[0035] The specific embodiments described below in conjunction with the accompanying drawings are descriptions of various configurations and do not represent the only configurations in which the concepts described herein can be practiced. In order to provide a thorough understanding of the various concepts, the specific embodiments include specific details. However, these concepts can be practiced without these specific details. In some cases, well-known structures and components are shown in block diagram form to avoid blurring such concepts.

[0036] Several aspects of the telecommunication system are also presented with reference to various apparatuses and methods. These apparatuses and methods are described in the following detailed description and are illustrated in the accompanying drawings by various blocks, components, circuits, processes, algorithms, etc. (collectively referred to as "elements"). These elements may be implemented using electronic hardware, computer software, or any combination thereof. Whether these elements are implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system.

[0037] As an example, an element, or any part of an element, or any combination of elements can be implemented as a "processing system", which includes one or more processors. Examples of processors include microprocessors, microcontrollers, graphics processing units (GPUs), central processing units (CPUs), application processors, digital signal processors (DSPs), reduced instruction set computing (RISC) processors, systems on chip (SoCs), baseband processors, field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gating logic, discrete hardware circuits, and other suitable hardware configured to perform various functionalities described throughout the present disclosure. One or more processors in a processing system can execute software. Whether referred to as software, firmware, middleware, microcode, hardware description language, or other terms, software should be broadly interpreted as meaning instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software components, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, or any combination thereof.

[0038] Thus, in one or more example aspects, specific implementations and / or use cases, the described functionality may be implemented in hardware, software, or any combination thereof. If implemented in software, the functionality may be stored or encoded as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media. Storage media may be any available media that can be accessed by a computer. For example, such computer-readable media may include random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), optical disk storage, magnetic disk storage, other magnetic storage devices, combinations of these types of computer-readable media, or any other medium that can be used to store computer-executable code in the form of instructions or data structures that can be accessed by a computer.

[0039] Although various aspects, specific implementations and / or use cases are described in this application by the illustration of some examples, additional or different aspects, specific implementations and use cases may be generated in many different arrangements and scenarios. The various aspects, specific implementations and / or use cases described herein can be implemented across many different platform types, devices, systems, shapes, sizes and packaging arrangements. For example, various aspects, specific implementations and / or use cases can be generated via integrated chip specific implementations and other devices based on non-module components (e.g., end-user devices, vehicles, communication equipment, computing equipment, industrial equipment, retail / purchase equipment, medical equipment, artificial intelligence (AI) enabled devices, etc.). Although some examples may or may not be specifically for use cases or applications, the described examples may have a wide range of applicability. Various aspects, specific implementations and / or use cases can be in the range from chip-level or modular components to non-modular, non-chip-level specific implementations, and further to the range of aggregated, distributed or original equipment manufacturer (OEM) equipment or systems in conjunction with one or more technologies herein. In some actual settings, the equipment combined with the various aspects and features described may also include additional components and features for implementing and practicing the claimed and described aspects. For example, the transmission and reception of wireless signals necessarily include several components for analog and digital purposes (e.g., hardware components including antennas, RF chains, power amplifiers, modulators, buffers, processors, interleavers, adders / summers, etc.). The various techniques described herein can be practiced in a wide variety of devices, chip-level components, systems, distributed arrangements, aggregated or disaggregated components, end-user devices, etc., of various sizes, shapes, and configurations.

[0040] The deployment of a communication system (such as a 5G NR system) can be arranged with various components or constituent parts in a variety of ways. In a 5G NR system or network, a network node, a network entity, a mobility element of a network, a radio access network (RAN) node, a core network node, a network element or a network equipment (such as a base station (BS), or one or more units (or one or more components) that perform base station functionality can be implemented in an aggregated or decomposed architecture. For example, a BS (such as a Node B (NB), an evolved NB (eNB), an NR BS, a 5G NB, an access point (AP), a transmit receive point (TRP) or a cell, etc.) can be implemented as an aggregated base station (also referred to as an independent BS or a monolithic BS) or a decomposed base station.

[0041] A clustered base station may be configured to utilize a radio protocol stack physically or logically integrated within a single RAN node. A decomposed base station may be configured to utilize a protocol stack physically or logically distributed between two or more units, such as one or more central or centralized units (CUs), one or more distributed units (DUs), or one or more radio units (RUs). In some aspects, a CU may be implemented within a RAN node, and one or more DUs may be co-located with the CU, or alternatively, may be geographically or virtually distributed in one or more other RAN nodes. A DU may be implemented to communicate with one or more RUs. Each of a CU, a DU, and a RU may be implemented as a virtual unit, i.e., a virtual central unit (VCU), a virtual distributed unit (VDU), or a virtual radio unit (VRU).

[0042] Base station operation or network design may take into account the aggregated nature of base station functionality. For example, a disaggregated base station may be used in an integrated access backhaul (IAB) network, an open radio access network (O-RAN (such as a network configuration sponsored by the O-RAN Alliance)), or a virtualized radio access network (vRAN, also known as a cloud radio access network (C-RAN)). Decomposition may include distributing functionality across two or more units at various physical locations, as well as virtually distributing functionality of at least one unit, which may enable flexibility in network design. Individual units of a disaggregated base station or disaggregated RAN architecture may be configured for wired or wireless communication with at least one other unit.

[0043] Figure 1 1 is a diagram 100 illustrating an example of a wireless communication system and an access network. The illustrated wireless communication system includes a disaggregated base station architecture. The disaggregated base station architecture may include one or more CUs 110 that may communicate directly with a core network 120 via a backhaul link, or indirectly with the core network 120 through one or more disaggregated base station units, such as a near real-time (near RT) RAN intelligent controller (RIC) 125 via an E2 link, or a non-real-time (non-RT) RIC 115 associated with a service management and orchestration (SMO) framework 105, or both. The CU 110 may communicate with one or more DUs 130 via corresponding midhaul links, such as an F1 interface. The DU 130 may communicate with one or more RUs 140 via corresponding fronthaul links. The RU 140 may communicate with corresponding UEs 104 via one or more radio frequency (RF) access links. In some implementations, a UE 104 may be served simultaneously by multiple RUs 140.

[0044] Each of the units (i.e., CU 110, DU 130, RU 140, and near-RT RIC 125, non-RT RIC 115, and SMO framework 105) may include or be coupled to one or more interfaces configured to receive or send signals, data, or information (collectively, signals) via a wired or wireless transmission medium. Each of the units or an associated processor or controller that provides instructions to the communication interfaces of these units may be configured to communicate with one or more of the other units via a transmission medium. For example, these units may include a wired interface that is configured to receive signals or send signals to one or more of the other units via a wired transmission medium. Additionally, these units may include a wireless interface that may include a receiver, transmitter, or transceiver (such as an RF transceiver) configured to receive and / or send signals to one or more of the other units via a wireless transmission medium.

[0045] In some aspects, CU 110 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), etc. Each control function may be implemented as having an interface configured to communicate signals with other control functions hosted by CU 110. CU110 may be configured to handle user plane functionality (i.e., central unit-user plane (CU-UP)), control plane functionality (i.e., central unit-control plane (CU-CP)), or a combination thereof. In some specific implementations, CU 110 may be logically split 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). As needed, CU 110 may be implemented to communicate with DU 130 for network control and signaling.

[0046] DU 130 may correspond to a logical unit that includes one or more base station functions for controlling the operation of one or more RUs 140. In some aspects, DU 130 may host one or more of a radio link control (RLC) layer, a medium access control (MAC) layer, and one or more high physical (PHY) layers (such as modules for forward error correction (FEC) encoding and decoding, scrambling, modulation and demodulation, etc.) at least in part according to a functional split such as those defined by 3GPP. In some aspects, DU 130 may further host 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) hosted by DU 130 or with control functions hosted by CU 110.

[0047] The lower layer functionality may be implemented by one or more RUs 140. In some deployments, the RU 140 controlled by the DU 130 may correspond to a logical node that hosts RF processing functions or low PHY layer functions (such as performing Fast Fourier Transform (FFT), Inverse FFT (iFFT), digital beamforming, physical random access channel (PRACH) extraction and filtering, etc.), or both, based at least in part on a functional split (such as a lower layer functional split). In such an architecture, the RU 140 may be implemented to handle over-the-air (OTA) communications with one or more UEs 104. In some implementations, real-time and non-real-time aspects of control and user plane communications with the RU 140 may be controlled by the corresponding DU 130. In some scenarios, this configuration may enable the DU 130 and the CU 110 to be implemented in a cloud-based RAN architecture (such as a vRAN architecture).

[0048] The SMO framework 105 may be configured to support RAN deployment and configuration of non-virtualized network elements and virtualized network elements. For non-virtualized network elements, the SMO framework 105 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 105 may be configured to interact with a cloud computing platform (such as an open cloud (O-Cloud) 190) to perform network element lifecycle management (such as 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 110, DU 130, RU 140, and near-RT RIC 125. In some specific implementations, the SMO framework 105 may communicate with hardware aspects of the 4G RAN (such as an open eNB (O-eNB) 111) via the O1 interface. Additionally, in some specific implementations, the SMO framework 105 may communicate directly with one or more RUs 140 via the O1 interface. The SMO framework 105 may also include a non-RT RIC 115 configured to support the functionality of the SMO framework 105 .

[0049] The non-RT RIC 115 may be configured to include logic functions that enable non-real-time control and optimization of RAN elements and resources, artificial intelligence (AI) / machine learning (ML) (AI / ML) workflows including model training and updating, or policy-based guidance of applications / features in the near-RT RIC 125. The non-RT RIC 115 may be coupled to or in communication with the near-RT RIC 125 (such as via an A1 interface). The near-RT RIC 125 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 through an interface (such as via an E2 interface) that connects one or more CUs 110, one or more DUs 130, or both, and the O-eNB with the near-RT RIC 125.

[0050] In some implementations, in order to generate an AI / ML model to be deployed in the near-RT RIC 125, the non-RT RIC 115 may receive parameters or external enrichment information from an external server. Such information may be utilized by the near-RT RIC 125 and may be received from a non-network data source or from a network function at the SMO framework 105 or the non-RT RIC 115. In some examples, the non-RT RIC 115 or the near-RT RIC 125 may be configured to tune RAN behavior or performance. For example, the non-RT RIC 115 may monitor long-term trends and patterns for performance and employ AI / ML models to perform corrective actions through the SMO framework 105 (such as via reconfiguration of O1) or via creation of RAN management policies (such as A1 policies).

[0051] At least one of the CU 110, the DU 130, and the RU 140 may be referred to as a base station 102. Therefore, the base station 102 may include one or more of the CU 110, the DU 130, and the RU 140 (each component is indicated by a dotted line to indicate that each component may be included in the base station 102 or may not be included in the base station). The base station 102 provides an access point to the core network 120 for the UE 104. The base station 102 may include a macro cell (a high-power cellular base station) and / or a small cell (a low-power cellular base station). Small cells include femto cells, pico cells, and micro cells. A network including both small cells and macro cells may be referred to as a heterogeneous network. A heterogeneous network may also include a home evolved Node B (eNB) (HeNB), which may provide services to a restricted group referred to as a closed subscriber group (CSG). The communication link between RU 140 and UE 104 may include uplink (UL) (also known as reverse link) transmission from UE 104 to RU 140 and / or downlink (DL) (also known as forward link) transmission from RU 140 to UE 104. The communication link may use multiple-input multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming and / or transmit diversity. The communication link may be through one or more carriers. For each carrier allocated in a carrier aggregation of up to Yx MHz (x component carriers) for transmission in each direction, the base station 102 / UE 104 may use a spectrum of up to Y MHz (e.g., 5MHz, 10MHz, 15MHz, 20MHz, 100MHz, 400MHz, etc.) bandwidth. The carriers may or may not be adjacent to each other. The allocation of carriers may be asymmetric with respect to DL and UL (e.g., more or fewer carriers may be allocated to DL compared to UL). The component carrier may include a primary component carrier and one or more secondary component carriers. The primary component carrier may be referred to as a primary cell (PCell) and the secondary component carrier may be referred to as a secondary cell (SCell).

[0052] Some UEs 104 may communicate with each other using a device-to-device (D2D) communication link 158. The D2D communication link 158 may use DL / UL wireless wide area network (WWAN) spectrum. 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), and a physical sidelink control channel (PSCCH). The D2D communication may be through a variety of wireless D2D communication systems, such as, for example, Bluetooth, Wi-Fi based on the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard, LTE, or NR.

[0053] The wireless communication system may also include a Wi-Fi AP 150 that communicates with a UE 104 (also referred to as a Wi-Fi station (STA)) via a communication link 154, for example, in a 5 GHz unlicensed spectrum, etc. When communicating in an unlicensed spectrum, the UE 104 / AP 150 may perform a clear channel assessment (CCA) to determine whether a channel is available prior to communication.

[0054] The electromagnetic spectrum is typically subdivided into various categories, bands, channels, etc. based on frequency / wavelength. In 5G NR, two initial operating bands have been identified as frequency range designations FR1 (410 MHz-7.125 GHz) and FR2 (24.25 GHz-52.6 GHz). Although a portion of FR1 is greater than 6 GHz, FR1 is often (interchangeably) referred to as the "sub-6 GHz" band in various documents and articles. A similar naming issue sometimes occurs with respect to FR2, which is often (interchangeably) referred to as the "millimeter wave" band in documents and articles, although different from the extremely high frequency (EHF) band (30 GHz-300 GHz) identified as the "millimeter wave" band by the International Telecommunication Union (ITU).

[0055] Frequencies between FR1 and FR2 are generally referred to as mid-band frequencies. Recent 5G NR research has identified the operating bands for these mid-band frequencies as frequency range designation FR3 (7.125GHz–24.25GHz). The bands falling within FR3 can inherit FR1 characteristics and / or FR2 characteristics, and therefore the features of FR1 and / or FR2 can be effectively extended to mid-band frequencies. In addition, higher frequency bands are currently being explored to extend 5G NR operations to more than 52.6GHz. For example, three higher operating bands have been identified as frequency range designations FR2-2 (52.6GHz–71GHz), FR4 (71GHz–114.25GHz), and FR5 (114.25GHz–300GHz). Each of these higher frequency bands falls within the EHF band.

[0056] In view of the above, unless otherwise specified, if the term "6 GHz or less" is used herein, it may broadly refer to frequencies that may be less than 6 GHz, may be within FR1, or may include mid-band frequencies. In addition, unless otherwise specified, if the term "millimeter wave" is used herein, it may broadly refer to frequencies that may include mid-band frequencies, may be within FR2, FR4, FR2-2, and / or FR5, or may be within the EHF band.

[0057] The base station 102 and the UE 104 may each include multiple antennas (such as antenna elements, antenna panels, and / or antenna arrays) to facilitate beamforming. The base station 102 may send a beamformed signal 182 to the UE 104 in one or more transmit directions. The UE 104 may receive the beamformed signal from the base station 102 in one or more receive directions. The UE 104 may also send a beamformed signal 184 to the base station 102 in one or more transmit directions. The base station 102 may receive the beamformed signal from the UE 104 in one or more receive directions. The base station 102 / UE 104 may perform beam training to determine the best receive direction and transmit direction for each of the base station 102 / UE 104. The transmit direction and receive direction of the base station 102 may be the same or different. The transmit direction and receive direction of the UE 104 may be the same or different.

[0058] The base station 102 may include and / or be referred to as a gNB, a Node B, an eNB, an access point, a base transceiver station, a radio base station, a radio transceiver, a transceiver function, a basic service set (BSS), an extended service set (ESS), a transmit receive point (TRP), a network node, a network entity, a network equipment, or some other suitable term. The base station 102 may be implemented as an integrated access and backhaul (IAB) node, a relay node, a sidelink node, a clustered (monolithic) base station with a baseband unit (BBU) (including a CU and a DU) and a RU, or as a decomposed base station including one or more of a CU, a DU, and / or a RU. A set of base stations that may include decomposed base stations and / or aggregated base stations may be referred to as a next generation (NG) RAN (NG-RAN).

[0059] The core network 120 may include an access and mobility management function (AMF) 161, a session management function (SMF) 162, a user plane function (UPF) 163, a unified data management (UDM) 164, one or more location servers, and other functional entities. AMF 161 is a control node that processes signaling between UE 104 and the core network 120. AMF 161 supports registration management, connection management, mobility management, and other functions. SMF 162 supports session management and other functions. UPF 163 supports packet routing, packet forwarding, and other functions. UDM 164 supports the generation of authentication and key agreement (AKA) credentials, user identity processing, access authorization, and subscription management. One or more location servers are exemplified as including a gateway mobile location center (GMLC) 165 and a location management function (LMF) 166. However, in general, the one or more location servers may include one or more location / positioning servers, which may include one or more of the GMLC 165, the LMF 166, a position determination entity (PDE), a serving mobile location center (SMLC), a mobile positioning center (MPC), etc. The GMLC 165 and the LMF 166 support UE location services. The GMLC 165 provides an interface for clients / applications (e.g., emergency services) to access UE positioning information. The LMF 166 receives measurements and assistance information from the NG-RAN and the UE 104 via the AMF 161 to calculate the location of the UE 104. The NG-RAN may determine the location of the UE 104 using one or more positioning methods. Positioning the UE 104 may involve signal measurements, position estimation, and optional speed calculations based on these measurements. Signal measurements may be performed by the UE 104 and / or the serving base station 102. The measured signals may be based on a satellite positioning system (SPS) 170 (e.g., one or more of a global navigation satellite system (GNSS), a global positioning system (GPS), a non-terrestrial network (NTN), or other satellite positioning / positioning systems), LTE signals, wireless local area network (WLAN) signals, Bluetooth signals, a terrestrial beacon system (TBS), sensor-based information (e.g., an atmospheric pressure sensor, a motion sensor), an NR enhanced cell ID (NR E-CID) method, NR signals (e.g., multi-round trip time (multi-RTT), DL angle of departure (DL-AoD), DL time difference of arrival (DL-TDOA), UL time difference of arrival (UL-TDOA), and UL angle of arrival (UL-AoA) positioning), and / or one or more of other systems / signals / sensors.

[0060] Examples of UE 104 include cellular phones, smart phones, session initiation protocol (SIP) phones, laptops, personal digital assistants (PDAs), satellite radios, global positioning systems, multimedia devices, video devices, digital audio players (e.g., MP3 players), cameras, game consoles, tablet computers, smart devices, wearable devices, vehicles, electric meters, gas pumps, large or small kitchen appliances, healthcare equipment, implants, sensors / actuators, displays, or any other similarly functional devices. Some of the UEs 104 may be referred to as IoT devices (e.g., parking meters, gas pumps, toasters, vehicles, heart monitors, etc.). UE 104 may also be referred to as stations, mobile stations, subscriber stations, mobile units, subscriber units, wireless units, remote units, mobile devices, wireless devices, wireless communication devices, remote devices, mobile subscriber stations, access terminals, mobile terminals, wireless terminals, remote terminals, mobile phones, user agents, mobile clients, clients, or some other suitable terminology. In some scenarios, the term UE may also be applied to one or more supporting devices, such as in a device constellation arrangement. One or more of these devices may access the network collectively and / or individually.

[0061] Reference again Figure 1 In certain aspects, the UE 104 and / or the base station 102 may include a machine learning component 198 configured to process information using machine learning associated with a model identifier (ID), a machine learning function, or a machine learning use case; and report data via wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case. The UE and / or the base station 102 may include a machine learning configuration component 199 configured to provide a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and receive a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case.

[0062] Although the following description may focus on 5G NR, the concepts described herein may be applicable to other similar areas such as LTE, LTE-A, CDMA, GSM, and other wireless technologies.

[0063] Figure 2A FIG200 is a diagram illustrating an example of a first subframe within a 5G NR frame structure. Figure 2B FIG230 is a diagram illustrating an example of DL channels within a 5G NR subframe. Figure 2C FIG250 is a diagram illustrating an example of a second subframe within a 5G NR frame structure. Figure 2DFIG280 is a diagram illustrating an example of UL channels within a 5G NR subframe. The 5G NR frame structure may be frequency division duplex (FDD), where for a particular set of subcarriers (carrier system bandwidth), subframes within that subcarrier set are dedicated to either DL or UL, or may be time division duplex (TDD), where for a particular set of subcarriers (carrier system bandwidth), subframes within that subcarrier set are dedicated to both DL and UL. Figure 2A , Figure 2C In the example provided, the 5G NR frame structure is assumed to be TDD, where subframe 4 is configured with slot format 28 (most of which are DL), where D is DL, U is UL, and F is flexible between DL / UL, and subframe 3 is configured with slot format 1 (all of which are UL). Although subframes 3 and 4 are shown as having slot formats 1 and 28, respectively, any particular subframe may be configured with any of the various available slot formats 0-61. Slot formats 0 and 1 are full DL and full UL, respectively. Other slot formats 2-61 include a mix of DL, UL and flexible symbols. The slot format is configured for the UE via a received slot format indicator (SFI) (dynamically configured via DL control information (DCI) or semi-statically / statically configured via radio resource control (RRC) signaling). Note that the following description also applies to the 5G NR frame structure as TDD.

[0064] FIG. 2A to FIG. 2D The frame structure is illustrated, and various aspects of the present disclosure may be applicable to other wireless communication technologies that may have different frame structures and / or different channels. A frame (10ms) may be divided into 10 subframes of equal size (1ms). Each subframe may include one or more time slots. A subframe may also include a microslot, which may include 7, 4, or 2 symbols. Each time slot may include 14 or 12 symbols, depending on whether the cyclic prefix (CP) is normal or extended. For a normal CP, each time slot may include 14 symbols, and for an extended CP, each time slot may include 12 symbols. The symbol on the DL may be a CP orthogonal frequency division multiplexing (OFDM) (CP-OFDM) symbol. The symbol on the UL may be a CP-OFDM symbol (for high throughput scenarios) or a discrete Fourier transform (DFT) extended OFDM (DFT-s-OFDM) symbol (also known as a single carrier frequency division multiple access (SC-FDMA) symbol) (for power-limited scenarios; limited to single-stream transmission). The number of slots within a subframe is based on the CP and the parameter set. The parameter set defines the subcarrier spacing (SCS) and effectively defines the symbol length / duration, which is equal to 1 / SCS.

[0065]

[0066] For normal CP (14 symbols / slot), different parameter sets μ0 to 4 allow 1, 2, 4, 8, and 16 slots per subframe, respectively. For extended CP, parameter set 2 allows 4 slots per subframe. Thus, for normal CP and parameter set μ, there are 14 symbols / slot and 2 μ timeslots / subframe. The subcarrier spacing can be equal to 2 μ *15kHz, where μ is parameter set 0 to 4. Therefore, the subcarrier spacing for parameter set μ=0 is 15kHz, and the subcarrier spacing for parameter set μ=4 is 240kHz. The symbol length / duration is inversely related to the subcarrier spacing. FIG. 2A to FIG. 2D An example of a parameter set μ=2 with a normal CP of 14 symbols per slot and 4 slots per subframe is provided. The slot duration is 0.25 ms, the subcarrier spacing is 60 kHz, and the symbol duration is about 16.67 μs. Within a frame set, there may be one or more different bandwidth parts (BWPs) that are frequency-division multiplexed (see Figure 2B ). Each BWP may have a specific parameter set and CP (normal or extended).

[0067] A resource grid may be used to represent the frame structure. Each slot includes a resource block (RB) (also called a physical RB (PRB)) extending over 12 consecutive subcarriers. The resource grid is divided into a number of resource elements (REs). The number of bits carried by each RE depends on the modulation scheme.

[0068] like Figure 2A As illustrated, some of the REs carry reference (pilot) signals (RS) for the UE. The RS may include a demodulation RS (DM-RS) (indicated as R for one particular configuration, but other DM-RS configurations are possible) and 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 a phase tracking RS (PT-RS).

[0069] Figure 2BExamples of various DL channels within a subframe of a frame are illustrated. A physical downlink control channel (PDCCH) carries DCI within one or more control channel elements (CCEs) (e.g., 1, 2, 4, 8, or 16 CCEs), each CCE including six RE groups (REGs), each REG including 12 consecutive REs in an OFDM symbol of an RB. A PDCCH within a BWP may be referred to as a control resource set (CORESET). The UE is configured to monitor PDCCH candidates in a PDCCH search space (e.g., a common search space, a UE-specific search space) during a PDCCH monitoring opportunity on a CORESET, wherein the PDCCH candidates have different DCI formats and different aggregation levels. Additional BWPs may be located at higher and / or lower frequencies on the channel bandwidth. A primary synchronization signal (PSS) may be within symbol 2 of a particular subframe of a frame. The PSS is used by UE 104 to determine subframe / symbol timing and physical layer identification. A secondary synchronization signal (SSS) may be within symbol 4 of a particular subframe of a frame. The SSS is used by the UE to determine the physical layer cell identity group number and the radio frame timing. Based on the physical layer identity and the physical layer cell identity group number, the UE can determine the physical cell identifier (PCI). Based on the PCI, the UE can determine the location of the DM-RS. 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 (also referred to as an SS block (SSB)). 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 not sent via the PBCH (such as a system information block (SIB)), and paging messages.

[0070] like Figure 2C As illustrated, some of the REs carry DM-RS (indicated as R for a particular configuration, but other DM-RS configurations are possible) for channel estimation at the base station. The UE may send a DM-RS for a physical uplink control channel (PUCCH) and a DM-RS for a physical uplink shared channel (PUSCH). The PUSCH DM-RS may be sent 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 DM-RS may be sent in different configurations. The UE may send a sounding reference signal (SRS). The SRS may be sent in the last symbol of a subframe. The SRS may have a comb structure, and the UE may send the SRS on one of the combs in the comb. The SRS may be used by the base station for channel quality estimation to achieve frequency-dependent scheduling of the UL.

[0071] Figure 2DExamples of various UL channels within a subframe of a frame are illustrated. The PUCCH may be located at a position as indicated in one configuration. The PUCCH carries uplink control information (UCI), such as scheduling requests, channel quality indicators (CQI), precoding matrix indicators (PMI), rank indicators (RI), and hybrid automatic repeat request (HARQ) acknowledgement (ACK) (HARQ-ACK) feedback (i.e., one or more HARQ ACK bits indicating one or more ACKs and / or negative ACKs (NACKs)). The PUSCH carries data and may additionally be used to carry buffer status reports (BSRs), power headroom reports (PHRs), and / or UCI.

[0072] Figure 3 375. The Controller / Processor 375 is a block diagram of a base station 310 in an access network communicating with a UE 350. In the DL, Internet Protocol (IP) packets may be provided to a controller / processor 375. The controller / processor 375 implements layer 3 and layer 2 functionality. Layer 3 includes a radio resource control (RRC) layer, and layer 2 includes a service data adaptation protocol (SDAP) layer, a packet data convergence protocol (PDCP) layer, a radio link control (RLC) layer, and a medium access control (MAC) layer. The controller / processor 375 provides RRC layer functionality associated with broadcasting of system information (e.g., MIB, SIB), RRC connection control (e.g., RRC connection paging, RRC connection establishment, RRC connection modification, and RRC connection release), inter-radio access technology (RAT) mobility, and measurement configuration for UE measurement reporting; PDCP layer functionality associated with header compression / decompression, security (encryption, decryption, integrity protection, integrity verification), and handover support functions; RLC layer functionality associated with transmission of upper layer packet data units (PDUs), error correction through ARQ, concatenation, segmentation and reassembly of RLC service data units (SDUs), re-segmentation of RLC data PDUs, and reordering of RLC data PDUs; and MAC layer functionality associated with mapping between logical channels and transport channels, multiplexing of MAC SDUs onto transport blocks (TBs), demultiplexing of MAC SDUs from TBs, scheduling information reporting, error correction through HARQ, priority handling, and logical channel prioritization.

[0073] The transmit (TX) processor 316 and the receive (RX) processor 370 implement layer 1 functionality associated with various signal processing functions. Layer 1 (which includes the physical (PHY) layer) may include error detection on the transport channel, forward error correction (FEC) coding / decoding of the transport channel, interleaving, rate matching, mapping to the physical channel, modulation / demodulation of the physical channel, and MIMO antenna processing. The TX processor 316 processes the mapping to the signal constellation based on various modulation schemes (e.g., binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), M-order phase shift keying (M-PSK), M-order quadrature amplitude modulation (M-QAM)). The coded and modulated symbols may then be separated into parallel streams. Each stream may then be mapped to an OFDM subcarrier, multiplexed with a reference signal (e.g., pilot) in the time and / or frequency domain, and then combined together using an inverse fast Fourier transform (IFFT) to produce a physical channel carrying a time domain OFDM symbol stream. The OFDM stream is spatially pre-coded to generate multiple spatial streams. Channel estimates from a channel estimator 374 may be used to determine the coding and modulation schemes, as well as for spatial processing. Channel estimates may be derived from reference signals and / or channel condition feedback sent by the UE 350. Each spatial stream may then be provided to a different antenna 320 via a separate transmitter 318Tx. Each transmitter 318Tx may modulate a radio frequency (RF) carrier with a corresponding spatial stream for transmission.

[0074] At the UE 350, each receiver 354Rx receives a signal through its corresponding antenna 352. Each receiver 354Rx recovers the information modulated onto the RF carrier and provides the information to a receive (RX) processor 356. The TX processor 368 and the RX processor 356 implement layer 1 functionality associated with various signal processing functions. The RX processor 356 can perform spatial processing on the information to recover any spatial stream destined for the UE 350. If multiple spatial streams are destined for the UE 350, they can be combined into a single OFDM symbol stream by the RX processor 356. The RX processor 356 then converts the OFDM symbol stream from the time domain to the frequency domain using a fast Fourier transform (FFT). The frequency domain signal includes a separate OFDM symbol stream for each subcarrier of the OFDM signal. The symbol on each subcarrier and the reference signal are recovered and demodulated by determining the signal constellation point most likely to be sent by the base station 310. These soft decisions can be based on the channel estimate calculated by the channel estimator 358. The soft decisions are then decoded and deinterleaved to recover the data and control signals originally sent on the physical channel by the base station 310. The data and control signals are then provided to the controller / processor 359, which implements layer 3 and layer 2 functionality.

[0075] The controller / processor 359 may be associated with a memory 360 that stores program codes and data. The memory 360 may be referred to as a computer readable medium. In the UL, the controller / processor 359 provides demultiplexing between transport channels and logical channels, packet reassembly, decryption, header decompression, and control signal processing to recover IP packets. The controller / processor 359 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operations.

[0076] Similar to the functionality described in conjunction with DL transmissions performed by the base station 310, the controller / processor 359 provides RRC layer functionality associated with system information (e.g., MIB, SIB) acquisition, RRC connection, and measurement reporting; PDCP layer functionality associated with header compression / decompression and security (encryption, decryption, integrity protection, integrity verification); RLC layer functionality associated with delivery of upper layer PDUs, error correction through ARQ, concatenation, segmentation and reassembly of RLC SDUs, resegmentation of RLC data PDUs, and reordering of RLC data PDUs; and MAC layer functionality associated with mapping between logical channels and transport channels, multiplexing of MAC SDUs onto TBs, demultiplexing of MAC SDUs from TBs, scheduling information reporting, error correction through HARQ, priority handling, and logical channel prioritization.

[0077] The TX processor 368 may use channel estimates derived by the channel estimator 358 from a reference signal or feedback transmitted by the base station 310 to select appropriate coding and modulation schemes and to facilitate spatial processing. The spatial streams generated by the TX processor 368 may be provided to different antennas 352 via separate transmitters 354Tx. Each transmitter 354Tx may modulate an RF carrier with a corresponding spatial stream for transmission.

[0078] UL transmissions are processed at the base station 310 in a manner similar to that described in conjunction with the receiver functionality at the UE 350. Each receiver 318Rx receives a signal through its corresponding antenna 320. Each receiver 318Rx recovers information modulated onto an RF carrier and provides the information to the RX processor 370.

[0079] The controller / processor 375 may be associated with a memory 376 that stores program codes and data. The memory 376 may be referred to as a computer readable medium. In the UL, the controller / processor 375 provides demultiplexing between transport channels and logical channels, packet reassembly, decryption, header decompression, control signal processing to recover IP packets. The controller / processor 375 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operations.

[0080] At least one of the TX processor 368, the RX processor 356, and the controller / processor 359 may be configured to perform operations related to Figure 1 Various aspects of the combination of the machine learning component 198 and / or the machine learning configuration component 199.

[0081] At least one of the TX processor 316, the RX processor 370, and the controller / processor 375 may be configured to perform operations related to Figure 1 Various aspects of the combination of the machine learning component 198 and / or the machine learning configuration component 199.

[0082] Figure 4 is an example of an AI / ML algorithm 400 that can be used in conjunction with wireless communication. The AI / ML algorithm 400 can include various functions, including data collection 402, model training function 404, model inference function 406, and actor 408.

[0083] Data collection 402 may be a function that provides input data to the model training function 404 and the model inference function 406. The data collection 402 function may include any form of data preparation, and it may not be specific to the specific implementation of the AI / ML algorithm (e.g., data preprocessing and cleaning, formatting, and transformation). Examples of input data may include, but are not limited to, radio measurements from a UE or network node (such as reference signal received power (RSRP) for a cell and / or beam), feedback from an actor 408, output from another AI / ML model. As an example, the measurement may be a reference signal such as a CSI-RS or a measurement of a pre-decoding metric, etc. Data collection 402 may include training data and inference data, where the training data refers to data to be transmitted as input to the AI / ML model training function 404, and the inference data refers to data to be transmitted as input to the AI / ML model inference function 406.

[0084] The model training function 404 may be a function that performs ML model training, validation, and testing, which may generate model performance metrics as part of the model testing procedure. The model training function 404 may also be responsible for data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) based on the training data delivered or received from the data collection 402 function. The model training function 404 may deploy or update the trained, validated, and tested AI / ML model to the model inference function 406, and receive model performance feedback from the model inference function 406.

[0085] The model inference function 406 may be a function that provides an AI / ML model inference output (e.g., a prediction or decision). The model inference function 406 may also perform data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) based on the inferred data delivered from the data collection 402 function. The output of the model inference function 406 may include the inference output of the AI / ML model generated by the model inference function 406. The details of the inference output may be use case specific. As an example, the output may include a predicted beam (e.g., for a beam measurement use case), a predicted CSI-RS RSRP and / or pre-decoding metrics (e.g., for a CSI feedback use case), a predicted UE position (e.g., for a positioning use case), and the like. In some aspects, the actor may be a UE. In some aspects, the actor may be a network node. In some aspects, the UE may report the output to a network node or another UE. In some aspects, the actor may be a network node. The network node may report the output to a UE or another network node.

[0086] Model performance feedback may refer to information derived from model inference functionality 406 that may be suitable for improving the AI / ML model trained in model training functionality 404. Feedback from actors 408 or other network entities (via data collection 402 functionality) may be implemented for model inference functionality 406 to create model performance feedback.

[0087] Actor 408 may be a function that receives output from model inference function 406 and triggers or performs corresponding actions. An actor may trigger actions for network entities including other network entities or itself. Actor 408 may also provide feedback information so that model training function 404 or model inference function 406 can derive training or inference data or performance feedback. Feedback may be sent back to data collection 402.

[0088] The network may use machine learning algorithms, deep learning algorithms, neural networks, reinforcement learning, regression, boosting, or advanced signal processing methods for various aspects of wireless communications, including TCI state-based DCI selection to identify neighboring cell TCI candidates for autonomous TCI candidate set updates.

[0089] In some aspects described herein, the network may train one or more neural networks to learn the dependence of the quality of the measurement on various parameters. Examples of machine learning models or neural networks that may be included in the network entity include, among others: artificial neural networks (ANNs); decision tree learning; convolutional neural networks (CNNs); deep learning architectures, in which the outputs of a first layer of neurons become inputs to a second layer of neurons, etc.; support vector machines (SVMs), for example, which include a separating hyperplane (e.g., a decision boundary) that classifies data; regression analysis; Bayesian networks; genetic algorithms; deep convolutional networks (DCNs), which are configured with additional pooling and normalization layers; and deep belief networks (DBNs).

[0090] A machine learning model, such as an artificial neural network (ANN), may include a group of interconnected artificial neurons (e.g., a neuron model) and may be a computing device or may represent a method to be performed by a computing device. The connections of the neuron model may be modeled as weights. The machine learning model may provide predictive models, adaptive control, and other applications by training via a data set. The model may be adaptive based on external or internal information processed by the machine learning model. Machine learning may provide nonlinear statistical data models or decision making, and may model complex relationships between input data and output information.

[0091] The machine learning model may include multiple layers and / or operations, which may be formed by cascading one or more of the cited operations. Examples of operations that may be involved include: extraction of various features of data, convolution operations, fully connected operations that may be activated or deactivated, compression, decompression, quantization, flattening, etc. As used herein, the "layer" of the machine learning model may be used to represent operations on input data. For example, convolution layers, fully connected layers, etc. may be used to refer to associated operations on data input to the layer. Convolution A×B operations refer to operations that convert multiple input features A into multiple output features B. "Kernel size" may refer to the number of neighboring coefficients combined in a dimension. As used herein, "weights" may be used to represent one or more coefficients used in operations for combining individual rows and / or columns of input data in each layer. For example, a fully connected layer operation may have an output y, which is determined at least in part based on the product of the input matrix x and the weight A (which may be a matrix) and the sum of the bias value B (which may be a matrix). The term "weight" may be generally used herein to refer to both weights and bias values. Weights and biases are examples of parameters of a trained machine learning model. Different layers of a machine learning model can be trained independently.

[0092] Machine learning models may include various connectivity patterns, for example, including any of feedforward networks, hierarchies, recursive architectures, feedback connections, and the like. The connections between layers of a neural network may be fully connected or partially connected. In a fully connected neural network, a neuron in a first layer may communicate its output to each neuron in a second layer, and each neuron in a second layer may receive input from each neuron in the first layer. In a locally connected network, a neuron in a first layer may be connected to a limited number of neurons in a second layer. In some aspects, a convolutional network may be locally connected and configured with a shared connection strength associated with the input of each neuron in the second layer. A locally connected layer of a network may be configured so that each neuron in the layer has the same or similar connectivity pattern, but with different connection strengths.

[0093] A machine learning model or neural network may be trained. For example, a machine learning model may be trained based on supervised learning. During training, the machine learning model may be presented with inputs that the model uses to calculate to produce an output. The actual output may be compared to the target output, and the difference may be used to adjust the parameters of the machine learning model (such as weights and biases) to provide an output closer to the target output. Before training, the output may be incorrect or less accurate, and the error or difference between the actual output and the target output may be calculated. The weights of the machine learning model may then be adjusted so that the output is more closely aligned with the target. In order to adjust the weights, the learning algorithm may calculate the gradient vector of the weights. The gradient may indicate the amount by which the error will increase or decrease if the weights are slightly adjusted. At the top layer, the gradient may directly correspond to the value of the weights connecting the activated neurons in the penultimate layer and the neurons in the output layer. In the lower layers, the gradient may depend on the value of the weights and the calculated error gradients of the higher layers. The weights may then be adjusted to reduce the error or move the output closer to the target. This way of adjusting the weights may be referred to as back propagation through a neural network. The process may continue until the achievable error rate stops decreasing, or until the error rate has reached a target level.

[0094] These machine learning models may include computational complexity and a large number of processors for training machine learning models. The output of one node is connected as an input to another node. The connection between nodes may be referred to as an edge, and weights may be applied to the connection / edge to adjust the output from one node, which is used as an input to another node. Nodes may apply thresholds to determine whether or when to provide output to a connected node. The output of each node may be calculated as a nonlinear function of the sum of the inputs to the node. A neural network may include any number of nodes and any type of connection between the nodes. A neural network may include one or more hidden nodes. Nodes may be aggregated into layers, and different layers of a neural network may perform different kinds of transformations on the input. A signal may travel from an input at the first layer through multiple layers of a neural network to an output at the last layer of the neural network, and may traverse these layers multiple times.

[0095] The performance of an AI or ML model's performance may depend on the quality of the dataset used in conjunction with the AI / ML model. Figure 5 The example shows that the labels (eg, L1, L2 to L m ) and features (e.g., F1, F2 to F n ) 500 . As an example, a large number of unnecessary feature spaces (which may be referred to as input parameters) may lead to overfitting. For example, overfitting may refer to a loss of accuracy (e.g., low accuracy due to a bias toward the training data) when used on field data (or real-world data) in order to fit model predictions or inferences of the training input data. The selection of feature spaces (e.g., input parameter sets) may help avoid overfitting. Principal component analysis (PCA) techniques may be used to select feature spaces (e.g., input parameter sets) to improve AI / ML performance by avoiding overfitting.

[0096] Similarly, an unbalanced dataset may have a negative impact on AI / ML model performance. For example, if there is a larger number of observations for one class than for another class, the AI / ML model may tend to overfit to the class with more observations. To avoid having unbalanced observations in a dataset, the data may be made independent and identically distributed (iid).

[0097] The various aspects presented herein provide AI / ML data collection, reporting and / or verification aspects that can improve AI / ML performance. The various aspects presented herein provide data collection and reporting (including processing of data for inference or training) to be configured (e.g., configured by a network node or UE for another UE, requested by a UE from a network node or another UE, configured by a network node for another network node, etc.). The various aspects presented herein also provide data verification to determine whether the collected data meets the included criteria for training and inference based on AI / ML models. The various aspects presented herein can help reduce the use of air resources by processing and reporting verified data and skipping processing and / or reporting of data that does not meet the verification criteria. The various aspects presented herein can also help reduce inference and online training delays by reducing data processing delays.

[0098] For applications in wireless networks, there are various AI / ML inference and / or training use cases and related data processing procedures. As a first example, AI / ML training and inference can be performed at the UE using UE data. In this example, a model designer (e.g., a UE or UE vendor) can determine the most significant feature space (e.g., a set of input parameters) that can use PCA. Data for training and inference can then be collected at the UE, for example, as combined with Figure 4 The UE may perform data validation, for example, to determine whether the data is balanced. In some aspects, the data validation may be based on iid so that the validated data complies with a defined set of statistics or standards.

[0099] In another example, training and inference may be performed at the UE using UE data and network data. In this example, a model designer (e.g., a UE or UE vendor) may determine the most significant feature space (e.g., a set of input parameters), which may use PCA. In some aspects, the UE may indicate to the network the data to be collected at the network. Data for training and / or inference may then be collected at the network and at the UE. The UE and the network may perform data validation, for example, to determine whether the data is balanced.

[0100] As another example, AI / ML training and inference can be performed at a network node using network data. In this example, a model designer (e.g., a network entity or network provider) can determine the most efficient feature space (e.g., a set of input parameters) that can use PCA. Data for training and inference can then be collected at the network node, for example, as combined with Figure 4 The network node may perform data validation, for example, to determine whether the data is balanced.

[0101] In another example, training and inference can be performed at a network node using UE data and network data. In this example, a model designer (e.g., a network entity or network provider) can determine the most effective feature space (e.g., a set of input parameters), which can use PCA. In some aspects, a network node can indicate to a UE the data to be collected at the UE. Data for training and / or inference can then be collected at the network and at the UE. The UE and the network can perform data validation, for example, to determine whether the data is balanced.

[0102] In another example, training and inference may be performed at the UE, e.g., as configured by a network node. In this example, a model designer (e.g., a network entity or a network provider) may determine a highest effective feature space (e.g., a set of input parameters), which may use PCA. The network node may configure the UE to perform data training and / or inference based on the AI / ML model. The network node may provide a scheme for data validation to the UE. The UE performs data validation for training and / or inference according to the data validation scheme provided by the network node. In some aspects, the UE may perform training when data validation is successful, and may skip training the AI / ML model if the data meets the validation criteria according to the validation scheme provided by the network node.

[0103] The UE may be configured by the network node with different measurement objects and identifiers for configuring the measurements. Figure 6 An example communication flow 600 between a UE 602 and a base station 604 is illustrated, where the base station 604 configures the UE with one or more measurement objects and reporting configurations at 606. At 608, the UE 602 performs measurements as configured by the base station. At 610, the UE 602 reports the measurements to the base station based on the configuration received at 606. As an example, the base station may indicate to the UE 602 to measure CSI-RS and report CSI to the base station.

[0104] The various aspects presented herein provide data collection, reporting and verification based on AI / ML performed by one or more devices in a wireless network. The various aspects enable timely reporting of input features for inference and online training, and can help minimize the processing delay associated with obtaining input features from raw data (e.g., data observed at UE and / or network nodes). Raw data may also be referred to as source data, atomic data, or raw data. Raw data is data that has not yet been processed for use. Raw data is data collected at a network or UE, and input features of a model can be derived from the data. The various aspects presented herein provide verification of these input features before the input features are provided as input to an inference and / or training engine. The various aspects may also provide verification of such data before the output data is reported to the network or UE.

[0105] A number of parameters may be configured at 606 for measurements performed by the UE 602. Different AI / ML models, AI / ML functions, and / or use cases may use different subsets of the configured measurements, such that at least one subset of the configured measurements may be irrelevant to a specific AI / ML model, AI / ML function, or use case. For different AI / ML models, input features (e.g., input parameters) may involve different conditions and periodicities for reporting. Various aspects presented herein enable different measurement collection and / or reporting configurations to be provided with different conditions for measurement collection and / or reporting for different AI / ML models, AI / ML functions, and / or use cases. Measurements may be processed in different ways for different AI / ML models, AI / ML functions, or use cases. For example, normalization, filtering, and other validation of measurements may be used in conjunction with one or more AI / ML models, AI / ML functions, or use cases, as data validation may affect AI / ML performance.

[0106] Figure 7 Illustrated is a communication flow 700 that includes example aspects of data collection and validation for a machine learning model in which inference, federated learning, and / or training occurs at a UE with network assistance. Figure 7 It is shown that at 708, the model designer 706 may provide inference or training initialization to the UE 704. In some examples, the model designer 706 may be a UE vendor or manufacturer and may configure the UE 704 with inference or training initialization for the AI / ML model. In some aspects, the model designer 706 may provide inference or training initialization for multiple AI / ML models to the UE 704. As illustrated at 710, the UE 704 may request data from a network node 702, such as a base station or a component of a base station. The UE 704 may request data for one or more AI / ML models and may indicate a request per model identifier (ID) associated with the corresponding AI / ML model. In another example, the UE may request data for one or more AI / ML models for different AI / ML use cases. In another example, the UE may request data for one or more AI / ML models for different AI / ML functions. As an example of different ML model IDs, the UE 704 may request data for ML model 1 and ML model 5. As Figure 7 As shown, the request may indicate data for the network node 702 to collect and / or report. The request may indicate a data validation configuration that provides criteria that the data must meet in order for the network node 702 to report the data to the UE 704. Fig.11At 712, the network node 702 may report the requested data or a failure indication to the UE 704. For example, if the network node 702 has collected and verified data in response to the request for ML model 1, the network node may report the verified data to the UE 704 at 712. If the data collected for ML model 5 does not meet the verification criteria for ML model 5, at 712, the network node 702 may send a verification failure indication to the UE 704. Although examples for different ML model IDs are described to illustrate the concepts, these aspects may be similarly applied to different AI / ML use cases or to different AI / ML functions instead of different ML model IDs.

[0107] Figure 8 Illustrated is a communication flow 800 including example aspects of data collection and validation for a machine learning model, where inference, federated learning, and / or training occurs at a UE 804 with network configuration and network assistance. Figure 8 It is shown that at 810, the model designer 806 can register the AI / ML for inference or training. During model registration, the model designer 806 can provide input features (e.g., input parameter sets), data processing modules, and / or data validation information for the registered AI / ML model. For example, the model can be registered and stored at a model repository 808 at the network. The model designer 806 can be a UE vendor or manufacturer, a network vendor, or a third party. As illustrated at 812, the UE 804 can receive inference or training initialization for the AI / ML model, for example, from the network node 802. For online training or federated learning at the UE, a training request for the AI / ML model can be received from the network node 802 according to a request from the model repository 808. The network node 802 can similarly receive inference and / or training initialization from the model repository 808. For example, the network node 802 can be a base station, a component of a base station, or can implement base station functionality. In some aspects, the model repository 808 can store model information for multiple AI / ML models. As illustrated at 814, the network node 802 may send configurations for one or more AI / ML models to the UE 804, and may indicate the configuration of each model identifier (ID) associated with the corresponding AI / ML model. In another example, the network node 802 may configure the UE 804 to collect / process / report data for one or more AI / ML models for different AI / ML use cases. In another example, the network node 802 may configure the UE 804 to collect / process / report data for one or more AI / ML models for different AI / ML functions. As an example of different ML model IDs, the network node 802 may configure the UE 804 to collect / process / report data for ML model 1 and ML model 5. Figure 8As shown, at 814, the configuration may instruct the UE to collect, process, verify data, and / or report data to the network node 802. The configuration at 814 may indicate an AI / ML model configuration for inference, for federated learning, for online learning, etc. The configuration may indicate a data verification configuration that provides criteria that the data must meet in order for the UE 804 to report the data to the network node 802. Aspects of verification are combined Fig.11 At 816, the network node 802 may provide data to the UE 804 and / or may request data from the UE 804. At 818, the UE 804 may report the requested data or a failure indication to the network node 802. For example, if the UE 804 has collected and verified data in response to the request for ML model 1, the UE 804 may report the verified data to the network node 802 at 818. If the data collected for ML model 5 does not meet the verification criteria for ML model 5, at 818, the UE 804 may send a verification failure indication to the network node 802. Although examples for different ML model IDs are described to illustrate the concepts, these aspects may be similarly applied to different AI / ML use cases or to different AI / ML functions instead of different ML model IDs.

[0108] Fig. 9 An example communication flow 900 is illustrated that includes example aspects of data collection and validation for a machine learning model in which inference and / or training occurs at the network with UE assistance. Fig. 9It is shown that at 910, a model designer 906 may register an AI / ML model for inference or training. During the registration process, the model designer 906 may provide input features (e.g., input parameter sets), data processing modules, and / or data validation information for the registered AI / ML model. For example, the model may be registered and stored at a model repository 908 at the network. The model designer 906 may be a UE vendor or manufacturer, a network vendor, or a third party. As illustrated at 912, the network node 902 may receive inference and / or training initialization from the model repository 908. For example, the network node 902 may be a base station, a component of a base station, or may implement base station functionality. In some aspects, the model repository 908 may store model information for multiple AI / ML models. As illustrated at 914, the network node 902 may send configurations for one or more AI / ML models to the UE 904, and may indicate the configuration of each model ID associated with the corresponding AI / ML model. In another example, the network node 902 may configure the UE 904 to collect and report data for one or more AI / ML models for different AI / ML use cases. In another example, the network node 902 may configure the UE 904 to collect and report data for one or more AI / ML models for different AI / ML functions. As an example of different ML model IDs, the network node 902 may configure the UE 904 to collect and report data for ML model 2 and ML model 3. Fig. 9 As shown, at 914, the configuration may instruct the UE to collect, verify data, and / or report data to the network node 902. The configuration at 914 may indicate a data verification configuration that provides criteria that the data must meet in order for the UE 904 to report the data to the network node 902. Fig.11 At 916, the UE 904 may report the requested data or a failure indication to the network node 902. For example, if the UE 904 has collected and verified data in response to the request for ML model 2, the UE 904 may report the verified data to the network node 902 at 916. If the data collected for ML model 3 does not meet the verification criteria for ML model 3, at 916, the UE 904 may send a verification failure indication to the network node 902. Although examples for different ML model IDs are described in order to illustrate the concepts, these aspects may be similarly applied to different AI / ML use cases or to different AI / ML functions instead of different ML model IDs.

[0109] exist Figure 7 , Figure 8 and Fig. 9In each of the , data collection and reporting per model ID, AI / ML use case, or AI / ML function helps reduce inference latency and online training latency by helping prepare data during the collection phase. Validation of inference and training data helps improve AI / ML performance by selecting data that adheres to a set of characteristics and / or statistics to ensure that the data is real data or accurate data for training and / or inference purposes. Validation helps avoid overfitting AI / ML models to inaccurate data.

[0110] Fig.10 An example communication flow 1000 is illustrated between a first device 1004 and a second device 1002 in a wireless network, including data requests and reports based on an AI / ML model ID. In some aspects, the first device 1004 may be a UE, and the second device 1002 may be a network node or a second UE. In other aspects, the first device 1004 may be a network node, and the second device 1002 may be a UE or another network node. Fig.10 It is illustrated that model registration 1008 may occur, wherein a model designer 1006 registers input parameters of an AI / ML model, the input parameters including raw data to be collected for the AI / ML model and from which the input parameters are obtained. The model registration may also include an indication of one or more data processing modules to be used to obtain the input parameters of the AI / ML model. The model registration may also include an indication of a list of one or more input parameters (e.g., feature spaces) of the AI / ML model. In some aspects, the model designer 1006 may use PCA to determine the most effective feature space (e.g., set of input parameters) for AI / ML model training and inference, for example, as combined with Figure 4 The model designer 1006 may be a UE, a UE vendor or manufacturer, a network node, a network vendor, a third party, or the like.

[0111] As illustrated at 1010, the second device 1002 can send a data report configuration to the first device 1004. The second device 1002 can be a base station, a component of a base station, or can implement base station functionality, or can be a UE or UE vendor, for example, based on a specific training and inference scenario. Figures 7 to 9Various examples of training and inference scenarios are described. The second device 1002 may send a data request per model ID of multiple AI / ML models, AI / ML use cases, or AI / ML functions, the request / configuration indicating the AI / ML model or use case input parameters, raw data / measurements for the first device 1004 to calculate the input parameters, raw data for obtaining model input parameters, data processing modules, or data reporting methods for the corresponding AI / ML model. As an example, the network node (e.g., as the second device 1002) may configure these parameters per model ID, per use case, or per AI / ML function, for example, in one or more RRC messages to the UE (e.g., as the first device 1004). In some aspects, the UE (e.g., as the second device 1002) may request data from the network node (e.g., as the first device 1004) in a message including UE assistance information (UAI).

[0112] In some aspects, at 1012, the second device 1002 may activate or deactivate the AI / ML model, use case of the function configured at 1010. As an example, the network node (e.g., as the second device 1002) may send a MAC-CE or DCI at 1012 to the UE (e.g., as the first device 1004), which activates or deactivates data reporting for one or more configured AI / ML model IDs, use cases, or functions at 1010. As another example, the UE (e.g., as the second device 1002) may send a MAC-CE or DCI at 1012 to the network node or another UE (e.g., as the first device 1004), which activates or deactivates data reporting for one or more configured AI / ML model IDs, use cases, or functions at 1010.

[0113] At 1014, the first device 1004 reports data according to the configured (e.g., and activated) AI / ML model ID, use case, or function. The first device 1004 can send data reports to the second device 1002 in any combination of RRC messages, MAC-CE, UCI, and / or DCI.

[0114] Fig.11 An example communication flow 1100 is illustrated between a first device 1104 and a second device 1102 in a wireless network, including data requests and reports based on an AI / ML model ID. In some aspects, the first device 1104 may be a UE, and the second device 1102 may be a network node or a second UE. In other aspects, the first device 1104 may be a network node, and the second device 1102 may be a UE or another network node. In some aspects, Fig.11 The communication flow in Fig.10 The communication flow in the implementation, for example, Figures 7 to 9 as shown in the various examples in .

[0115] At 1106, the second device 1002 may send a data validation configuration to the first device 1004. For example, the data validation configuration may be provided to the first device 1004 in association with a data reporting configuration (e.g., such as 1010). In the data validation configuration, the configuring or requesting device (whether a UE or a network node) may provide rules or standards for data validation. As an example, the second device 1002 may indicate a set of data statistics and / or attributes that the AI / ML inference and training data will follow for a particular model ID, use case, or AI / ML function. For example, errors in new data observed or measured by the first device 1004 may be checked by comparing the newly observed data at 1107 with the set of predefined data attributes, statistics, standards, or rules in the data validation configuration for the corresponding model ID, use case, or function. The data validation configuration may be provided to the second device 1002 per model ID / use case / or AI / ML function, and may be provided together with the data collection and reporting configuration for the corresponding model ID, use case, or function, for example, as combined Figure 8 and Fig. 9 Similarly, data validation configuration for a model ID, use case or function may be provided from the UE to the network in a UAI or other message, such as in conjunction with Figure 7 as described.

[0116] As illustrated at 1108, if the observed data does not meet the configured validation criteria, the device 1104 may provide a data validation failure indication. Otherwise, the device may report the data, e.g., Figures 7 to 10 As illustrated in any of . For inference or training (e.g., online training) including AI / ML models, data validation failures may be reported by device 1104 in a timely manner so that other devices 1102 may take appropriate actions. For example, if the observed data is different from what is expected for the ML model, use case, or function, the second device 1002 may fall back to a different procedure. In some aspects, the device may fall back to a non-AI / ML model procedure that may be referred to as a legacy procedure. In some aspects, device 1104 may further indicate information about data validation to device 1102 (e.g., the extent to which newly observed inference and / or training data is closely related to a defined set of statistics or attributes). Device 1104 may provide validation failure information for each inference / online training opportunity including a failure indication. In some aspects, device 1104 may provide information about data validation in batches (e.g., for multiple inference or training opportunities) and failure indications.

[0117] In some aspects, if the data indicates a failure, the network node and / or UE may fall back to a different procedure, such as a non-AI / ML measurement and reporting procedure. In some aspects, the verification failure indication at 1108 may indicate that the device 1104 is to switch to a different reporting procedure and / or may indicate that the device 1102 uses a non-AI / ML procedure to wirelessly communicate with the device 1104. In some aspects, the reported data may be used for joint learning. In such aspects, the verification failure indication 1108 may indicate to the device 1102 that the device 1104 did not obtain the weights for a particular period.

[0118] As an example, the network may request or indicate to the UE the data to be collected, processed and / or reported per model ID, per use case, or per AI / ML function. As an example, for AI / ML training, federated learning, or inference at a network node, the network node may indicate the data processing module / technique to be used. Fig.12 A data processing diagram 1200 is illustrated, showing that raw data or multiple observations from multiple sources (e.g., at 1202a, 1202b, 1202c, 1202d) can be processed at 1206 using one or more of multiple data processing modules 1204a or 1204b to provide prepared data 1208. The data processing modules 1204a and 1204b may include filtering, reconstruction, transformation (including feature scaling, normalization, converting qualitative variables into quantitative variables, and other examples) and other modules / techniques to be used when a UE or network node collects measurements for training or inference for a given model ID, use case, or AI / ML function. At 1210, the prepared data can then be verified by comparing the processed data to a validation criterion (e.g., statistics, attributes, rules, etc.). If the processed data meets the validation criterion, then at 1212, the data can be reported, used for inference, and / or used for training. Figure 4 Various aspects of training and inference are illustrated, while Figures 7 to 10 Various aspects of the report are illustrated. Fig.11 Various aspects of providing a validation failure indication and / or validation failure information when the processed data does not meet the validation criteria are illustrated. The network may provide a configuration for training and inference data validation, for example, in conjunction with Figures 7 to 11 For federated learning, the network may, for example, configure data processing and validation before the data is used for federated learning.

[0119] Similarly, the UE may request or indicate to a network node or another UE the data to be collected, processed, and / or reported per model ID, per use case, or per AI / ML function. In some aspects, the UE may request training or inference data per model ID, use case, or AI / ML function from the network using UAI. In some aspects, the UE may indicate that the feature space / parameters of the network node or other UE are to be reported to the requesting UE. In some aspects, the UE may indicate a set of raw data from which the input features / parameters of the AI / ML model will be obtained. In some aspects, the UE may indicate the data and processing modules / techniques (e.g., 1204a or 1204b) to be used for data collection and / or reporting. In some aspects, the UE may indicate the configuration for training and / or inference data validation at 1210 to be performed before reporting data to the UE.

[0120] As combined Figure 7 As described, the UE may request network assistance data per model ID / use case / ML function for inference, federated learning, or training (online and / or offline training) at the UE, for example, by providing a list of input features (e.g., input parameters) of the model ID / use case / function, data processing modules, and / or verification methods for the model ID / use case / function.

[0121] For inference, joint learning or training (online and offline training) at the network, for example, Figure 8 or Fig. 9 In the example above, the network may provide data collection, reporting and verification configuration per Model ID / Use Case / ML Function. In response, the UE may collect and report the requested measurements per Model ID / Use Case / ML Function. If a verification failure occurs, for example, in conjunction with Fig.11 As described above, the UE may indicate to the network that the verification has failed.

[0122] From the perspective of the network nodes, the network can provide configurations for data collection, reporting and verification per model ID / use case / ML function, such as in combination with Figures 8 to 10 The network may provide a data processing model to be used to process the data before reporting to the network. The network may provide a data validation method, standard, or statistics to be followed by the collected inference or training data to be considered legal for inference or training of the AI / ML model. Following a data request from the UE (including a data validation method), the network may report the data to the UE after validation, for example, as combined with Figure 7 If the requested data is not available or does not satisfy the verification scheme, the network may indicate to the UE, for example, as described in conjunction with Fig.11 as described.

[0123] As combined Figures 7 to 10As described in any of , a model designer may register a model and, when registering, may provide a list of input features, data processing modules used to obtain the input features, and / or data validation schemes (such as baseline statistics to which inference or training data will follow). The information provided by the model designer may be used by the network node and / or UE to configure, request, collect, process, validate, and / or report data in conjunction with the AI / ML model, use case, and / or function.

[0124] Fig.13 An example communication flow 1300 between a UE 1302 and a base station 1304 is illustrated, illustrating the concept of configuring multiple AI / ML models and one or more of the configured models being activated for data collection / processing / reporting. Although the aspects are described with respect to a UE and a network, the aspects performed by the UE 1302 may be performed by a network node and / or the aspects performed by the base station 1304 may be performed by the UE, for example, depending on the device requesting the collection and reporting of data, for example, as in conjunction with Figures 7 to 10 As described in the different AI / ML scenarios.

[0125] At 1306, UE 1302 may receive configurations for multiple AI / ML models (e.g., AI / ML Model 1, AI / ML Model 2, AI / ML Model 3, and AI / ML Model 4). UE 1302 may receive these configurations individually or together. At 1308, base station 1304 may send a MAC-CE or DCI to activate AI / ML Model 1. At 1310, UE 1302 may collect data for training, inference, and / or reporting based on AI / ML Model 1. At 1306, UE may also use criteria from the configuration to validate the data, such as in combination with Fig.11 If the data meets the validation criteria for AI / ML model 1, then at 1312, the UE may report the data. If the data does not meet the validation criteria for AI / ML model 1, then at 1314, the UE may provide a failure indication.

[0126] At 1316, the base station 1304 may send a MAC-CE or DCI activating AI / ML Model 2. In some aspects, the base station may deactivate AI / ML Model 1, and the UE may stop training / collecting / validating / reporting based on Model 1. In some aspects, the activation of Model 2 may indicate the deactivation of Model 1. In other aspects, the UE may continue to collect and report data based on Model 1, and may also collect and report data based on Model 2 in response to the activation. At 1318, the UE 1302 may collect data for training, inference, and / or reporting based on AI / ML Model 2. At 1306, the UE may also validate the data using criteria from the configuration, such as in conjunction with Fig.11If the data meets the validation criteria for AI / ML model 2, then at 1320, the UE may report the data. If the data does not meet the validation criteria for AI / ML model 1, then at 1322, the UE may provide a failure indication.

[0127] Fig.14A 1400 is a flow chart of a method of wireless communication. In some aspects, the method may be performed by a UE (e.g., UE 104, 350, 602, 704, 804, 904, 1302; first device 1004, 1104; apparatus 1604). In some aspects, the method may be performed by a network node (e.g., base station 102, 310, 1304; network node 702, 802, 902; first device 1004, 1104; network entity 1702) such as a base station, a component of a base station, or a device implementing base station functionality.

[0128] At 1406, the device processes the information using machine learning associated with the model ID, machine learning function, or machine learning use case. The processing may be performed, for example, by machine learning component 198. The processing may include combining Figure 4 , Figure 5 or Figures 7 to 13 Any of the described aspects.

[0129] At 1410, the device reports data via wireless communication based on a configuration associated with a model ID, a machine learning function, or a machine learning use case. The reporting may be performed, for example, by machine learning component 198. The reporting may include, for example, combining Figures 7 to 10 , Fig.12 or Fig.13 The configuration associated with a model ID, a machine learning function, or a machine learning use case may indicate one or more of the following: a data reporting method, at least one input parameter for machine learning, unprocessed data for obtaining a model input parameter, at least one measurement for obtaining a model input parameter, at least one data processing module, timing for data reporting, conditions for data reporting, or periodicity for data reporting.

[0130] Fig. 14B 1450 is a flow chart of a method of wireless communication. In some aspects, the method may be performed by a UE (e.g., UE 104, 350, 602, 704, 804, 904, 1302; first device 1004, 1104; apparatus 1604). In some aspects, the method may be performed by a network node (e.g., base station 102, 310, 1304; network node 702, 802, 902; first device 1004, 1104; network entity 1702) such as a base station, a component of a base station, or a device implementing base station functionality.

[0131] At 1406, the device processes the information using machine learning associated with the model ID, machine learning function, or machine learning use case. The processing may be performed, for example, by machine learning component 198. The processing may include combining Figure 4 , Figure 5 or Figures 7 to 13 Any of the described aspects.

[0132] At 1410, the device reports data via wireless communication based on a configuration associated with a model ID, a machine learning function, or a machine learning use case. The reporting may be performed, for example, by machine learning component 198. The reporting may include, for example, combining Figures 7 to 10 , Fig.12 or Fig.13 The data may be reported in at least one of an RRC message, a MAC-CE, a UCI, or a DCI. In some aspects, at 1410, the device may report different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case.

[0133] As illustrated at 1402, the device may receive a configuration identifying a model ID, a machine learning function, or a machine learning use case, wherein reporting the data includes sending the data based on a condition, timing, or periodicity indicated in the configuration for the model ID, machine learning function, or machine learning use case. The receiving may be performed, for example, by the machine learning component 198. The configuration associated with the model ID, machine learning function, or machine learning use case may indicate one or more of the following: a data reporting method, at least one input parameter for machine learning, unprocessed data for obtaining a model input parameter, at least one measurement for obtaining a model input parameter, at least one data processing module, timing for data reporting, conditions for data reporting, or periodicity for data reporting. In some aspects, the method may be performed at a UE, wherein the configuration is received from a network node, and the data is reported to the network node. In some aspects, the method may be performed at a network node, the configuration may be received from a UE, and the data is reported to the UE. In some aspects, the method may be performed at a first network node, the configuration may be received from a second network node, and the data is reported to the second network node. In some aspects, the method may be performed at a first UE, the configuration may be received from a second UE, and the data may be reported to the second UE.

[0134] As illustrated at 1404, the device can receive an activation for data reporting for a model ID, machine learning function, or machine learning use case, wherein the data is reported in response to the activation. The receiving can be performed, for example, by the machine learning component 198. For example, the activation can be performed in MAC-CE or DCI. Fig.13 To describe the example activation.

[0135] As illustrated at 1412, the device can receive a deactivation of data reporting for a model ID, machine learning function, or machine learning use case. For example, the deactivation can be performed in MAC-CE or DCI. The receiving can be performed, for example, by the machine learning component 198. Fig.13 To describe an example deactivation. In response to the deactivation, the device may stop data reporting for the model ID, machine learning function, or machine learning use case at 1414 in response to the deactivation. Fig.13 An example of a device that stops data reporting in response to deactivation is illustrated. The stop can be performed, for example, by the machine learning component 198.

[0136] The configuration associated with the model ID, machine learning function, or machine learning use case may include a data validation configuration. As illustrated at 1408, the device may validate the data before reporting based on the criteria of the data validation configuration for the model ID, machine learning function, or machine learning use case. The validation may be performed, for example, by the machine learning component 198. For example, in conjunction with Figures 7 to 9 , Fig.11 , Fig.12 and Fig.13 to describe example aspects of validation. The data validation configuration may include one or more of the following: at least one rule for data validation associated with a model ID, a machine learning function, or a machine learning use case, at least one data statistic associated with a model ID, a machine learning function, or a machine learning use case, or at least one data attribute associated with a model ID, a machine learning function, or a machine learning use case.

[0137] In some aspects, validating the data at 1408 may include identifying at least one of the machine learning-based inference or training outputs that does not satisfy a validation criterion of a data validation configuration associated with the model ID, machine learning function, or machine learning use case. In response, the device may indicate a data validation failure according to the configuration for the model ID, machine learning function, or machine learning use case, e.g., Fig.11 or Fig.13 In some aspects, indicating a data validation failure may also indicate a transition to a procedure without machine learning.

[0138] Fig.15A1500 is a flow chart of a method of wireless communication. In some aspects, the method may be performed by a UE (e.g., UE 104, 350, 602, 704, 804, 904, 1302; second device 1002, 1102; apparatus 1604). In some aspects, the method may be performed by a network node (e.g., base station 102, 310, 1304; network node 702, 802, 902; second device 1002, 1102; network entity 1702) such as a base station, a component of a base station, or a device implementing base station functionality.

[0139] At 1502, the device provides a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case. The provision of the configuration can be performed, for example, by a machine learning configuration component 199. The configuration associated with a model ID, a machine learning function, or a machine learning use case can indicate one or more of the following: a data reporting method, at least one input parameter for machine learning, unprocessed data for obtaining a model input parameter, at least one measurement for obtaining a model input parameter, at least one data processing module, timing for data reporting, conditions for data reporting, or periodicity for data reporting. Figures 7 to 13 Various examples of configurations provided for collecting, processing, validating, and / or reporting data using AI / ML models are described.

[0140] At 1506, the device receives a data report based on the configuration associated with a model ID, a machine learning function, or a machine learning use case. The receiving may be performed, for example, by a machine learning configuration component 199. In some aspects, the method may be performed at a UE, wherein the configuration is provided to a network node and the data is received from the network node. In some aspects, the method may be performed at a network node, the configuration may be provided to a UE, and the data may be reported by the UE. In some aspects, the method may be performed at a first network node, and the configuration may be provided to a second network node, which reports the data to the second network node. In some aspects, the method may be performed at a first UE, and the configuration may be provided to a second UE, which reports the data to the first UE. The report may include, for example, in conjunction with Figures 7 to 10 , Fig.12 or Fig.13 Any of the aspects described in any one of the aspects. The data may be reported in at least one of an RRC message, a MAC-CE, a UCI or a DCI.

[0141] In some aspects, the method may be performed at a UE, and wherein the configuration is provided to a network node, and the data is received from the network node. In some aspects, the method may be performed at a network node, and wherein the configuration is provided to a UE, and the data is received from the UE. In some aspects, the method may be performed at a first network node, and wherein the configuration is provided to a second network node, and wherein the data is received from the second network node. In some aspects, the method may be performed at a first UE, and wherein the configuration is provided to a second UE, and wherein the data is received from the second UE.

[0142] Fig. 15B 1550 is a flow chart of a method of wireless communication. In some aspects, the method may be performed by a UE (e.g., UE 104, 350, 602, 704, 804, 904, 1302; second device 1002, 1102; apparatus 1604). In some aspects, the method may be performed by a network node (e.g., base station 102, 310, 1304; network node 702, 802, 902; second device 1002, 1102; network entity 1702) such as a base station, a component of a base station, or a device implementing base station functionality.

[0143] At 1502, the device provides a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case. The provision of the configuration can be performed, for example, by a machine learning configuration component 199. The configuration associated with a model ID, a machine learning function, or a machine learning use case can indicate one or more of the following: a data reporting method, at least one input parameter for machine learning, unprocessed data for obtaining a model input parameter, at least one measurement for obtaining a model input parameter, at least one data processing module, timing for data reporting, conditions for data reporting, or periodicity for data reporting. Figures 7 to 13 Various examples of configurations provided for collecting, processing, validating, and / or reporting data using AI / ML models are described.

[0144] At 1506, the device receives a data report based on the configuration associated with a model ID, a machine learning function, or a machine learning use case. The receiving may be performed, for example, by a machine learning configuration component 199. In some aspects, the method may be performed at a UE, wherein the configuration is provided to a network node and the data is received from the network node. In some aspects, the method may be performed at a network node, the configuration may be provided to a UE, and the data may be reported by the UE. In some aspects, the method may be performed at a first network node, and the configuration may be provided to a second network node, which reports the data to the second network node. In some aspects, the method may be performed at a first UE, and the configuration may be provided to a second UE, which reports the data to the first UE. The report may include, for example, in conjunction with Figures 7 to 10 , Fig.12 or Fig.13 The data may be reported in at least one of an RRC message, a MAC-CE, a UCI, or a DCI. In some aspects, the device may receive reports of different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case. The receiving may be performed, for example, by a machine learning configuration component 199. For example, the reporting of data may include combining Figures 7 to 10 , Fig.12 or Fig.13 Any of the aspects described in any one of the aspects. The data may be received in at least one of an RRC message, a MAC-CE, or a DCI.

[0145] In some aspects, the method may be performed at a UE, and wherein the configuration is provided to a network node, and the data is received from the network node. In some aspects, the method may be performed at a network node, and wherein the configuration is provided to a UE, and the data is received from the UE. In some aspects, the method may be performed at a first network node, and wherein the configuration is provided to a second network node, and wherein the data is received from the second network node. In some aspects, the method may be performed at a first UE, and wherein the configuration is provided to a second UE, and wherein the data is received from the second UE.

[0146] As illustrated at 1504, the device can also provide activation of data reporting for a model ID, machine learning function, or machine learning use case, wherein the data is received in response to the activation. The provision of the activation can be performed, for example, by the machine learning configuration component 199. Fig.13 To describe the example activation.

[0147] As illustrated at 1510, the device can also provide for deactivation of data reporting for a model ID, machine learning function, or machine learning use case. The provision of the deactivation can be performed, for example, by the machine learning configuration component 199. Fig.13 To describe example deactivation.

[0148] The configuration associated with the model ID may include a data validation configuration, which includes one or more of the following: at least one rule for data validation associated with the model ID, machine learning function, or machine learning use case, at least one data statistic associated with the model ID, machine learning function, or machine learning use case, or at least one data attribute associated with the model ID, machine learning function, or machine learning use case. As illustrated at 1508, the device may receive a data validation failure indication, which indicates that at least one of the validation criteria of the data validation configuration associated with the model ID, machine learning function, or machine learning use case is not met in the machine learning-based inference or training output. The receiving may be performed, for example, by the machine learning configuration component 199. For example, in conjunction with Fig.11 and Fig.13 1 to describe example aspects of data validation failure. Data validation failure may also indicate a transition to a procedure without machine learning.

[0149] Fig.161600 is a diagram illustrating an example of a hardware implementation for an apparatus 1604. The apparatus 1604 may be a UE, a component of a UE, or may implement UE functionality. In some aspects, the apparatus 1604 may include a cellular baseband processor 1624 (also referred to as a modem) coupled to one or more transceivers 1622 (e.g., a cellular RF transceiver). The cellular baseband processor 1624 may include on-chip memory 1624'. In some aspects, the apparatus 1604 may also include one or more subscriber identity module (SIM) cards 1620 and an application processor 1606 coupled to a secure digital (SD) card 1608 and a screen 1610. The application processor 1606 may include on-chip memory 1606'. In some aspects, the device 1604 may also include a Bluetooth module 1612, a WLAN module 1614, an SPS module 1616 (e.g., a GNSS module), one or more sensor modules 1618 (e.g., an atmospheric pressure sensor / altimeter; a motion sensor such as an inertial management unit (IMU), a gyroscope, and / or an accelerometer; light detection and ranging (LIDAR), radio-aided detection and ranging (RADAR), sound navigation and ranging (SONAR), a magnetometer, audio, and / or other technologies for positioning), an additional memory module 1626, a power source 1630, and / or a camera 1632. The Bluetooth module 1612, the WLAN module 1614, and the SPS module 1616 may include an on-chip transceiver (TRX) (or, in some cases, only a receiver (RX)). The Bluetooth module 1612, the WLAN module 1614, and the SPS module 1616 may include their own dedicated antennas and / or communicate using an antenna 1680. The cellular baseband processor 1624 communicates with the UE 104 and / or with the RU associated with the network entity 1602 through the transceiver 1622 via one or more antennas 1680. The cellular baseband processor 1624 and the application processor 1606 may each include a computer-readable medium / memory 1624', 1606', respectively. The additional memory module 1626 may also be considered as a computer-readable medium / memory. Each computer-readable medium / memory 1624', 1606', 1626 may be non-transitory. The cellular baseband processor 1624 and the application processor 1606 are each responsible for general processing, including executing software stored on a computer-readable medium / memory. The software enables the cellular baseband processor 1624 / application processor 1606 to perform the various functions described above when executed by the cellular baseband processor 1624 / application processor 1606. The computer-readable medium / memory can also be used to store data manipulated by the cellular baseband processor 1624 / application processor 1606 when executing the software.The cellular baseband processor 1624 / application processor 1606 may be a component of the UE 350 and may include the memory 360 and / or at least one of the TX processor 368, the RX processor 356, and the controller / processor 359. In one configuration, the device 1604 may be a processor chip (modem and / or application) and include only the cellular baseband processor 1624 and / or the application processor 1606, while in another configuration, the device 1604 may be the entire UE (e.g., see. Figure 3 350) and includes additional modules of device 1604.

[0150] As discussed herein, the machine learning component 198 can be configured to process information using machine learning associated with a model ID, a machine learning function, or a machine learning use case; and report data via wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case. The machine learning component 198 can also be configured to perform a combination of Fig.14A and / or Fig. 14B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The first device 1104, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 UE in any one of Figure 7 As described above, the machine learning configuration component 199 can be configured to provide a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and receive a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case. The machine learning configuration component 199 can also be configured to perform a combination of Fig.15A and / or Fig. 15B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The second device 1102, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 A network node in any of Figure 7Any of the aspects performed by the UE in . The machine learning component 198 and / or the machine learning configuration component 199 may be within the cellular baseband processor 1624, within the application processor 1606, or within both the cellular baseband processor 1624 and the application processor 1606. The machine learning component 198 and / or the machine learning configuration component may be one or more hardware components that are specifically configured to perform the process / algorithm, implemented by one or more processors configured to perform the process / algorithm, stored in a computer-readable medium for implementation by one or more processors, or some combination thereof. As shown, the device 1604 may include a variety of components configured for various functions. In one configuration, the device 1604 (and more specifically, the cellular baseband processor 1624 and / or the application processor 1606) may include the device 1604, which may include a component for processing information using machine learning associated with a model ID, a machine learning function, or a machine learning use case; and a component for reporting data via wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case. The device 1604 may also include a component for receiving a configuration identifying a model ID, a machine learning function, or a machine learning use case, wherein reporting the data includes sending the data based on a condition, timing, or periodicity indicated in the configuration for the model ID, machine learning function, or machine learning use case. The device 1604 may also include a component for reporting different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case. The device 1604 may also include a component for receiving activation of data reporting for a model ID, a machine learning function, or a machine learning use case, wherein the data is reported in response to the activation. The device 1604 may also include a component for receiving deactivation of data reporting for a model ID, a machine learning function, or a machine learning use case; and a component for stopping data reporting for a model ID, a machine learning function, or a machine learning use case in response to the deactivation. The configuration associated with the model ID, the machine learning function, or the machine learning use case may include a data validation configuration, and the device 1604 may also include a component for validating the data before reporting based on the criteria of the data validation configuration for the model ID, the machine learning function, or the machine learning use case. The device 1604 may also include means for identifying at least one of the machine learning-based inference or training output that does not satisfy the validation criteria of the data validation configuration associated with the model ID, machine learning function, or machine learning use case; and means for indicating a data validation failure according to the configuration for the model ID, machine learning function, or machine learning use case. The device 1604 may include means for performing a combination Fig.14A and / or Fig. 14B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11The first device 1104, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 UE in any one of Figure 7 Components for any of the aspects performed by the network node in . In some aspects, device 1604 may include a component for providing a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and a component for receiving a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case. Device 1604 may also include a component for receiving a report of different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case. Device 1604 may also include a component for providing activation of a data report for a model ID, a machine learning function, or a machine learning use case, wherein the data is received in response to the activation. Device 1604 may also include a component for providing deactivation of a data report for a model ID, a machine learning function, or a machine learning use case. Device 1604 may also include a component for receiving a data validation failure, which indicates that at least one of the validation criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case is not satisfied in the machine learning-based inference or training output. Device 1604 may also include a component for performing a combination Fig.15A and / or Fig. 15B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The second device 1102, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 A network node in any of Figure 7 The component may be a machine learning component 198 and / or a machine learning configuration component 199 of the device 1604 configured to perform the functions recorded by the component. As described above, the device 1604 may include a TX processor 368, an RX processor 356, and a controller / processor 359. Therefore, in one configuration, the component may be a TX processor 368, an RX processor 356, and / or a controller / processor 359 configured to perform the functions recorded by the component.

[0151] Fig.171700 is a diagram illustrating an example of a hardware implementation for a network entity 1702. The network entity 1702 may be a BS, a component of a BS, or may implement BS functionality. The network entity 1702 may include at least one of a CU 1710, a DU 1730, or a RU 1740. For example, depending on the layer functionality processed by the machine learning configuration component 199 and / or the machine learning component 198, the network entity 1702 may include a CU 1710; both a CU 1710 and a DU 1730; each of the CU 1710, the DU 1730, and the RU 1740; the DU 1730; both a DU 1730 and a RU 1740; or a RU 1740. The CU 1710 may include a CU processor 1712. The CU processor 1712 may include an on-chip memory 1712'. In some aspects, the CU 1710 may also include an additional memory module 1714 and a communication interface 1718. CU 1710 communicates with DU 1730 via a midhaul link (such as an F1 interface). DU 1730 may include a DU processor 1732. DU processor 1732 may include on-chip memory 1732'. In some aspects, DU 1730 may also include additional memory modules 1734 and communication interfaces 1738. DU 1730 communicates with RU 1740 via a fronthaul link. RU 1740 may include a RU processor 1742. RU processor 1742 may include on-chip memory 1742'. In some aspects, RU 1740 may also include additional memory modules 1744, one or more transceivers 1746, antenna 1780, and communication interface 1748. RU 1740 communicates with UE 104. On-chip memory 1712', 1732', 1742' and additional memory modules 1714, 1734, 1744 may each be considered a computer-readable medium / memory. Each computer readable medium / memory may be non-transitory. Each of processors 1712, 1732, 1742 is responsible for general processing, including executing software stored on the computer readable medium / memory. The software, when executed by the corresponding processor, causes the processor to perform the various functions described above. The computer readable medium / memory may also be used to store data manipulated by the processor when executing the software.

[0152] As discussed herein, the machine learning component 198 can be configured to process information using machine learning associated with a model ID, a machine learning function, or a machine learning use case; and report data via wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case. The machine learning component 198 can also be configured to perform a combination of Fig.14A and / or Fig. 14B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The first device 1104, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 UE in any one of Figure 7 As described above, the machine learning configuration component 199 can be configured to provide a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and receive a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case. The machine learning configuration component 199 can also be configured to perform a combination of Fig.15A and / or Fig. 15B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The second device 1102, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 A network node in any of Figure 7Any of the aspects performed by the UE in the . The machine learning configuration component 199 and / or the machine learning component 198 may be within one or more processors of one or more of the CU 1710, the DU 1730, and the RU 1740. The machine learning configuration component 199 and / or the machine learning component 198 may be one or more hardware components that are specifically configured to perform the process / algorithm, implemented by one or more processors configured to perform the process / algorithm, stored in a computer-readable medium so as to be implemented by one or more processors, or some combination thereof. The network entity 1702 may include a variety of components configured for various functions. In one configuration, the network entity 1702 may include a component for processing information using machine learning associated with a model ID, a machine learning function, or a machine learning use case; and a component for reporting data via wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case. The network entity 1702 may also include a component for receiving a configuration identifying a model ID, a machine learning function, or a machine learning use case, wherein reporting the data includes sending the data based on a condition, timing, or periodicity indicated in the configuration for the model ID, the machine learning function, or the machine learning use case. The network entity 1702 may also include a component for reporting different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case. The network entity 1702 may also include a component for receiving activation of data reporting for a model ID, a machine learning function, or a machine learning use case, wherein the data is reported in response to the activation. The network entity 1702 may also include a component for receiving deactivation of data reporting for a model ID, a machine learning function, or a machine learning use case; and a component for stopping data reporting for a model ID, a machine learning function, or a machine learning use case in response to the deactivation. The configuration associated with the model ID, the machine learning function, or the machine learning use case may include a data validation configuration, and the network entity 1702 may also include a component for validating data before reporting based on the criteria of the data validation configuration for the model ID, the machine learning function, or the machine learning use case. The network entity 1702 may also include a component for identifying at least one of the validation criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case in the inference or training output based on machine learning; and a component for indicating data validation failure according to the configuration for the model ID, the machine learning function, or the machine learning use case. The network entity 1702 may include a Fig.14A and / or Fig. 14B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The first device 1104, Figure 6 , Figure 8, Fig. 9 , Fig.10 or Fig.13 UE in any one of Figure 7 Components of any of the aspects performed by the network node in . In some aspects, the network entity 1702 may include a component for providing a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and a component for receiving a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case. The network entity 1702 may also include a component for receiving a report of different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case. The network entity 1702 may also include a component for providing activation of a data report for a model ID, a machine learning function, or a machine learning use case, wherein the data is received in response to the activation. The network entity 1702 may also include a component for providing deactivation of a data report for a model ID, a machine learning function, or a machine learning use case. The network entity 1702 may also include a component for receiving a data validation failure, which indicates that at least one of the validation criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case is not satisfied in the machine learning-based inference or training output. The network entity may include a component for performing a combination Fig.15A and / or Fig. 15B Flowchart in Figure 4 The aspects described in the algorithm in, and / or by Fig.11 The second device 1102, Figure 6 , Figure 8 , Fig. 9 , Fig.10 or Fig.13 A network node in any of Figure 7 The component may be a machine learning configuration component 199 and / or a machine learning component 198 of the network entity 1702 configured to perform the functions recorded by the component. As described above, the network entity 1702 may include a TX processor 316, an RX processor 370, and a controller / processor 375. Therefore, in one configuration, the component may be a TX processor 316, an RX processor 370, and / or a controller / processor 375 configured to perform the functions recorded by the component.

[0153] Fig.18 1800 is a flow chart of a method of registering an AI / ML model in accordance with various aspects presented herein. At 1802, a model designer may register a machine learning model for collecting and reporting data based on wireless communications. Figures 7 to 10 Illustrate various aspects of a model designer registering an AI / ML model.

[0154] At 1804, the model designer may provide at least one of input features for the machine learning model, a data processing module for obtaining the input features for the machine learning model, or a data validation scheme for the machine learning model. The data validation scheme may include one or more of the following: at least one rule for data validation associated with the machine learning model, at least one data statistic associated with the training data for the machine learning model, at least one data statistic associated with the inferred data for the machine learning model, at least one data attribute associated with the training data for the machine learning model, or at least one data attribute associated with the inferred data for the machine learning model.

[0155] It should be understood that the specific order or hierarchy of the blocks in the disclosed process / flowchart is merely an illustration of the exemplary method. It should be understood that the specific order or hierarchy of the blocks in the process / flowchart may be rearranged based on design preferences. Further, some blocks may be combined or omitted. The attached method claims provide the elements of each block in a sample order, but are not limited to the specific order or hierarchy provided.

[0156] Fig.19 1900 is a diagram illustrating an example of a hardware implementation for a model registration entity 1902. The model registration entity 1902 may include an on-chip memory 1912'. In some aspects, the model registration entity 1902 may also include an additional memory module 1914 and a communication interface 1918. The on-chip memory 1912' and the additional memory module 1914 may each be considered a computer-readable medium / memory. Each computer-readable medium / memory may be non-transitory. The processor 1912 may be responsible for general processing, including executing software stored on the computer-readable medium / memory. The software, when executed by the corresponding processor, causes the processor to perform the various functions described above. The computer-readable medium / memory may also be used to store data manipulated by the processor when executing the software.

[0157] As described above, the registration component 1920 may be configured to register a machine learning model for collecting and reporting data based on wireless communication; and provide the network entity 1904 model with at least one of input features for the machine learning model, a data processing module for obtaining input features for the machine learning model, or a data validation scheme for the machine learning model. The registration component 1920 may be configured to perform Fig.18and / or any aspects of the aspects thereof. The registration component 1920 may be within one or more processors of the model registration entity 1902. The registration component 1920 may be one or more hardware components specifically configured to execute the process / algorithm, implemented by one or more processors configured to execute the process / algorithm, stored in a computer-readable medium for implementation by one or more processors, or some combination thereof. The model registration entity 1902 may include a variety of components configured for various functions. In one configuration, the model registration entity 1902 may include a component for registering a machine learning model for collecting and reporting data based on wireless communications; and a component for providing at least one of input features for the machine learning model, a data processing module for obtaining input features for the machine learning model, or a data validation scheme for the machine learning model. The model registration entity 1902 may include a component for performing a combination of Fig.18 The various aspects described by the flowchart in, and / or by Figures 7 to 10 The component may be a registration component 1920 of the model registration entity 1902 configured to perform the functions described by the component.

[0158] The foregoing description is provided to enable any person skilled in the art to practice the various aspects described herein. 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. Therefore, the claims are not limited to the various aspects described herein, but should be given the full scope consistent with the language claims. Unless otherwise specified, reference to an element in the singular form does not mean "one and only one", but "one or more". Terms such as "if", "when ......" and "while" do not mean a direct temporal relationship or reaction. That is, these phrases (such as "when ......") do not mean an immediate action in response to the occurrence of an action or during the occurrence of an action, but simply imply that if the conditions are met, then the action will occur, but no specific or immediate time limit is required for the occurrence of the action. The word "exemplary" is used herein to mean "used as an example, instance, or illustration". Any aspect described as "exemplary" herein is not necessarily interpreted as being preferred or having advantages over other aspects. Unless otherwise specified, the term "some" refers to one or more. Combinations such as "at least one of A, B, or C", "one or more of A, B, or C", "at least one of A, B and C", "one or more of A, B, and C", and "A, B, C, or any combination thereof" include any combination of A, B, and / or C, and may include multiple A, multiple B, or multiple C. Specifically, combinations such as "at least one of A, B, or C", "one or more of A, B, or C", "at least one of A, B, and C", "one or more of A, B, and C", and "A, B, C, or any combination thereof" may be only A, only B, only C, A and B, A and C, B and C, or A and B and C, wherein any such combination may contain one or more members of A, B, or C. A set should be interpreted as a set of elements, wherein the number of elements is one or more. Thus, for a set of X, X will include one or more elements. If the first device receives data from the second device or sends data to the second device, data can be directly received / sent between the first device and the second device, or indirectly received / sent between the first device and the second device through a collection of devices. All structural and functional equivalents of the elements throughout the various aspects described in this disclosure that are known or will be known later to those of ordinary skill in the art are expressly incorporated herein by reference and are 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. Words such as "module", "mechanism", "element", "device" cannot replace the word "component". Therefore, no claim element will be understood as a component plus function unless the element is explicitly recorded using the phrase "component for..."

[0159] As used herein, the phrase "based on" should not be interpreted as referring to a closed set of information, one or more conditions, one or more factors, etc. In other words, the phrase "based on A" (where "A" can be information, a condition, a factor, etc.) should be interpreted as "based at least on A" unless specifically stated differently.

[0160] The following aspects are merely illustrative and may be combined with other aspects or teachings described herein without limitation.

[0161] Aspect 1 is a method of wireless communication, the method comprising: processing information using machine learning associated with a model ID, a machine learning function, or a machine learning use case; and reporting data via the wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case.

[0162] In aspect 2, the method according to aspect 1 also includes: receiving the configuration identifying the model ID, the machine learning function, or the machine learning use case, wherein reporting the data includes sending the data based on the conditions, timing, or periodicity indicated in the configuration for the model ID, the machine learning function, or the machine learning use case.

[0163] In aspect 3, the method according to aspect 2 further comprises: the method is performed at a UE, and wherein the configuration is received from a network node, and the data is reported to the network node.

[0164] In aspect 4, the method according to aspect 2 further comprises: the method is performed at a network node, and wherein the configuration is received from a UE and the data is reported to the UE.

[0165] In aspect 5, the method according to aspect 2 further comprises: the method is performed at a first network node, and wherein the configuration is received from a second network node, and the data is reported to the second network node.

[0166] In aspect 6, the method according to aspect 2 further includes: the method is performed at a first UE, and wherein the configuration is received from a second UE, and the data is reported to the second UE.

[0167] In aspect 7, the method according to any one of aspects 1 to 6 also includes: the configuration associated with the model ID, the machine learning function or the machine learning use case indicates one or more of the following: a data reporting method, at least one input parameter for the machine learning, unprocessed data for obtaining model input parameters, at least one measurement for obtaining the model input parameters, at least one data processing module, timing for data reporting, conditions for the data reporting or periodicity for the data reporting.

[0168] In aspect 8, the method according to any one of aspects 1 to 7 further includes: reporting different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case.

[0169] In aspect 9, the method according to any one of aspects 1 to 8 further includes: receiving a data reporting activation for the model ID, the machine learning function, or the machine learning use case, wherein the data is reported in response to the activation.

[0170] In aspect 10, the method according to any one of aspects 1 to 9 also includes: receiving a deactivation of a data report for the model ID, the machine learning function, or the machine learning use case; and in response to the deactivation, stopping the reporting of the data for the model ID, the machine learning function, or the machine learning use case.

[0171] In aspect 11, the method according to any one of aspects 1 to 10 further includes: the data is reported in at least one of an RRC message, MAC-CE, UCI or DCI.

[0172] In aspect 12, the method according to any one of aspects 1 to 11 also includes: the configuration associated with the model ID, the machine learning function, or the machine learning use case includes a data validation configuration, and the method also includes: validating the data before the report based on the standards of the data validation configuration used for the model ID, the machine learning function, or the machine learning use case.

[0173] In aspect 13, the method according to aspect 12 also includes: the data validation configuration includes one or more of the following: at least one rule for data validation associated with the model ID, the machine learning function, or the machine learning use case, at least one data statistic associated with the model ID, the machine learning function, or the machine learning use case, or at least one data attribute associated with the model ID, the machine learning function, or the machine learning use case.

[0174] In aspect 14, the method according to aspect 12 or aspect 13 also includes: identifying at least one of the criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case in the inference or training output based on the machine learning; and indicating data validation failure based on the configuration used for the model ID, the machine learning function, or the machine learning use case.

[0175] In aspect 15, the method according to aspect 14 further comprises: indicating that the data validation failure also indicates a transition to a procedure without the machine learning.

[0176] Aspect 16 is an apparatus for wireless communication, the apparatus comprising means for performing the method according to any one of aspects 1 to 15.

[0177] Aspect 17 is an apparatus for performing wireless communications, the apparatus comprising: a memory; and at least one processor, the at least one processor being coupled to the memory and being configured to perform a method according to any one of Aspects 1 to 15 based at least in part on information stored in the memory.

[0178] In aspect 18, the apparatus according to aspect 16 or aspect 17 further comprises at least one transceiver or at least one antenna coupled to the at least one processor.

[0179] Aspect 19 is a non-transitory computer-readable medium storing computer-executable code, which, when executed by at least one processor, causes the at least one processor to implement the method as described in any one of aspects 1 to 15.

[0180] Aspect 20 is a method of wireless communication, the method comprising: providing a configuration for machine learning associated with a model ID, a machine learning function, or a machine learning use case; and receiving a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case.

[0181] In aspect 21, the method according to aspect 20 also includes: the configuration associated with the model ID, the machine learning function or the machine learning use case indicates one or more of the following: a data reporting method, at least one input parameter for the machine learning, unprocessed data for obtaining model input parameters, at least one measurement for obtaining the model input parameters, at least one data processing module, timing for data reporting, conditions for the data reporting or periodicity for the data reporting.

[0182] In aspect 22, the method according to aspect 20 or aspect 21 also includes: receiving a report of different data based on multiple configurations, each configuration being associated with a different model ID, a different machine learning function, or a different machine learning use case.

[0183] In aspect 23, the method according to any one of aspects 20 to 22 also includes: providing a data reporting activation for the model ID, the machine learning function, or the machine learning use case, wherein the data is received in response to the activation.

[0184] In aspect 24, the method according to any one of aspects 20 to 23 further includes: providing data reporting deactivation for the model ID, the machine learning function, or the machine learning use case.

[0185] In aspect 25, the method according to any one of aspects 20 to 24 further comprises: the data is received in at least one of an RRC message, a MAC-CE, a UCI or a DCI.

[0186] In aspect 26, the method according to any one of aspects 20 to 25 also includes: the configuration associated with the model ID includes a data validation configuration, and the data validation configuration includes one or more of the following: at least one rule for data validation associated with the model ID, the machine learning function, or the machine learning use case, at least one data statistic associated with the model ID, the machine learning function, or the machine learning use case, or at least one data attribute associated with the model ID, the machine learning function, or the machine learning use case.

[0187] In aspect 27, the method according to aspect 26 also includes: receiving a data validation failure, wherein the data validation failure indicates that at least one of the validation criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case is not satisfied in the inference or training output based on the machine learning.

[0188] In aspect 28, the method according to aspect 27 further comprises: indicating that the data validation failure also indicates a transition to a procedure without the machine learning.

[0189] In aspect 29, the method according to any one of aspects 20 to 28 further comprises: the method is performed at a UE, and wherein the configuration is provided to a network node, and the data is received from the network node.

[0190] In aspect 30, the method according to any one of aspects 20 to 28 further comprises: the method is performed at a network node, and wherein the configuration is provided to a UE and the configuration is received from the UE.

[0191] In aspect 31, the method according to any one of aspects 20 to 28 further comprises the method being performed at a first network node, and wherein the configuration is provided to a second network node, and wherein the data is received from the second network node.

[0192] In aspect 32, the method according to any one of aspects 20 to 28 further comprises the method being performed at a first UE, and wherein the configuration is provided to a second UE, and the data is received from the second UE.

[0193] Aspect 33 is an apparatus for wireless communication, the apparatus comprising means for performing the method according to any one of aspects 20 to 32.

[0194] Aspect 34 is an apparatus for wireless communication, the apparatus comprising: a memory; and at least one processor, the at least one processor being coupled to the memory and being configured to perform a method according to any one of Aspects 20 to 32 based at least in part on information stored in the memory.

[0195] In aspect 35, the apparatus according to aspect 33 or aspect 34 further comprises at least one transceiver or at least one antenna coupled to the at least one processor.

[0196] Aspect 36 is a non-transitory computer readable medium storing computer executable code, which, when executed by at least one processor, causes the at least one processor to implement the method as described in any one of aspects 20 to 32.

[0197] Aspect 37 is a method of wireless communication, the method comprising: registering a machine learning model for collecting and reporting data based on the wireless communication; and providing at least one of input features for the machine learning model, a data processing module for obtaining the input features for the machine learning model, or a data validation scheme for the machine learning model.

[0198] In aspect 38, the method according to aspect 37 also includes: the data validation scheme includes one or more of the following: at least one rule for data validation associated with the machine learning model, at least one first data statistic associated with the training data used for the machine learning model, at least one second data statistic associated with the inferred data used for the machine learning model, at least one first data attribute associated with the training data used for the machine learning model, or at least one second data attribute associated with the inferred data used for the machine learning model.

[0199] Aspect 39 is an apparatus for wireless communication, the apparatus comprising means for performing a method according to any one of aspects 37 to 38.

[0200] Aspect 40 is an apparatus for wireless communication, the apparatus comprising: a memory; and at least one processor, the at least one processor being coupled to the memory and configured to perform a method according to any one of Aspects 37 to 38 based at least in part on information stored in the memory.

[0201] In aspect 41, the apparatus according to aspect 39 or aspect 40 further comprises at least one transceiver or at least one antenna coupled to the at least one processor.

[0202] Aspect 42 is a non-transitory computer-readable medium storing computer-executable code, which, when executed by at least one processor, causes the at least one processor to implement the method as described in any one of aspects 37 to 38.

Claims

1. A device for wireless communication, the device comprising: Memory; and at least one processor coupled to the memory and based at least in part on the information stored in the memory, the at least one processor configured to: Processing the information using machine learning associated with a model identifier (ID), machine learning capability, or machine learning use case; and Data is reported via the wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case.

2. The device according to claim 1, further comprising: a transceiver coupled to the at least one processor, wherein the at least one processor is further configured to: Receiving the configuration identifying the model ID, the machine learning function, or the machine learning use case, wherein reporting the data includes sending the data based on a condition, timing, or periodicity indicated in the configuration for the model ID, the machine learning function, or the machine learning use case.

3. The apparatus of claim 2, wherein the apparatus is for performing the wireless communication at a user equipment (UE), and the at least one processor is configured to receive the configuration from a network node and report the data to the network node.

4. The apparatus of claim 2, wherein the apparatus is for performing the wireless communication at a network node, and the at least one processor is configured to receive the configuration from a user equipment (UE) and report the data to the UE.

5. The apparatus of claim 2, wherein the apparatus is for performing the wireless communication at a first network node, and the at least one processor is configured to receive the configuration from a second network node and report the data to the second network node.

6. The apparatus of claim 2, wherein the apparatus is for performing the wireless communication at a first user equipment (UE), and the at least one processor is configured to receive the configuration from a second UE and report the data to the second UE.

7. The apparatus of claim 1 , wherein the configuration associated with the model ID, the machine learning function, or the machine learning use case indicates one or more of the following: Data reporting methods, at least one input parameter for said machine learning, Unprocessed data used to obtain model input parameters, at least one measurement for obtaining said model input parameter, at least one data processing module, For timing of data reporting, the conditions for reporting of said data, or The periodicity used for the data reporting.

8. The apparatus of claim 1, wherein the at least one processor is further configured to: Report different data based on multiple configurations, each configuration associated with a different model ID, a different machine learning function, or a different machine learning use case.

9. The apparatus of claim 1 , wherein the at least one processor is further configured to: Receiving a data reporting activation for the model ID, the machine learning function, or the machine learning use case, and reporting the data in response to the activation.

10. The apparatus of claim 1, wherein the at least one processor is further configured to: receiving a data report deactivation for the model ID, the machine learning function, or the machine learning use case; and In response to the deactivation, the reporting of the data for the model ID, the machine learning function, or the machine learning use case is stopped.

11. The apparatus of claim 1 , wherein the at least one processor is configured to report the data in at least one of a radio resource control (RRC) message, a medium access control-control element (MAC-CE), uplink control information (UCI), or downlink control information (DCI).

12. The apparatus of claim 1 , wherein the configuration associated with the model ID, the machine learning function, or the machine learning use case comprises a data validation configuration, wherein the at least one processor is further configured to: Based on criteria of the data validation configuration for the model ID, the machine learning function, or the machine learning use case, the data is validated prior to reporting.

13. The apparatus of claim 12, wherein the data validation configuration comprises one or more of: at least one rule for data validation associated with the model ID, the machine learning function, or the machine learning use case, at least one data statistic associated with the model ID, the machine learning function, or the machine learning use case, or At least one data attribute associated with the model ID, the machine learning function, or the machine learning use case.

14. The apparatus of claim 12, wherein the at least one processor is further configured to: identifying at least one of the inference or training output based on the machine learning that does not satisfy the criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case; and A data validation failure is indicated according to the configuration for the model ID, the machine learning function, or the machine learning use case.

15. The apparatus of claim 14, wherein indicating that the data validation failed also indicates a transition to a procedure without the machine learning.

16. An apparatus for wireless communication, the apparatus comprising: Memory; and at least one processor coupled to the memory and based at least in part on the information stored in the memory, the at least one processor configured to: Providing a configuration for machine learning associated with a model identifier (ID), a machine learning function, or a machine learning use case; and Receive a data report based on the configuration associated with the model ID, the machine learning function, or the machine learning use case.

17. The apparatus of claim 16, wherein the configuration associated with the model ID, the machine learning function, or the machine learning use case indicates one or more of the following: Data reporting methods, at least one input parameter for said machine learning, Unprocessed data used to obtain model input parameters, at least one measurement for obtaining said model input parameter, at least one data processing module, For timing of data reporting, the conditions for reporting of said data, or The periodicity used for the data reporting.

18. The apparatus according to claim 16, further comprising: a transceiver coupled to the at least one processor, wherein the at least one processor is further configured to: Receive reports with different data based on multiple configurations, each associated with a different model ID, a different machine learning function, or a different machine learning use case.

19. The apparatus of claim 16, wherein the at least one processor is further configured to: Providing a data reporting activation for the model ID, the machine learning function, or the machine learning use case, wherein the data is received in response to the activation.

20. The apparatus of claim 16, wherein the at least one processor is further configured to: Provide data reporting deactivation for the model ID, the machine learning function, or the machine learning use case.

21. The apparatus of claim 16, wherein the at least one processor is configured to receive the data in at least one of a radio resource control (RRC) message, a medium access control-control element (MAC-CE), uplink control information (UCI), or downlink control information (DCI).

22. The apparatus of claim 16, wherein the configuration associated with the model ID comprises a data validation configuration, the data validation configuration comprising one or more of: at least one rule for data validation associated with the model ID, the machine learning function, or the machine learning use case, at least one data statistic associated with the model ID, the machine learning function, or the machine learning use case, or At least one data attribute associated with the model ID, the machine learning function, or the machine learning use case.

23. The apparatus of claim 22, wherein the at least one processor is further configured to: A data validation failure is received, the data validation failure indicating that at least one of the validation criteria of the data validation configuration associated with the model ID, the machine learning function, or the machine learning use case is not satisfied in an inference or training output based on the machine learning.

24. An apparatus according to claim 23, wherein indicating that the data validation failed also indicates a transition to a procedure without the machine learning.

25. An apparatus for registering a machine learning model, the apparatus comprising: Memory; and at least one processor coupled to the memory and based at least in part on the information stored in the memory, the at least one processor configured to: registering the machine learning model for collecting and reporting data based on wireless communications; and Provide at least one of input features for the machine learning model, a data processing module for obtaining the input features for the machine learning model, or a data validation scheme for the machine learning model.

26. The apparatus of claim 25, wherein the data verification scheme comprises one or more of: at least one rule for data validation associated with the machine learning model, at least one first data statistic associated with training data for the machine learning model, at least one second data statistic associated with the inferred data for the machine learning model, at least one first data attribute associated with the training data for the machine learning model, or At least one second data attribute associated with the inferred data for use with the machine learning model.

27. A method of wireless communication, the method comprising: Processing the information using machine learning associated with a model identifier (ID), a machine learning function, or a machine learning use case; as well as Data is reported via the wireless communication based on a configuration associated with the model ID, the machine learning function, or the machine learning use case.

28. The method according to claim 27, further comprising: Receiving the configuration identifying the model ID, the machine learning function, or the machine learning use case, wherein reporting the data includes sending the data based on a condition, timing, or periodicity indicated in the configuration for the model ID, the machine learning function, or the machine learning use case.

29. The method according to claim 27, further comprising: Report different data based on multiple configurations, each configuration associated with a different model ID, a different machine learning function, or a different machine learning use case.

30. The method of claim 27, wherein the configuration associated with the model ID, the machine learning function, or the machine learning use case comprises a data validation configuration, the method further comprising: The data is validated prior to the reporting based on criteria of the data validation configuration for the model ID, the machine learning function, or the machine learning use case.

Images