Combined performance indicators for machine learning models
By sending and receiving performance parameter information of machine learning models in a wireless communication system and dynamically adjusting the model combination, the problems of low communication efficiency and resource utilization in the prior art are solved, and more efficient communication and resource management are achieved.
Patent Information
- Application Number
- CN202380029500.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-04-13
- Filing Date
- 2023-02-23
- Publication Date
- 2024-11-08
AI Technical Summary
When existing wireless communication systems combine machine learning models, it is difficult to effectively manage and optimize performance parameters, resulting in a decrease in communication efficiency and resource utilization.
Performance parameters of the machine learning model are indicated by sending and receiving capability information between user equipment (UE) and network nodes, for selecting and combining appropriate models to optimize the communication process. The method includes the UE sending capability information to indicate supported model combinations and based on the received information, the network node selects an appropriate ML model to improve communication efficiency and save resources.
The efficiency and resource utilization of wireless communication are improved, and the combination of machine learning models is dynamically adjusted, unnecessary power and computing resources are reduced, and communication performance is optimized.
Smart Images

Figure CN118922837A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This patent application claims priority to U.S. non-provisional patent application No. 17 / 659,129, filed on April 13, 2022, entitled “PERFORMANCE INDICATORS FOR COMBINATIONS OF MACHINE LEARNING MODELS,” which is hereby expressly incorporated herein by reference. Technical Field
[0003] Aspects of the present disclosure relate generally to wireless communications, and to techniques and apparatus for combining parameters of machine learning models. Background Art
[0004] Wireless communication systems are widely deployed to provide a variety of telecommunication services, such as telephony, video, data, messaging, and broadcasting. Typical wireless communication systems may employ multiple access technologies that can support communication with multiple users by sharing available system resources (e.g., bandwidth, transmit power, etc.). 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, time division synchronous code division multiple access (TD-SCDMA) systems, and long term evolution (LTE). LTE / LTE-Advanced is a set of enhancements to the Universal Mobile Telecommunications System (UMTS) mobile standard promulgated by the Third Generation Partnership Project (3GPP).
[0005] A wireless network may include one or more base stations that support communications for a user equipment (UE) or multiple UEs. A UE may communicate with a base station via downlink communications and uplink communications. A "downlink" (or "DL") refers to the communication link from a base station to a UE, and an "uplink" (or "UL") refers to the communication link from a UE to a base station.
[0006] The above-mentioned multiple access technologies have been adopted in various telecommunication standards to provide a common protocol that enables different UEs to communicate at a city, country, region and / or global level. New Radio (NR), which may be referred to as 5G, is a set of enhancements to the LTE mobile standard promulgated by 3GPP. NR is designed to better integrate with other open standards by improving spectrum efficiency, reducing costs, improving services, utilizing new spectrum, and using orthogonal frequency division multiplexing (OFDM) (CP-OFDM) with cyclic prefix (CP) on the downlink, using CP-OFDM and / or single carrier frequency division multiplexing (SC-FDM) (also known as discrete Fourier transform spread OFDM (DFT-s-OFDM)) on the uplink, as well as supporting beamforming, multiple input multiple output (MIMO) antenna technology and carrier aggregation, so as to better support mobile broadband Internet access. As the demand for mobile broadband access continues to increase, further improvements in LTE, NR and other radio access technologies remain useful. Summary of the invention
[0007] Some aspects described herein relate to a method of wireless communication performed by a user equipment (UE). The method may include sending capability information indicating support for one or more model combinations of machine learning (ML) models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML model relative to a model combination of the one or more model combinations including the ML model. The method may include receiving one or more indications to use one or more ML models in the ML model based at least in part on the capability information.
[0008] Some aspects described herein relate to a method of wireless communication performed by a network node. The method may include receiving capability information indicating support of one or more model combinations of ML models by a UE, wherein the capability information further indicates one or more performance parameters of an ML model of the one or more ML models relative to a model combination of the one or more model combinations including the ML model. The method may include sending one or more indications of using one or more ML models of the ML models based at least in part on the capability information.
[0009] Some aspects described herein relate to a UE for wireless communication. The UE may include a memory and one or more processors coupled to the memory. The one or more processors may be configured to send capability information indicating support for one or more model combinations of ML models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML model relative to a model combination in the one or more model combinations including the ML model. The one or more processors may be configured to receive one or more indications to use one or more ML models in the ML model based at least in part on the capability information.
[0010] Some aspects described herein relate to a network node for wireless communication. The network node may include a memory and one or more processors coupled to the memory. The one or more processors may be configured to receive capability information indicating support of one or more model combinations of ML models by the UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination in the one or more model combinations including the ML model. The one or more processors may be configured to send one or more indications of using one or more ML models in the ML models based at least in part on the capability information.
[0011] Some aspects described herein relate to a non-transitory computer-readable medium storing an instruction set for wireless communication by a UE. The instruction set, when executed by one or more processors of the UE, may cause the UE to send capability information indicating support for one or more model combinations of ML models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML model relative to a model combination including the ML model in the one or more model combinations. The instruction set, when executed by one or more processors of the UE, may cause the UE to receive one or more indications of using one or more ML models in the ML model based at least in part on the capability information.
[0012] Some aspects described herein relate to a non-transitory computer-readable medium storing an instruction set for wireless communication by a network node. The instruction set, when executed by one or more processors of the network node, can cause the network node to receive capability information indicating support of one or more model combinations of ML models by a UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination including the ML model in the one or more model combinations. The instruction set, when executed by one or more processors of the network node, can cause the network node to send one or more indications of using one or more ML models in the ML model based at least in part on the capability information.
[0013] Some aspects described herein are directed to an apparatus for wireless communication. The apparatus may include means for sending capability information indicating support for one or more model combinations of ML models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML model relative to a model combination of the one or more model combinations that includes the ML model. The apparatus may include means for receiving one or more indications of using one or more ML models in the ML model based at least in part on the capability information.
[0014] Some aspects described herein are directed to an apparatus for wireless communication. The apparatus may include means for receiving capability information indicating support of one or more model combinations of ML models by a UE, wherein the capability information further indicates one or more performance parameters of an ML model of the one or more ML models relative to a model combination of the one or more model combinations including the ML model. The apparatus may include means for sending one or more indications of using one or more ML models of the ML models based at least in part on the capability information.
[0015] The various aspects generally include methods, apparatuses, systems, computer program products, non-transitory computer-readable media, user equipment, base stations, wireless communication devices and / or processing systems as fully described herein with reference to and illustrated by the drawings and description.
[0016] The features and technical advantages of examples according to the present disclosure have been outlined quite extensively above so that the following specific embodiments may be better understood. Additional technical features and advantages will be described below. The disclosed concepts and specific examples may be easily used as a basis for modifying or designing other structures for achieving the same purpose of the present disclosure. Such equivalent constructions do not depart from the scope of protection of the appended claims. The characteristics of the concepts disclosed herein, both in terms of their organization and method of operation, and the associated advantages will be better understood by considering the following description in conjunction with the accompanying drawings. Each of the accompanying drawings is provided for the purpose of illustration and description, and not as a definition of limitations of the claims.
[0017] Although various aspects are described in the present disclosure by the description of some examples, it will be understood by those skilled in the art that such aspects can be implemented in many different arrangements and scenarios. The technology described herein can be implemented using different platform types, devices, systems, shapes, sizes and / or packaging arrangements. For example, some aspects can be implemented via integrated chip implementations or other devices based on non-module components (e.g., end-user devices, vehicles, communication equipment, computing equipment, industrial equipment, retail / purchasing equipment, medical equipment and / or artificial intelligence (AI) devices). Various aspects can be implemented in chip-level components, modular components, non-modular components, non-chip-level components, device-level components and / or system-level components. The equipment combined with the various aspects and features described may include additional components and features for implementing and practicing the various aspects claimed and described. For example, the transmission and reception of wireless signals may include one or more components for analog and digital purposes (e.g., hardware components, including antennas, radio frequency (RF) chains, power amplifiers, modulators, buffers, processors, interleavers, adders and / or summers). The various aspects described herein are intended to be practiced in a wide variety of devices, components, systems, distributed arrangements, and / or end-user devices of different sizes, shapes, and configurations. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to be able to understand the above-mentioned features of the present disclosure in detail, a more specific description briefly summarized above may be obtained by reference to various aspects (some of which are illustrated in the accompanying drawings). However, it should be noted that the accompanying drawings illustrate only certain typical aspects of the present disclosure and are therefore not to be considered as limiting the scope thereof, as the specification may admit of other equally effective aspects. The same reference numerals in different drawings may identify the same or similar elements.
[0019] Figure 1 is a diagram illustrating an example of a wireless network according to the present disclosure.
[0020] Figure 2 is a diagram illustrating an example of communication between a base station and a user equipment (UE) in a wireless network according to the present disclosure.
[0021] Figure 3 is a diagram illustrating an example 300 decomposed base station architecture according to the present disclosure.
[0022] Figure 4 is a diagram illustrating an example associated with conveying UE capabilities for machine learning models according to the present disclosure.
[0023] Figures 5 to 7 is a diagram illustrating an example associated with combined parameters for a machine learning model according to the present disclosure.
[0024] Figure 8 and Fig. 9 is a diagram illustrating an example process associated with combining parameters for a machine learning model according to the present disclosure.
[0025] Fig.10 and Fig.11 is a diagram of an example apparatus for wireless communications according to the present disclosure. DETAILED DESCRIPTION
[0026] The various aspects of the present disclosure are described more fully below with reference to the accompanying drawings. However, the present disclosure can be embodied in many different forms, and should not be construed as being limited to any specific structure or function presented throughout the present disclosure. Instead, these aspects are provided so that the present disclosure will be thorough and complete, and the protection scope of the present disclosure will be fully conveyed to those skilled in the art. It should be appreciated by those skilled in the art that the scope of the present disclosure is intended to cover any aspect of the present disclosure disclosed herein, whether it is implemented independently or in combination with any other aspect of the present disclosure. For example, any number of aspects set forth herein may be used to implement a device or practice method. In addition, the scope of the present disclosure is intended to cover such a device or method that is practiced using other structures, functionality, or structure and functionality in addition to or different from the various aspects of the present disclosure set forth herein. It should be understood that any aspect of the present disclosure disclosed herein may be embodied by one or more elements of the claims.
[0027] Several aspects of telecommunication systems will now be presented with reference to various devices and techniques. These devices and techniques will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as "elements"). These elements can be implemented using hardware, software, or a 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.
[0028] Although aspects may be described herein using terminology generally associated with 5G or new radio (NR) radio access technology (RAT), aspects of the present disclosure may be applied to other RATs, such as 3G RAT, 4G RAT, and / or post-5G RATs (e.g., 6G).
[0029] Figure 11 is a diagram illustrating an example of a wireless network 100 according to the present disclosure. The wireless network 100 may be a 5G (e.g., NR) network and / or a 4G (e.g., Long Term Evolution (LTE)) network, or may include elements of a 5G (e.g., NR) network and / or elements of a 4G (e.g., Long Term Evolution (LTE)) network, among other examples. The wireless network 100 may include one or more base stations 110 (shown as BS110a, BS110b, BS110c, and BS110d), a user equipment (UE) 120 or multiple UEs 120 (shown as UE120a, UE 120b, UE 120c, UE 120d, and UE 120e), and / or other network entities. The base station 110 is an entity that communicates with the UE 120. Base station 110 (sometimes referred to as BS) may include, for example, an NR base station, an LTE base station, a Node B, an eNB (e.g., in 4G), a gNB (e.g., in 5G), an access point, and / or a transmit receive point (TRP). Each base station 110 may provide communication coverage for a particular geographic area. In the Third Generation Partnership Project (3GPP), the term "cell" may refer to a coverage area of a base station 110 and / or a base station subsystem serving the coverage area, depending on the context in which the term is used.
[0030] The base station 110 may provide communication coverage for a macro cell, a pico cell, a femto cell, and / or another type of cell. A macro cell may cover a relatively large geographic area (e.g., a radius of several kilometers) and may allow unrestricted access by a UE 120 with a service subscription. A pico cell may cover a relatively small geographic area and may allow unrestricted access by a UE 120 with a service subscription. A femto cell may cover a relatively small geographic area (e.g., a home) and may allow restricted access by a UE 120 that has an association with the femto cell (e.g., a UE 120 in a closed subscriber group (CSG)). A base station 110 for a macro cell may be referred to as a macro base station. A base station 110 for a pico cell may be referred to as a pico base station. A base station 110 for a femto cell may be referred to as a femto base station or a home base station. In Figure 1 In the example shown in , BS 110a may be a macro base station for macro cell 102a, BS 110b may be a pico base station for pico cell 102b, and BS 110c may be a femto base station for femto cell 102c. A base station may support one or more (eg, three) cells.
[0031] In some examples, the cell may not necessarily be stationary, and the geographic area of the cell may move according to the location of the mobile base station 110 (e.g., a mobile base station). In some examples, the base stations 110 may be interconnected with each other and / or with one or more other base stations 110 or network nodes (not shown) in the wireless network 100 through various types of backhaul interfaces (such as direct physical connections or virtual networks) using any suitable transport network.
[0032] The wireless network 100 may include one or more relay stations. A relay station is an entity that can receive transmissions of data from an upstream station (e.g., a base station 110 or a UE 120) and transmit transmissions of data to a downstream station (e.g., a UE 120 or a base station 110). A relay station may be a UE 120 that is capable of relaying transmissions for other UEs 120. Figure 1 In the example shown in FIG, BS 110d (eg, a relay base station) can communicate with BS 110a (eg, a macro base station) and UE 120d to facilitate communication between BS 110a and UE 120d. Base station 110 that relays communication may be referred to as a relay station, relay base station, relay, or the like.
[0033] The wireless network 100 may be a heterogeneous network that includes different types of base stations 110, such as macro base stations, pico base stations, femto base stations, relay base stations, etc. These different types of base stations 110 may have different transmit power levels, different coverage areas, and / or different impacts on interference in the wireless network 100. For example, a macro base station may have a high transmit power level (e.g., 5 watts to 40 watts), while a pico base station, a femto base station, and a relay base station may have a lower transmit power level (e.g., 0.1 watt to 2 watts).
[0034] A network controller 130 may be coupled to or in communication with a set of base stations 110 and may provide coordination and control for the base stations 110. The network controller 130 may communicate with the base stations 110 via backhaul communication links. The base stations 110 may also communicate directly with each other or indirectly via wireless or wired backhaul communication links.
[0035] UE 120 can be dispersed throughout the wireless network 100, and each UE 120 can be stationary or mobile. UE 120 can include, for example, access terminals, terminals, mobile stations, and / or subscriber units. UE 120 can be a cellular phone (e.g., a smart phone), a personal digital assistant (PDA), a wireless modem, a wireless communication device, a handheld device, a laptop computer, a cordless phone, a wireless local loop (WLL) station, a tablet computer, a camera, a gaming device, a netbook, a smartbook, an ultrabook, a medical device, a biometric device, a wearable device (e.g., a smart watch, smart clothing, smart glasses, a smart wristband, smart jewelry (e.g., a smart ring or a smart bracelet)), an entertainment device (e.g., a music device, a video device, and / or a satellite radio), a vehicle component or sensor, a smart meter / sensor, an industrial manufacturing equipment, a global positioning system device, and / or any other suitable device configured to communicate via a wireless medium.
[0036] Some UEs 120 may be considered as machine type communication (MTC) or evolved or enhanced machine type communication (eMTC) UEs. MTC UEs and / or eMTC UEs may include, for example, robots, drones, remote devices, sensors, meters, monitors, and / or location tags that may communicate with a base station, another device (e.g., a remote device), or some other entity. Some UEs 120 may be considered as Internet of Things (IoT) devices and / or may be implemented as NB-IoT (narrowband IoT) devices. Some UEs 120 may be considered as customer premises equipment. UE 120 may be included inside a housing that houses components of UE 120, such as a processor component and / or a memory component. In some examples, the processor component and the memory component may be coupled together. For example, a processor component (e.g., one or more processors) and a memory component (e.g., a memory) are operationally coupled, communicatively coupled, electronically coupled, and / or electrically coupled.
[0037] In general, any number of wireless networks 100 may be deployed in a given geographic area. Each wireless network 100 may support a specific RAT and may operate on one or more frequencies. RAT may be referred to as a radio technology, air interface, etc. Frequency may be referred to as a carrier, frequency channel, etc. Each frequency may support a single RAT in a given geographic area to avoid interference between wireless networks of different RATs. In some cases, NR or 5G RAT networks may be deployed.
[0038] In some examples, two or more UEs 120 (e.g., shown as UE 120a and UE 120e) may communicate directly (e.g., without using base station 110 as an intermediary to communicate with each other) using one or more side link channels. For example, UE 120 may communicate using peer-to-peer (P2P) communication, device-to-device (D2D) communication, vehicle-to-everything (V2X) protocols (e.g., which may include vehicle-to-vehicle (V2V) protocols, vehicle-to-infrastructure (V2I) protocols, or vehicle-to-pedestrian (V2P) protocols), and / or mesh networks. In such examples, UE 120 may perform scheduling operations, resource selection operations, and / or other operations described elsewhere herein as being performed by base station 110.
[0039] The devices of the wireless network 100 may communicate using an electromagnetic spectrum, which may be subdivided into various categories, bands, channels, etc., based on frequency or wavelength. For example, the devices of the wireless network 100 may communicate using one or more operating bands. In 5G NR, two initial operating bands have been identified as frequency ranges designated FR1 (410 MHz-7.125 GHz) and FR2 (24.25 GHz-52.6 GHz). It should be understood that, although a portion of FR1 is greater than 6 GHz, FR1 is often (interchangeably) referred to as a "below 6 GHz" band in various documents and articles. A similar naming problem sometimes occurs with respect to FR2, which is often (interchangeably) referred to as a "millimeter wave" band in documents and articles, although different from the extremely high frequency (EHF) band (30 GHz-300 GHz) identified as a "millimeter wave" band by the International Telecommunication Union (ITU).
[0040] 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, so 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 FR4a or FR4-1 (52.6GHz-71GHz), FR4 (52.6GHz-114.25GHz) and FR5 (114.25GHz-300GHz). Each of these higher frequency bands falls within the EHF band.
[0041] Considering the above examples, unless otherwise specifically stated, it should be understood that if the term "below 6 GHz" or the like is used herein, the term may broadly refer to frequencies that may be lower than 6 GHz, may be within FR1, or may include mid-band frequencies. In addition, unless otherwise specifically stated, it should be understood that if the term "millimeter wave" or the like is used herein, the term may broadly refer to frequencies that may include mid-band frequencies, may be within FR2, FR4, FR4-a or FR4-1, and / or FR5, or may be within the EHF band. It is contemplated that the frequencies included in these operating bands (e.g., FR1, FR2, FR3, FR4, FR4-a, FR4-1, and / or FR5) may be modified, and the techniques described herein are applicable to those modified frequency ranges.
[0042] In some aspects, the UE 120 may include a communication manager 140. As described in more detail elsewhere herein, the communication manager 140 may send capability information indicating support for one or more model combinations of ML models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML model relative to a model combination in the one or more model combinations that includes the ML model; and receive one or more indications to use one or more ML models in the ML model based at least in part on the capability information. Additionally or alternatively, the communication manager 140 may perform one or more other operations described herein.
[0043] In some aspects, a network node (e.g., base station 110) may include a communication manager 150. As described in more detail elsewhere herein, the communication manager 150 may receive capability information indicating support of one or more model combinations of ML models by the UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination including the ML model in the one or more model combinations; and send one or more indications to use one or more ML models in the ML models based at least in part on the capability information. Additionally or alternatively, the communication manager 150 may perform one or more other operations described herein.
[0044] In some aspects, the term "base station" (e.g., base station 110) or "network node" or "network entity" may refer to a converged base station, a decomposed base station (e.g., a combined Fig. 9), integrated access and backhaul (IAB) nodes, relay nodes, and / or one or more components thereof. For example, in some aspects, a "base station", "network node", or "network entity" may refer to a central unit (CU or centralized unit), a distributed unit (DU), a radio unit (RU), a near real-time (near RT) RAN intelligent controller (RIC) or a non-real-time (non-RT) RIC, or a combination thereof. In some aspects, the term "base station", "network node", or "network entity" may refer to a device configured to perform one or more functions (such as those described herein in conjunction with base station 110). In some aspects, the term "base station", "network node", or "network entity" may refer to a plurality of devices configured to perform one or more functions. For example, in some distributed systems, each of a plurality of different devices (which may be located in the same geographic location or in different geographic locations) may be configured to perform at least a portion of a function, or to repeatedly perform at least a portion of the function, and the term "base station", "network node", or "network entity" may refer to any one or more of these different devices. In some aspects, the term "base station", "network node", or "network entity" may refer to one or more virtual base stations and / or one or more virtual base station functions. For example, in some aspects, two or more base station functions may be instantiated on a single device. In some aspects, the terms "base station," "network node," or "network entity" may refer to one of the base station functions but not another. In this way, a single device may include more than one base station.
[0045] As indicated above, Figure 1 are provided as examples. Other examples can be found in the Figure 1 The content described is different.
[0046] Figure 2 2 is a diagram illustrating an example 200 of base station 110 communicating with UE 120 in wireless network 100 according to the present disclosure. Base station 110 may be equipped with a set of antennas 234a to 234t, such as T antennas (T≥1). UE 120 may be equipped with a set of antennas 252a to 252r, such as R antennas (R≥1).
[0047] At the base station 110, the transmit processor 220 may receive data intended for the UE 120 (or a set of UEs 120) from the data source 212. The transmit processor 220 may select one or more modulation and coding schemes (MCS) for the UE 120 based at least in part on one or more channel quality indicators (CQI) received from the UE 120. The base station 110 may process (e.g., encode and modulate) the data for the UE 120 based at least in part on the MCS selected for the UE 120, and provide data symbols for the UE 120. The transmit processor 220 may process system information (e.g., for semi-static resource partitioning information (SRPI)) and control information (e.g., CQI requests, grants, and / or upper layer signaling), and provide overhead symbols and control symbols. The transmit processor 220 may generate reference symbols for reference signals (e.g., cell-specific reference signals (CRS) or demodulation reference signals (DMRS)) and synchronization signals (e.g., primary synchronization signals (PSS) or secondary synchronization signals (SSS)). The transmit (TX) multiple-input multiple-output (MIMO) processor 230 may perform spatial processing (e.g., pre-coding) on data symbols, control symbols, overhead symbols, and / or reference symbols, where applicable, and may provide a set of output symbol streams (e.g., T output symbol streams) to a corresponding set of modems 232 (e.g., T modems) (shown as modems 232a to 232t). For example, each output symbol stream may be provided to a modulator component (shown as MOD) of the modem 232. Each modem 232 may process a corresponding output symbol stream (e.g., for OFDM) using a corresponding modulator component to obtain an output sample stream. Each modem 232 may further process (e.g., convert to analog, amplify, filter, and / or up-convert) the output sample stream using a corresponding modulator component to obtain a downlink signal. The modems 232a to 232t may transmit a set of downlink signals (e.g., T downlink signals) via a corresponding set of antennas 234 (e.g., T antennas) (shown as antennas 234a to 234t).
[0048] At the UE 120, a set of antennas 252 (shown as antennas 252a to 252r) may receive downlink signals from the base station 110 and / or other base stations 110 and may provide a set of received signals (e.g., R received signals) to a set of modems 254 (e.g., R modems) (shown as modems 254a to 254r). For example, each received signal may be provided to a demodulator component (shown as DEMOD) of the modem 254. Each modem 254 may use a corresponding demodulator component to condition (e.g., filter, amplify, downconvert, and / or digitize) the received signal to obtain input samples. Each modem 254 may use a demodulator component to further process the input samples (e.g., for OFDM) to obtain received symbols. A MIMO detector 256 may obtain received symbols from the modem 254, may perform MIMO detection on the received symbols where applicable, and may provide detected symbols. The receive processor 258 may process (e.g., demodulate and decode) the detected symbols, may provide decoded data for the UE 120 to the data sink 260, and may provide decoded control information and system information to the controller / processor 280. The term "controller / processor" may refer to one or more controllers, one or more processors, or a combination thereof. The channel processor may determine a reference signal received power (RSRP) parameter, a received signal strength indicator (RSSI) parameter, a reference signal received quality (RSRQ) parameter, and / or a CQI parameter, among other things. In some examples, one or more components of the UE 120 may be included in the housing 284.
[0049] The network controller 130 may include a communication unit 294, a controller / processor 290, and a memory 292. The network controller 130 may include, for example, one or more devices in a core network. The network controller 130 may communicate with the base station 110 via the communication unit 294.
[0050] One or more antennas (e.g., antennas 234a to 234t and / or antennas 252a to 252r) may include or be included within one or more antenna panels, one or more antenna groups, one or more sets of antenna elements, and / or one or more antenna arrays, etc. Antenna panels, antenna groups, sets of antenna elements, and / or antenna arrays may include one or more antenna elements (in a single housing or multiple housings), sets of coplanar antenna elements, sets of non-coplanar antenna elements, and / or may be coupled to one or more transmit and / or receive components (such as, Figure 2 One or more antenna elements of one or more components in.
[0051] On the uplink, at the UE 120, the transmit processor 264 may receive and process data from the data source 262 and control information from the controller / processor 280 (e.g., for reports including RSRP, RSSI, RSRQ, and / or CQI). The transmit processor 264 may generate reference symbols for one or more reference signals. The symbols from the transmit processor 264 may be pre-decoded by the TX MIMO processor 266, where applicable, further processed by the modem 254 (e.g., for DFT-s-OFDM or CP-OFDM), and transmitted to the base station 110. In some examples, the modem 254 of the UE 120 may include a modulator and a demodulator. In some examples, the UE 120 includes a transceiver. The transceiver may include any combination of an antenna 252, a modem 254, a MIMO detector 256, a receive processor 258, a transmit processor 264, and / or a TX MIMO processor 266. The transceiver may be used by a processor (eg, controller / processor 280) and memory 282 to perform the functions described herein (eg, with reference to Figures 5 to 11 )Aspects of any of the methods described herein.
[0052] At the base station 110, uplink signals from the UE 120 and / or other UEs may be received by the antenna 234, processed by the modem 232 (e.g., a demodulator component of the modem 232 shown as DEMOD), detected by the MIMO detector 236 (if applicable), and further processed by the receive processor 238 to obtain decoded data and control information transmitted by the UE 120. The receive processor 238 may provide the decoded data to the data sink 239 and the decoded control information to the controller / processor 240. The base station 110 may include a communication unit 244 and may communicate with the network controller 130 via the communication unit 244. The base station 110 may include a scheduler 246 to schedule one or more UEs 120 for downlink and / or uplink communications. In some examples, the modem 232 of the base station 110 may include a modulator and a demodulator. In some examples, the base station 110 includes a transceiver. The transceiver may include any combination of antenna 234, modem 232, MIMO detector 236, receive processor 238, transmit processor 220, and / or TX MIMO processor 230. The transceiver may be used by a processor (e.g., controller / processor 240) and memory 242 to perform operations described herein (e.g., with reference to Figures 5 to 11 )Aspects of any of the methods described herein.
[0053] The controller / processor 240 of the base station 110, the controller / processor 280 of the UE 120, and / or Figure 2Any other component in may perform one or more techniques associated with combining parameters for ML models, as described in more detail elsewhere herein. In some aspects, the network nodes described herein are Figure 2 The base station 110 is shown, is included in the base station 110, or includes one or more components of the base station 110. For example, the controller / processor 240 of the base station 110, the controller / processor 280 of the UE 120, and / or Figure 2 Any other component of the Figure 8 The process of 800 Fig. 9 900 and / or operations of other processes as described herein. Memory 242 and memory 282 may store data and program codes for base station 110 and UE 120, respectively. In some examples, memory 242 and / or memory 282 may include a non-transitory computer-readable medium storing one or more instructions (e.g., code and / or program code) for wireless communication. For example, the one or more instructions, when executed by one or more processors of base station 110 and / or UE 120 (e.g., directly, or after compilation, conversion and / or interpretation), may cause the one or more processors, UE 120 and / or base station 110 to perform or direct, for example, Figure 8 The process of 800 Fig. 9 The process 900 and / or operations of other processes as described herein. In some examples, executing instructions may include running instructions, converting instructions, compiling instructions, and / or interpreting instructions, etc.
[0054] In some aspects, a UE (e.g., UE 120) includes: means for sending capability information indicating support for one or more model combinations of ML models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML model relative to a model combination of the one or more model combinations including the ML model; and / or means for receiving one or more indications of using one or more ML models in the ML model based at least in part on the capability information. Means for the UE to perform operations described herein may include, for example, one or more of the communications manager 140, antenna 252, modem 254, MIMO detector 256, receive processor 258, transmit processor 264, TX MIMO processor 266, controller / processor 280, or memory 282.
[0055] In some aspects, a network node (e.g., base station 110) includes: means for receiving capability information indicating support of one or more model combinations of ML models by the UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination including the ML model in the one or more model combinations; and / or means for sending one or more indications of using one or more ML models in the ML models based at least in part on the capability information. In some aspects, means for the network node to perform operations described herein may include, for example, one or more of the communication manager 150, the transmit processor 220, the TX MIMO processor 230, the modem 232, the antenna 234, the MIMO detector 236, the receive processor 238, the controller / processor 240, the memory 242, or the scheduler 246.
[0056] Although Figure 2 The blocks in the 200 and 210 are illustrated as distinct components, but the functionality described above for these blocks may be implemented in a single hardware, software, or combined component or in various combinations of components. For example, the functionality described with respect to the transmit processor 264, the receive processor 258, and / or the TX MIMO processor 266 may be performed by or under the control of the controller / processor 280.
[0057] As indicated above, Figure 2 are provided as examples. Other examples can be found in the Figure 2 The content described is different.
[0058] Figure 3 is a diagram illustrating an example 300 decomposed base station architecture according to the present disclosure.
[0059] 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 RAN node, a core network node, a network element or a network equipment (such as a base station (BS, e.g., base station 110) or one or more units (or one or more components) performing base station functionality can be implemented in an aggregated or decomposed architecture. For example, a BS (such as a Node B (NB), an eNB, an NR BS, a 5G NB, an access node (AP), a TRP, a cell, etc.) can be implemented as an aggregated base station (also called a stand-alone BS or a monolithic BS) or a decomposed base station.
[0060] A converged 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 CUs, one or more DUs, or one or more 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 also be implemented as a virtual unit, namely a virtual centralized unit (VCU), a virtual distributed unit (VDU), or a virtual radio unit (VRU).
[0061] Base station type operations 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 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.
[0062] Figure 3 The decomposed base station architecture shown in FIG. 1 may include one or more CUs 310 that may communicate directly with a core network 320 via a backhaul link, or indirectly with a core network 320 through one or more decomposed base station units, such as a near-RT RAN RIC 325 via an E2 link, or a non-RT RIC 315 associated with a service management and orchestration (SMO) framework 305, or both. The CU 310 may communicate with one or more DUs 330 via corresponding midhaul links, such as an F1 interface. The DU 330 may communicate with one or more RUs 340 via corresponding fronthaul links. The RU 340 may communicate with corresponding UEs 120 via one or more radio frequency (RF) access links. In some implementations, a UE 120 may be served simultaneously by multiple RUs 340.
[0063] Each of these units (e.g., CU 310, DU 330, RU 340) and near-RT RIC 325, non-RTRIC 315, and SMO framework 305 may include one or more interfaces, or be coupled to one or more interfaces configured to receive or send signals, data, or information (collectively referred to as signals) via a wired or wireless transmission medium. Each of the units or an associated processor or controller that provides instructions to the communication interface 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 or send signals to one or more of the other units via a wired transmission medium. In addition, these units may include a wireless interface that may include a receiver, a transmitter, or a transceiver (such as an RF transceiver) that is configured to receive or send signals, or both, to one or more of the other units on a wireless transmission medium.
[0064] In some aspects, CU 310 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 310. CU310 may be configured to handle user plane functionality (e.g., central unit-user plane (CU-UP)), control plane functionality (e.g., central unit-control plane (CU-CP)), or a combination thereof. In some specific implementations, CU 310 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 310 may be implemented to communicate with DU 330 for network control and signaling.
[0065] DU 330 may correspond to a logical unit that includes one or more base station functions for controlling the operation of one or more RU 340. In some aspects, DU 330 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 defined by the Third Generation Partnership Project (3GPP). In some aspects, DU 330 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 330 or with control functions hosted by CU 310.
[0066] The lower layer functionality may be implemented by one or more RUs 340. In some deployments, a RU 340 controlled by a DU 330 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 340 may be implemented to handle over-the-air (OTA) communications with one or more UEs 120. In some implementations, real-time and non-real-time aspects of control plane and user plane communications with the RU 340 may be controlled by the corresponding DU 330. In some scenarios, this configuration may enable the implementation of the DU 330 and the CU 310 in a cloud-based RAN architecture (such as a vRAN architecture).
[0067] The SMO framework 305 may be configured to support RAN deployment and provisioning of non-virtualized network elements and virtualized network elements. For non-virtualized network elements, the SMO framework 305 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 305 may be configured to interact with a cloud computing platform (such as an open cloud (O-Cloud) 390) 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 310, DU 330, RU 340, and near-RT RIC 325. In some specific implementations, the SMO framework 305 may communicate with hardware aspects of the 4G RAN (such as an open eNB (O-eNB) 311) via the O1 interface. In addition, in some specific implementations, the SMO framework 305 may communicate directly with one or more RUs 340 via the O1 interface. The SMO framework 305 may also include a non-RT RIC 315 configured to support the functionality of the SMO framework 305 .
[0068] The non-RT RIC 315 may be configured to include logic functions that enable non-real-time control and optimization of RAN elements and resources, artificial intelligence / machine learning (AI / ML) workflows including model training and updating, or policy-based guidance of applications / features in the near-RT RIC 325. The non-RT RIC 315 may be coupled to or communicate with the near-RT RIC 325 (such as via an A1 interface). The near-RT RIC 325 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 310, one or more DUs 330, or both, and the O-eNB with the near-RT RIC 325.
[0069] In some implementations, in order to generate an AI / ML model to be deployed in the near-RT RIC 325, the non-RT RIC 315 may receive parameters or external enrichment information from an external server. Such information may be utilized by the near-RT RIC 325 and may be received from a non-network data source or from a network function at the SMO framework 305 or the non-RT RIC 315. In some examples, the non-RT RIC 315 or the near-RT RIC 325 may be configured to tune RAN behavior or performance. For example, the non-RT RIC 315 may monitor long-term trends and patterns of performance and employ AI / ML models to perform corrective actions through the SMO framework 305 (such as via reconfiguration of O1) or via creation of RAN management policies (such as A1 policies).
[0070] As indicated above, Figure 3 are provided as examples. Other examples can be found in the Figure 3 The content described is different.
[0071] Figure 4 4 is a diagram illustrating an example 400 associated with communicating UE capabilities for ML models according to the present disclosure. Figure 4 As shown, a network node and a UE may communicate via a wireless network, such as wireless network 100. The UE may support one or more ML models associated with communicating with the network node.
[0072] The UE may receive a UE capability query and the network node may send the UE capability query, as indicated by reference numeral 405. For example, the network node may request an indication of supported ML models and / or the performance of each of the ML models.
[0073] As indicated by reference numeral 410, the UE may send a capability report associated with each ML model and the network node may receive the capability report. For example, the capability report may indicate each supported ML model and may indicate performance parameters for each of the supported ML models.
[0074] As indicated by reference numeral 415, the UE may receive an indication of a set of one or more ML models used in communicating with a network node, and the network node may send the indication. The ML models may include ML models that the UE may execute to improve communication efficiency when communicating via the network node.
[0075] The UE may indicate a value of a performance parameter for each of the ML models when used alone. However, the value associated with the first ML model may change based at least in part on other ML models used in combination with the first ML model.
[0076] The ML models used in the combination of ML models may compete for resources of the UE, such as central processing unit (CPU) resources, neural processing unit (NPU) resources, graphics processing unit (GPU) resources, memory resources, and / or input / output (I / O) resources of the UE, etc. For example, the ML models may share and / or compete for special resources other than general resources, such as special hardware acceleration modules (e.g., fast Fourier transform (FFT) modules), etc.
[0077] Additionally or alternatively, the multiple ML models used in combination may target the same or related network functions, such as a first ML model that controls radio resource management (RRM) measurements and a second ML model that controls cell reselection. In this case, the UE may consume resources for the first ML model and the second ML model (which may be unnecessary and / or duplicated), and may consume resources that could have been used when applying another ML model. In addition, the multiple ML models used in combination may have dependencies, such as the output of the first ML model being used as the input of the second ML model.
[0078] In some examples, the first ML model may include a CSI report ML model for determining a subband set for reporting. Accurate selection of the subband set using the first ML model may improve communication efficiency and / or reduce overhead. However, the second ML model consumes resources (e.g., computing resources and / or memory resources, etc.) required by the first ML model, and the UE may not be able to identify the subband set before the CSI report becomes outdated and / or obsolete.
[0079] In these cases, among other examples, the reported values of the performance parameters of each of the ML models when used alone may not be representative of the performance parameters of the ML models when used in the model combination of the ML models. For example, the performance of the ML models when used in the model combination may be degraded and / or reduced to a point where use of the ML models reduces communication efficiency and / or consumes unnecessary power, communication, network, and / or computing resources based at least in part on, for example, latency in generating outputs of the ML models.
[0080] As indicated above, Figure 4 are provided as examples. Other examples can be found in the Figure 4 The content described is different.
[0081] In some aspects described herein, a UE may indicate performance parameters of an ML model when used in different model combinations. For example, a UE may send capability information indicating support for one or more model combinations of ML models, wherein the capability information indicates one or more performance parameters of an ML model in the ML model relative to a model combination including the ML model in the one or more model combinations. In some aspects, a mapping between a model identification and an indication of a performance category is not fixed, but is based at least in part on the combination in which the associated ML model is used. In some aspects, a performance parameter may be based at least in part on an ML model competing for resources of the UE (such as CPU resources, NPU resources, GPU resources, memory resources, and / or I / O resources of the UE, etc.).
[0082] In some aspects, the UE may indicate supported model combinations and / or may indicate unsupported model combinations (eg, in the same or separate communication as an indication of performance parameters).
[0083] The UE may send capability information including a list of model combinations (e.g., indicating combinations of ML models that the UE may support). The capability information may also include information for a first model combination indicating a first ML model identifier and an associated performance indication (e.g., a model performance category) of the first ML model when used in the first model combination, and a second ML model identifier and an associated performance indication of the second ML model when used in the first model combination. The information for the first model combination may include an indication of the performance of any number of ML models supported by the UE for combination. The capability information may also include information for a second model combination indicating a third ML model identifier (e.g., the same ML model identifier as the first ML model identifier or the second ML model identifier, or an additional ML model identifier) and an associated performance indication of the third ML model when used in the second model combination, and a fourth ML model identifier (e.g., the same ML model identifier as the first ML model identifier or the second ML model identifier, or an additional ML model identifier) and an associated performance indication of the fourth ML model when used in the first model combination. In some aspects, the ML models may have different indications of performance when used in different model combinations.
[0084] In some aspects, an indication of performance may be mapped to a set of performance metrics and / or values. For example, an indication of performance may include an indication of a performance category that is mapped to a value of a performance parameter related to, for example, an AI engine, memory, quantization (e.g., granularity of model output), and / or latency in generating output. The value of the performance parameter may be an explicit value and / or may be relative to the value of the performance parameter when the ML model is used independently (e.g., not combined with other ML models). In some aspects, the mapping may be defined in a table or other data storage structure. The mapping may be defined in a communication protocol, a communication standard, and / or via bilateral or multilateral coordination. For example, a UE vendor and an infrastructure vendor may share the same understanding of the meaning of each performance category and / or other indications of performance.
[0085] In some aspects, the UE may send capability information in an RRC message or other message type indicating capabilities. The UE may report capability information via a portion of UE radio capabilities (e.g., received by a base station and forwarded to a core network network node), UE core network capabilities (e.g., received by a core network network node and forwarded to a network node), and / or UE ML capabilities (e.g., newly indicated or combined with UE radio capabilities or core network capabilities information). In some aspects, based at least in part on the UE reporting capability information associated with the model combination to the first network node, the first network node may forward the capability information to the second network node and / or the additional network node.
[0086] Based at least in part on the performance parameter of the UE-indicated ML model when used in combination with other ML models, the UE may provide an indication of the performance of the ML model to the network node with improved accuracy. In this way, the network node may configure the UE to use the combination of ML models based at least in part on the improved accuracy, such that the performance of the ML model may be sufficient to improve communication efficiency and / or save power, communication, network, and / or computing resources that might otherwise be consumed in attempting to use additional ML models with further reduced performance.
[0087] Figure 5 is a diagram of an example 500 associated with combined parameters for an ML model according to the present disclosure. Figure 5 As shown, one or more network nodes (e.g., base station 110, core network node, CU, DU, and / or RU) may communicate with a UE (e.g., UE 120). In some aspects, the network node and the UE may be part of a wireless network (e.g., wireless network 100). The UE and the network node may be in communication with each other. Figure 5 The wireless connection has been established before the operation shown in .
[0088] As indicated by reference numeral 505, a network node of the one or more network nodes may send configuration information, and the UE may receive the configuration information. In some aspects, the UE may receive the configuration information via one or more of RRC signaling, one or more medium access control (MAC) control elements (CEs), and / or downlink control information (DCI), etc. In some aspects, the configuration information may include an indication of one or more configuration parameters for the UE to select (e.g., known to the UE and / or previously indicated by a network node or other network device) and / or explicit configuration information for the UE to configure the UE, etc.
[0089] In some aspects, the configuration information may indicate that the UE will send capability information for using ML models. In some aspects, the configuration information may indicate that the UE will send an indication of the maximum number of ML models that can be combined (e.g., configured for use during communication). Additionally or alternatively, the configuration information may indicate that the UE will send an indication of supported and / or unsupported model combinations. In some aspects, the configuration information may indicate a configuration for the UE to send one or more indications of capability information. For example, the configuration information may indicate that the UE will send capability information via RRC communications. In some aspects, the configuration information may indicate that the UE will send capability information and the same communication or a different communication as the indication of the maximum number of ML models that can be combined and / or the indication of supported and / or unsupported model combinations.
[0090] In some aspects, the configuration information may indicate that the capability information will indicate one or more performance parameters for a respective ML model based at least in part on a model combination of additional ML models used with the respective ML model. In some aspects, the configuration information may indicate that the capability information will indicate a model combination and one or more performance parameters (e.g., values of the one or more performance parameters) for one or more ML models included in the model combination, wherein the one or more performance parameters are based at least in part on other ML models used in the model combination. For example, the configuration information may indicate that the capability information will indicate a performance parameter for each of the model combinations that include the ML model. The ML models may have different performance parameters for different model combinations.
[0091] The UE may configure itself based at least in part on the configuration information.In some aspects, the UE may be configured to perform one or more operations described herein based at least in part on the configuration information.
[0092] As indicated by reference numeral 510, the UE may identify one or more model combinations of ML models supported by the UE and / or not supported by the UE (e.g., ML models associated with communication with the same network node, additional network node, and / or application server). For example, the UE may identify support for multiple ML models (e.g., communication-based ML models) based at least in part on overlap of resource requirements (e.g., primarily using different processing units), based at least in part on total resource requirements supported by the UE (e.g., based at least in part on hardware and / or software of the UE), based at least in part on target network functionality, and / or based at least in part on dependencies, etc.
[0093] In some aspects, the ML models may include one or more decision tree models, one or more decision forest models, one or more convolutional neural network models, one or more cluster models, one or more linear regression models, one or more feed-forward neural network models, and / or one or more recursive neural network models, etc. A combination of models may include one or more types of ML models.
[0094] As indicated by reference numeral 515, the UE may determine one or more performance parameters of the ML model relative to one or more model combinations. For example, the UE may determine a first set of one or more performance parameters associated with the ML model (e.g., AI engine performance, memory performance, quantization (e.g., granularity of model outputs), and / or latency in generating outputs) based at least in part on the ML model being in a first model combination. Additionally or alternatively, the UE may determine a second set of one or more performance parameters associated with the ML model based at least in part on the ML model being in a second model combination (e.g., a combination of the ML model with a set of one or more ML models different from the first model combination).
[0095] As indicated by reference numeral 520, the UE may send capability information indicating support for one or more model combinations and / or one or more performance parameters of an ML model (e.g., an ML model for communication) relative to the one or more model combinations, and the network node may receive the capability information. For example, the capability information may indicate corresponding performance parameters of the corresponding ML model when used in the model combination. In some aspects, the capability information may indicate a first performance parameter for a first ML model in the model combination, a second performance parameter for a second ML model in the model combination, and / or a third performance parameter for a third ML model in the model combination, etc. In some aspects, the performance parameters for the ML models may be different when indicated for different model combinations.
[0096] In some aspects, the capability information may indicate support for one or more model combinations based at least in part on including an indication of supported model combinations and / or an indication of unsupported model combinations. In some aspects, the UE may send the capability information in a capability report. In some aspects, the capability information includes or is included in UE radio capability information, ML capability information, and / or core network capability information. In some aspects, the UE may send the capability information to the first network node for forwarding to the second network node.
[0097] In some aspects, the model combination can be based at least in part on a first set of ML models associated with a RAN network node, a second set of ML models associated with a core network network node, and / or a third set of ML models associated with an application server. In some aspects, each ML model in the model combination is associated with only one of a RAN network node, a core network network node, or an application server.
[0098] In some aspects, the UE may indicate one or more performance parameters based at least in part on including one or more indications. For example, the one or more indications may include a hash of a value of a performance metric set of the ML model relative to the model combination, an indicator mapped to the value of the performance metric set (e.g., a mapping indicator), or a performance category of the ML model relative to the model combination. In some aspects, the indicator may be mapped to the value of the performance metric set based at least in part on a communication protocol and / or based at least in part on a definition of the UE and / or a network node configured to communicate with the UE. In some aspects, the indicator may be mapped to the value of the performance metric set based at least in part on a definition based at least in part on a core network node and / or a RAN network node (e.g., associated with the ML model).
[0099] In some aspects, the one or more performance parameters indicate values of processing resources available for application to the ML model when the ML model is used in a model combination, memory resources available for application to the ML model when the ML model is used in a model combination, quantization of the ML model when the ML model is used in a model combination, and / or latency when the associated ML model is used in a model combination.
[0100] In some aspects, the UE may send capability information via an RRC message or another type of communication. In some aspects, the RRC message or another type of communication may include an indication of the UE to a set of one or more model combinations that are supported or not supported. In some aspects, the RRC message or another type of communication may include an indication of the maximum number of supported ML models to be included in the model combination. In some aspects, the RRC message or another type of communication may include an indication of a first model combination, the indication including a first indication of model parameters associated with a first ML model in the model combination. The indication of the first model combination may include a second indication of model parameters associated with a second ML model in the model combination. In some aspects, the RRC message or another type of communication may include an indication of a second model combination, the indication including a third indication of model parameters associated with a third ML model in the second model combination, and a fourth indication of model parameters associated with a fourth ML model in the second model combination. In some aspects, the first ML model or the second ML model may be the same ML model as the third ML model or the fourth ML model, wherein the same ML model has different indications of model parameters based at least in part on using the same ML model in different model combinations. For example, the first ML model and the third ML model may be the same ML model, the first indication of a model parameter may be associated with a first set of one or more values, the third indication of the model parameter may be associated with a second set of one or more values, and the first set of one or more values may be different from the second set of one or more values.
[0101] As indicated by reference numeral 525, the network node may forward the capability information to an additional network node (e.g., a core network node or a RAN network node) of the one or more network nodes. For example, the network node may forward the capability information to the additional network node based at least in part on the additional network node being associated with an ML model of the one or more model combinations. For example, the core network node may control, support, and / or be affected by operations associated with an ML model for higher layer communications (e.g., L3 or above). Similarly, the RAN network node may control, support, and / or be affected by operations associated with an ML model for lower layer communications (e.g., L2 or below).
[0102] In some aspects, the network node may modify the capability information (e.g., add or remove a portion of the capability information) before forwarding the capability information to the additional network node. For example, a first portion of the capability information may be intended for the network node, and a second portion of the capability information may be intended for the additional network node (e.g., a core network network node or an additional RAN network node). The network node may forward only the portion of the capability information that is intended for the additional network node (e.g., via a target identification and / or based at least in part on the type of the capability information).
[0103] As indicated by reference numeral 530, the network node may identify a model combination for the UE to use for communication. The network node may identify the model combination based at least in part on the capability information. In some aspects, the network node may identify the model combination based at least in part on one or more performance parameters of the ML models in the model combination to optimize communication efficiency and / or power consumption.
[0104] For example, the network node may identify a first model combination that would provide the highest amount of communication efficiency if the UE is able to provide full resources for each ML model in the first model combination. However, the first model combination may not provide the highest amount of communication efficiency based at least in part on the UE indicating that one or more of the ML models in the first model combination will have reduced available resources (e.g., based at least in part on ML models competing for resources). For example, a second model combination may include a different set of ML models that may not be as efficient as the set of ML models in the first model combination if both have full available resources, where the UE indicates an amount of available resources for the models in the second model combination that would cause the second model combination to provide a higher amount of communication efficiency than the first model combination. In this way, the network node may select a model combination based at least in part on the resources available for the ML models in the different model combinations and the amount of communication efficiency obtained by using the ML models that have resources available for the ML models.
[0105] As indicated by reference numeral 535, the UE may receive an indication of a model combination, and the network node may send the indication. For example, the UE may receive an indication of a model combination based at least in part on the capability information. In this manner, the UE may receive an indication to use one or more of the ML models based at least in part on the capability information. For example, the UE may receive an indication to use an ML model belonging to the indicated model combination.
[0106] As shown by reference numeral 540, the UE and the network node may communicate based at least in part on the ML model in the model combination. In some aspects, the UE may configure wireless communications (e.g., RAN configuration) of the UE using the selected model combination. For example, based at least in part on the selected model combination, the UE may configure connection operations (e.g., random access channel (RACH) or physical RACH (PRACH) configurations), traffic management operations, timing synchronization operations, measurement operations, reporting operations (e.g., channel state information (CSI) reporting), reference signal configurations, handover operations, and / or resource configuration operations, among others.
[0107] In some aspects, communicating based at least in part on the ML model may include applying the ML model in the model combination to improve communication efficiency. The UE may apply the ML model set to one or more network functions, such as RRM measurement, CSI reporting, and / or cell reselection, etc.
[0108] Based at least in part on the performance parameter of the UE-indicated ML model when used in combination with other ML models, the UE may provide an indication of the performance of the ML model to the network node with improved accuracy. In this way, the network node may configure the UE to use the combination of ML models based at least in part on the improved accuracy, such that the performance of the ML model may be sufficient to improve communication efficiency and / or save power, communication, network, and / or computing resources that might otherwise be consumed in attempting to use additional ML models with further reduced performance.
[0109] As indicated above, Figure 5 are provided as examples. Other examples can be found in the Figure 5 The content described is different. For example, combined with Figure 5 One or more of the described operations may be omitted or performed by an additional node (eg, a network node or a UE).
[0110] Figure 6 is a diagram of an example 600 associated with a combination of parameters for an ML model according to the present disclosure. Figure 6 , one or more network nodes (e.g., base station 110, core network node, CU, DU, and / or RU) may communicate with a UE (e.g., UE 120). In some aspects, the network node and the UE may be part of a wireless network (e.g., wireless network 100). The UE and the network node may be in Figure 6 The wireless connection has been established before the operation shown in .
[0111] like Figure 6As shown, an indication 605 of a performance parameter (e.g., a performance category) for a combination of ML models may be based at least in part on a mapping of the indication 605 of the performance parameter to a parameter set 610 and a set of performance metric values 615 for the parameter set. For example, an indication associated with performance indicator A (e.g., performance category A) when a first ML model is in a first model combination indicates that when the first ML model is used in the first model combination, the first ML model is expected to be executed using a performance value 615 of the parameter set 610 associated with performance indicator A (e.g., a performance value in the same row as indicator A). Similarly, an indication associated with performance indicator C when the first ML model is in a second model combination indicates that when the first ML model is used in the second model combination, the first ML model is expected to be executed using a performance value 615 of the parameter set 610 associated with performance indicator C (e.g., a performance value in the same row as indicator C). Other performance indicators map to other combinations of performance values and may be associated with other ML models in one or more model combinations, not just the first ML model.
[0112] In some aspects, the performance parameter 605 may similarly comprise a hash of the values of the parameter set 610. In this manner, the indication of the performance parameter 605 may use a reduced amount of overhead when compared to the amount of overhead required to explicitly indicate each of the parameter set 610 without hashing. In some aspects, the indicator may comprise a bit value associated with the indicator, or may comprise a bitmap having a bit associated with each candidate indicator.
[0113] In some aspects, the capability information may indicate a performance indicator for the ML models within each indicated model combination. For example, the capability information may include an indication of a set of model combinations indicating all reported model combinations. The elements of the set of model combinations may be model combinations supported by the UE. For a model combination (e.g., for each model combination), the UE may indicate a list of ML models in the model combination and one or more model parameters. The one or more model parameters may include a model identification for the included ML models and a performance indicator for the ML models (e.g., a performance indicator for each of the included ML models).
[0114] As indicated above, Figure 6 are provided as examples. Other examples can be found in the Figure 6 The content described is different.
[0115] Figure 7 is a diagram of an example 700 associated with a combination of parameters for an ML model according to the present disclosure. Figure 7, one or more network nodes (e.g., base station 110, core network node, CU, DU, and / or RU) may communicate with a UE (e.g., UE 120). In some aspects, the network node and the UE may be part of a wireless network (e.g., wireless network 100). The UE and the network node may be in Figure 7 The wireless connection has been established before the operation shown in .
[0116] like Figure 7 As shown, the capability information may indicate that the UE supports model combinations of up to four ML models. The model combination list may include a first model combination 705, a second model combination 710, a third model combination 715, a fourth model combination 720, a fifth model combination 725, and / or a sixth model combination 730, etc. In some aspects, the model combinations may have a uniform number of model identifiers (IDs), or the model combinations may have different numbers of model IDs. For example, each model combination may have the same number of model IDs, or some model combinations may have different numbers of model IDs.
[0117] The capability information may indicate that when used in the first combination 705, model identification (ID) 1 (e.g., the ML model associated with model ID 1) has performance indicator 1, model ID 2 has performance indicator 2, model ID 3 has performance indicator 3, and model ID 4 has performance indicator 4. The performance indicators may be mapped to the same or different performance parameters. For example, performance indicator 1 and performance indicator 3 may be Figure 6 Indicator A in, performance indicator 2 can be Figure 6 Indicator B in and / or performance indicator 4 may be Figure 6 Indicators C in , each of which is associated with a set of one or more performance values.
[0118] Similarly, the second model combination 710, the third model combination 715, the fourth model combination 720, the fifth model combination 725, and / or the sixth model combination 730 may indicate a model ID and an associated indicator for the model combination. In some aspects, the model ID has an associated indicator that, when used in different combinations, maps to different performance values based at least in part on other ML models used in the different combinations. In some aspects, the model ID has an associated indicator that, when used in different combinations, maps to the same performance value based at least in part on other ML models used in the different combinations.
[0119] As indicated above, Figure 7 are provided as examples. Other examples can be found in the Figure 7 The content described is different.
[0120] Figure 8 8 is a diagram illustrating an example process 800 performed, for example, by a UE according to the present disclosure. Example process 800 is an example in which a UE (eg, UE 120) performs operations associated with a combined performance indicator for an ML model.
[0121] like Figure 8 As shown, in some aspects, process 800 may include sending capability information indicating support for one or more model combinations of ML models, wherein the capability information further indicates one or more performance parameters of the ML model in the ML model relative to the model combination of the one or more model combinations that includes the ML model (block 810). Fig.10 The communication manager 140 and / or the sending component 1004 depicted in the figure can send capability information indicating support for one or more model combinations of the ML model, wherein the capability information further indicates one or more performance parameters of the ML model in the ML model relative to the model combination including the ML model in the one or more model combinations, as described above.
[0122] like Figure 8 As further shown, in some aspects, process 800 may include receiving one or more indications to use one or more of the ML models based at least in part on the capability information (block 820). Fig.10 The communications manager 140 and / or receiving component 1002 depicted in FIG. 1004 may receive one or more indications to use one or more of the ML models based at least in part on the capability information, as described above.
[0123] Process 800 may include additional aspects, such as any single aspect or any combination of aspects described below and / or in conjunction with one or more other processes described elsewhere herein.
[0124] In a first aspect, a performance parameter in the one or more performance parameters comprises one or more of: a hash of a value of a set of performance metrics of the ML model relative to the model combination, an indicator mapped to a value of the set of performance metrics, or a performance category of the ML model relative to the model combination.
[0125] In a second aspect, alone or in combination with the first aspect, the indicator is mapped to a value of a set of performance metrics based at least in part on one or more of a communication protocol, or a definition based at least in part on one or more of a UE or a network node configured to communicate with the UE.
[0126] In a third aspect, alone or in combination with one or more of the first and second aspects, the one or more performance parameters indicate values of one or more of: processing resources available for application to the ML model, memory resources available for application to the ML model, quantization of the ML model, or latency when using the associated ML model.
[0127] In a fourth aspect, alone or in combination with one or more of the first to third aspects, sending the capability information includes sending the capability information via an RRC message.
[0128] In a fifth aspect, alone or in combination with one or more of the first to fourth aspects, the RRC message includes one or more of the following: an indication of a set of one or more model combinations; an indication of a model combination, the indication comprising: a first indication of model parameters associated with an ML model in the model combination, wherein the ML model is a first ML model; and a second indication of model parameters associated with a second ML model in the model combination; or an indication of an additional model combination, the indication comprising: a third indication of model parameters associated with a third ML model in the additional model combination, and a fourth indication of model parameters associated with a fourth ML model in the additional model combination.
[0129] In a sixth aspect, either alone or in combination with one or more of the first to fifth aspects, the first ML model and the third ML model are the same ML model, wherein a first indication of a model parameter is associated with a first set of one or more values, wherein a third indication of the model parameter is associated with a second set of one or more values, and wherein the first set of one or more values is different from the second set of one or more values.
[0130] In a seventh aspect, alone or in combination with one or more of the first to sixth aspects, the capability information comprises one or more of the following: UE radio capability information, ML capability information, or core network capability information.
[0131] In an eighth aspect, alone or in combination with one or more of the first to seventh aspects, sending the capability information includes: sending the capability information to the first network node to forward to the second network node.
[0132] In a ninth aspect, alone or in combination with one or more of aspects one to eight, one or more model combinations of ML models are based at least in part on one or more of: a first set of ML models associated with a radio access network (RAN) network node, a second set of ML models associated with a core network (CN) network node, or a third set of ML models associated with an application server.
[0133] In a tenth aspect, alone or in combination with one or more of the first to ninth aspects, process 800 includes sending one or more of an indication of a supported model combination or an indication of an unsupported model combination.
[0134] although Figure 8 An example block diagram of process 800 is shown, but in some aspects, process 800 may include Figure 8 Additional blocks, fewer blocks, different blocks, or blocks arranged in a different manner than those depicted in the process 800. Additionally or alternatively, two or more of the blocks in the process 800 may be performed in parallel.
[0135] Fig. 9 900 is a diagram illustrating an example process 900 performed, for example, by a network node according to the present disclosure. The example process 900 is a diagram in which a network node (e.g., Figure 5 An example of a base station 110 and / or network node) performing operations associated with a combined performance indicator for an ML model.
[0136] like Fig. 9 As shown, in some aspects, process 900 may include receiving capability information indicating support of one or more model combinations of ML models by the UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination including the ML model in the one or more model combinations (block 910). For example, a network node (e.g., using Fig.11 The communication manager 150 and / or receiving component 1102 depicted in the figure may receive capability information indicating support of one or more model combinations of ML models by the UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination including the ML model in the one or more model combinations, as described above.
[0137] like Fig. 9 As further shown, in some aspects, process 900 may include sending one or more indications to use one or more of the ML models based at least in part on the capability information (block 920). Fig.11 The communication manager 150 and / or the sending component 1104 depicted in FIG. 100 may send one or more indications to use one or more of the ML models based at least in part on the capability information, as described above.
[0138] Process 900 may include additional aspects, such as any single aspect or any combination of aspects described below and / or in conjunction with one or more other processes described elsewhere herein.
[0139] In a first aspect, a performance parameter in the one or more performance parameters comprises one or more of: a hash of a value of a set of performance metrics of the ML model relative to the model combination, an indicator mapped to a value of the set of performance metrics, or a performance category of the ML model relative to the model combination.
[0140] In a second aspect, alone or in combination with the first aspect, the indicator is mapped to a value of a set of performance metrics based at least in part on one or more of a communication protocol, or a definition based at least in part on one or more of a UE or a network node configured to communicate with the UE.
[0141] In a third aspect, alone or in combination with one or more of the first and second aspects, the one or more performance parameters indicate values of one or more of: processing resources available for application to the ML model, memory resources available for application to the ML model, quantization of the ML model, or latency when using the associated ML model.
[0142] In a fourth aspect, alone or in combination with one or more of the first to third aspects, receiving the capability information comprises receiving the capability information via an RRC message.
[0143] In a fifth aspect, alone or in combination with one or more of the first to fourth aspects, the RRC message includes one or more of the following: an indication of a set of one or more model combinations; an indication of a model combination, the indication comprising: a first indication of model parameters associated with an ML model in the model combination, wherein the ML model is a first ML model; and a second indication of model parameters associated with a second ML model in the model combination; or an indication of an additional model combination, the indication comprising: a third indication of model parameters associated with a third ML model in the additional model combination, and a fourth indication of model parameters associated with a fourth ML model in the additional model combination.
[0144] In a sixth aspect, either alone or in combination with one or more of the first to fifth aspects, the first ML model and the third ML model are the same ML model, wherein a first indication of a model parameter is associated with a first set of one or more values, wherein a third indication of the model parameter is associated with a second set of one or more values, and wherein the first set of one or more values is different from the second set of one or more values.
[0145] In a seventh aspect, alone or in combination with one or more of the first to sixth aspects, the capability information comprises one or more of the following: UE radio capability information, ML capability information, or core network capability information.
[0146] In an eighth aspect, alone or in combination with one or more of the first to seventh aspects, process 900 includes forwarding at least a portion of the capability information to an additional network node.
[0147] In a ninth aspect, alone or in combination with one or more of aspects one to eight, one or more model combinations of ML models are based at least in part on one or more of: a first set of ML models associated with a RAN network node, a second set of ML models associated with a CN network node, or a third set of ML models associated with an application server.
[0148] In a tenth aspect, alone or in combination with one or more of the first to ninth aspects, process 900 includes receiving one or more of an indication of a supported model combination or an indication of an unsupported model combination.
[0149] although Fig. 9 An example block diagram of process 900 is shown, but in some aspects, process 900 may include Fig. 9 Additional blocks, fewer blocks, different blocks, or blocks arranged in a different manner than those depicted in the process 900. Additionally or alternatively, two or more of the blocks in the process 900 may be performed in parallel.
[0150] Fig.10 1 is a diagram of an example apparatus 1000 for wireless communication. Apparatus 1000 may be a UE, or a UE may include apparatus 1000. In some aspects, apparatus 1000 includes a receiving component 1002 and a transmitting component 1004 that may communicate with each other (e.g., via one or more buses and / or one or more other components). As shown, apparatus 1000 may communicate with another apparatus 1006 (such as a UE, a base station, or another wireless communication device) using receiving component 1002 and transmitting component 1004. As further shown, apparatus 1000 may include a communication manager 1008 (e.g., communication manager 140). Communication manager 1008 may send control signaling to transmitting component 1004 and / or may receive control signaling via receiving component 1002 to control communication of apparatus 1000.
[0151] In some aspects, the apparatus 1000 may be configured to perform Figures 5 to 7 Additionally or alternatively, the apparatus 1000 may be configured to perform one or more processes described herein, such as Figure 8 The process 800. In some aspects, Fig.10 The apparatus 1000 and / or one or more components shown in FIG. 1 may include a combination of Figure 2Additionally or alternatively, Fig.10 One or more of the components shown in the figure may be combined with Figure 2 Additionally or alternatively, one or more of the set of components may be implemented at least in part as software stored in a memory. For example, a component (or a portion of a component) may be implemented as instructions or codes stored in a non-transitory computer-readable medium and executable by a controller or processor to perform the function or operation of the component.
[0152] The receiving component 1002 may receive communications, such as reference signals, control information, data communications, or combinations thereof, from the device 1006. The receiving component 1002 may provide the received communications to one or more other components of the device 1000. In some aspects, the receiving component 1002 may perform signal processing (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, demapping, equalization, interference cancellation, or decoding, etc.) on the received communications and may provide the processed signals to one or more other components of the device 1000. In some aspects, the receiving component 1002 may include combining Figure 2 One or more antennas, modems, demodulators, MIMO detectors, receive processors, controllers / processors, memories, or combinations thereof of the described UE.
[0153] The transmitting component 1004 may transmit communications, such as reference signals, control information, data communications, or combinations thereof, to the device 1006. In some aspects, one or more other components of the device 1000 may generate communications and may provide the generated communications to the transmitting component 1004 for transmission to the device 1006. In some aspects, the transmitting component 1004 may perform signal processing (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, etc.) on the generated communications and may transmit the processed signals to the device 1006. In some aspects, the transmitting component 1004 may include combining Figure 2 One or more antennas, modems, modulators, transmit MIMO processors, transmit processors, controllers / processors, memories, or combinations thereof of the described UE. In some aspects, the transmit component 1004 can be co-located with the receive component 1002 in a transceiver.
[0154] The sending component 1004 can send capability information indicating support for one or more model combinations of the ML models, wherein the capability information further indicates one or more performance parameters of the ML model in the ML model relative to the model combination including the ML model in the one or more model combinations. The receiving component 1002 can receive one or more indications to use one or more ML models in the ML model based at least in part on the capability information.
[0155] The sending component 1004 can send one or more of an indication of a supported model combination or an indication of an unsupported model combination.
[0156] Fig.10 The number and arrangement of components shown in are provided as examples. In practice, there may be Fig.10 Additional components, fewer components, different components, or components arranged in a different manner than those shown in FIG. Fig.10 Two or more components shown in may be implemented in a single component, or Fig.10 The single component shown in can be implemented as multiple distributed components. Additionally or alternatively, Fig.10 A set (one or more) of components shown in the figure may perform the operations described as being performed by Fig.10 One or more functions performed by another set of components shown in FIG.
[0157] Fig.11 1 is a diagram of an example apparatus 1100 for wireless communication. Apparatus 1100 may be a network node, or a network node may include apparatus 1100. In some aspects, apparatus 1100 includes a receiving component 1102 and a transmitting component 1104 that may communicate with each other (e.g., via one or more buses and / or one or more other components). As shown, apparatus 1100 may communicate with another apparatus 1106 (such as a UE, a base station, or another wireless communication device) using receiving component 1102 and transmitting component 1104. As further shown, apparatus 1100 may include a communication manager 1108 (e.g., communication manager 150). Communication manager 1108 may send control signaling to transmitting component 1104 and / or may receive control signaling via receiving component 1102 to control communication of apparatus 1100.
[0158] In some aspects, the apparatus 1100 may be configured to perform Figures 5 to 7 Additionally or alternatively, the apparatus 1100 may be configured to perform one or more processes described herein, such as Fig. 9 The process 900. In some aspects, Fig.11 The apparatus 1100 and / or one or more components shown in FIG. 1 may include a combination of Figure 2 Additionally or alternatively, Fig.11 One or more of the components shown in the figure may be combined with Figure 2Additionally or alternatively, one or more of the set of components may be implemented at least in part as software stored in a memory. For example, a component (or a portion of a component) may be implemented as instructions or codes stored in a non-transitory computer-readable medium and executable by a controller or processor to perform the function or operation of the component.
[0159] The receiving component 1102 may receive communications from the device 1106, such as reference signals, control information, data communications, or combinations thereof. The receiving component 1102 may provide the received communications to one or more other components of the device 1100. In some aspects, the receiving component 1102 may perform signal processing (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, demapping, equalization, interference cancellation, or decoding, etc.) on the received communications and may provide the processed signals to one or more other components of the device 1100. In some aspects, the receiving component 1102 may include combining Figure 2 One or more antennas, modems, demodulators, MIMO detectors, receive processors, controllers / processors, memories, or combinations thereof, of the described network nodes.
[0160] The transmitting component 1104 may transmit communications, such as reference signals, control information, data communications, or combinations thereof, to the device 1106. In some aspects, one or more other components of the device 1100 may generate communications and may provide the generated communications to the transmitting component 1104 for transmission to the device 1106. In some aspects, the transmitting component 1104 may perform signal processing (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, etc.) on the generated communications and may transmit the processed signals to the device 1106. In some aspects, the transmitting component 1104 may include combining Figure 2 One or more antennas, modems, modulators, transmit MIMO processors, transmit processors, controllers / processors, memories, or combinations thereof of the described network nodes. In some aspects, the transmit component 1104 can be co-located with the receive component 1102 in a transceiver.
[0161] The receiving component 1102 can receive capability information indicating support of one or more model combinations of ML models by the UE, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination in the one or more model combinations including the ML model. The sending component 1104 can send one or more indications of using one or more ML models in the ML models based at least in part on the capability information.
[0162] Communications manager 1108 and / or sending component 1104 can forward at least a portion of the capability information to additional network nodes.
[0163] Receiving component 1102 can receive one or more of an indication of a supported model combination or an indication of an unsupported model combination.
[0164] Fig.11 The number and arrangement of components shown in are provided as examples. In practice, there may be Fig.11 Additional components, fewer components, different components, or components arranged in a different manner than those shown in FIG. Fig.11 Two or more components shown in may be implemented in a single component, or Fig.11 The single component shown in can be implemented as multiple distributed components. Additionally or alternatively, Fig.11 A set (one or more) of components shown in the figure may perform the operations described as being performed by Fig.11 One or more functions performed by another set of components shown in FIG.
[0165] The following provides an overview of some aspects of the disclosure:
[0166] Aspect 1: A method of wireless communication performed by a user equipment (UE), comprising: sending capability information indicating support for one or more model combinations of machine learning (ML) models, wherein the capability information further indicates one or more performance parameters of an ML model in the ML models relative to a model combination including the ML model in the one or more model combinations; and receiving one or more indications of using one or more ML models in the ML models based at least in part on the capability information.
[0167] Aspect 2: A method according to Aspect 1, wherein the performance parameter among the one or more performance parameters comprises one or more of the following: a hash of the value of a performance metric set of the ML model relative to the model combination, an indicator mapped to the value of the performance metric set, or a performance category of the ML model relative to the model combination.
[0168] Aspect 3: A method according to Aspect 2, wherein the indicator is mapped to the value of the performance metric set based at least in part on one or more of the following: a communication protocol, or a definition based at least in part on one or more of the UE or a network node configured to communicate with the UE.
[0169] Aspect 4: A method according to any one of Aspects 1 to 3, wherein the one or more performance parameters indicate values of one or more of the following: processing resources available for application to the ML model, memory resources available for application to the ML model, quantization of the ML model, or latency when using the associated ML model.
[0170] Aspect 5: The method according to any one of aspects 1 to 4, wherein sending the capability information comprises: sending the capability information via a radio resource control (RRC) message.
[0171] Aspect 6: A method according to Aspect 5, wherein the RRC message includes one or more of the following: an indication of a set of the one or more model combinations; an indication of the model combination, the indication including: a first indication of model parameters associated with the ML model in the model combination, wherein the ML model is a first ML model, and a second indication of model parameters associated with the second ML model in the model combination; or an indication of an additional model combination, the indication including: a third indication of model parameters associated with a third ML model in the additional model combination, and a fourth indication of model parameters associated with a fourth ML model in the additional model combination.
[0172] Aspect 7: A method according to Aspect 6, wherein the first ML model and the third ML model are the same ML model, wherein the first indication of the model parameter is associated with a first set of one or more values, wherein the third indication of the model parameter is associated with a second set of one or more values, and wherein the first set of one or more values is different from the second set of one or more values.
[0173] Aspect 8: The method according to any one of aspects 1 to 7, wherein the capability information comprises one or more of the following: UE radio capability information, ML capability information or core network capability information.
[0174] Aspect 9: The method according to any one of aspects 1 to 8, wherein sending the capability information comprises: sending the capability information to a first network node to be forwarded to a second network node.
[0175] Aspect 10: A method according to any one of Aspects 1 to 9, wherein the one or more model combinations of the ML models are based at least in part on one or more of: a first set of ML models associated with a radio access network (RAN) network node, a second set of ML models associated with a core network (CN) network node, or a third set of ML models associated with an application server.
[0176] Aspect 11: The method according to any one of aspects 1 to 10, further comprising sending one or more of the following: an indication of a supported model combination or an indication of an unsupported model combination.
[0177] Aspect 12: A method of wireless communication performed by a network node, comprising: receiving capability information indicating support of a user equipment (UE) for one or more model combinations of machine learning (ML) models, wherein the capability information further indicates one or more performance parameters of an ML model in the one or more ML models relative to a model combination including the ML model in the one or more model combinations; and sending one or more indications of using one or more ML models in the ML models based at least in part on the capability information.
[0178] Aspect 13: A method according to Aspect 12, wherein the performance parameter among the one or more performance parameters comprises one or more of the following: a hash of the value of a performance metric set of the ML model relative to the model combination, an indicator mapped to the value of the performance metric set, or a performance category of the ML model relative to the model combination.
[0179] Aspect 14: A method according to any one of Aspects 12 to 13, wherein the indicator is mapped to the value of the performance metric set based at least in part on one or more of the following: a communication protocol, or a definition based at least in part on one or more of the UE or a network node configured to communicate with the UE.
[0180] Aspect 15: A method according to any one of Aspects 12 to 14, wherein the one or more performance parameters indicate the value of one or more of the following: processing resources available for application to the ML model, memory resources available for application to the ML model, quantization of the ML model, or latency when using the associated ML model.
[0181] Aspect 16: The method according to any one of aspects 12 to 15, wherein receiving the capability information comprises: receiving the capability information via a radio resource control (RRC) message.
[0182] Aspect 17: A method according to Aspect 16, wherein the RRC message includes one or more of the following: an indication of a set of the one or more model combinations; an indication of the model combination, the indication including: a first indication of model parameters associated with the ML model in the model combination, wherein the ML model is a first ML model, and a second indication of model parameters associated with the second ML model in the model combination; or an indication of an additional model combination, the indication including: a third indication of model parameters associated with a third ML model in the additional model combination, and a fourth indication of model parameters associated with a fourth ML model in the additional model combination.
[0183] Aspect 18: A method according to Aspect 17, wherein the first ML model and the third ML model are the same ML model, wherein the first indication of the model parameter is associated with a first set of one or more values, wherein the third indication of the model parameter is associated with a second set of one or more values, and wherein the first set of one or more values is different from the second set of one or more values.
[0184] Aspect 19: The method according to any one of aspects 12 to 18, wherein the capability information comprises one or more of the following: UE radio capability information, ML capability information or core network capability information.
[0185] Aspect 20: The method according to any one of aspects 12 to 19, further comprising: forwarding at least a portion of the capability information to an additional network node.
[0186] Aspect 21: A method according to any one of Aspects 12 to 20, wherein the one or more model combinations of the ML models are based at least in part on one or more of: a first set of the ML models associated with a radio access network (RAN) network node, a second set of the ML models associated with a core network (CN) network node, or a third set of the ML models associated with an application server.
[0187] Aspect 22: The method according to any one of aspects 12 to 21, further comprising receiving one or more of: an indication of a supported model combination or an indication of an unsupported model combination.
[0188] Aspect 23: An apparatus for performing wireless communications at a device, comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform one or more of the methods described in Aspects 1 to 22.
[0189] Aspect 24: A device for wireless communication, comprising: a memory; and one or more processors, the one or more processors coupled to the memory, the one or more processors configured to execute the method according to one or more of aspects 1 to 22.
[0190] Aspect 25: An apparatus for wireless communication, comprising at least one means for performing the method according to one or more of aspects 1 to 22.
[0191] Aspect 26: A non-transitory computer-readable medium storing code for wireless communication, the code comprising instructions executable by a processor to perform the method according to one or more of aspects 1 to 22.
[0192] Aspect 27: A non-transitory computer-readable medium storing an instruction set for wireless communication, the instruction set comprising one or more instructions that, when executed by one or more processors of a device, cause the device to perform one or more of the methods described in aspects 1 to 22.
[0193] The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the aspects to the precise forms disclosed. Modifications and variations may be made in light of the above disclosure or may be acquired from practice of these aspects.
[0194] As used herein, the term "component" is intended to be broadly interpreted as a combination of hardware and / or hardware and software. Whether referred to as software, firmware, middleware, microcode, hardware description language or other names, "software" should be broadly interpreted as meaning instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, processes and / or functions, etc. As used herein, a "processor" is implemented in a combination of hardware and / or hardware and software. It will be apparent that the system or method described herein can be implemented by a combination of hardware and / or hardware and software in different forms. The actual dedicated control hardware or software code used to implement these systems and / or methods does not limit various aspects. Therefore, there is no reference to a specific software code herein to describe the operation and behavior of the system and / or method, because it will be understood by those skilled in the art that software and hardware can be designed to implement the system and / or method at least in part based on the description herein.
[0195] As used herein, "satisfying a threshold" may refer to a value being greater than a threshold, greater than or equal to a threshold, less than a threshold, less than or equal to a threshold, equal to a threshold, not equal to a threshold, etc., depending on the context.
[0196] Although the specific combination of features is set forth in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of various aspects. Many of these features can be combined in a manner that is not specifically listed in the claims and / or is not disclosed in the specification. The disclosure of various aspects includes each dependent claim combined with each other claim in the claim set. As used herein, the phrase "at least one of" the list of items refers to any combination of these items (which includes a single member). As an example, "at least one of a, b or c" is intended to cover a, b, c, a+b, a+c, b+c and a+b+c, and any combination with multiple identical elements (e.g., a+a, a+a+a, a+a+b, a+a+c, a+b+b, a+c+c, b+b, b+b+b, b+b+c, c+c and c+c+c, or any other ordering of a, b and c).
[0197] Any element, action or instruction used herein should not be interpreted as key or necessary unless explicitly stated. In addition, as used herein, the articles "one" and "an" are intended to include one or more items and can be used interchangeably with "one or more". In addition, as used herein, the article "said" is intended to include one or more items connected to the article "said", and can be used interchangeably with "one or more". In addition, as used herein, the terms "set" and "group" are intended to include one or more items and can be used interchangeably with "one or more". If you only want to refer to an item, the phrase "only one" or similar terms will be used. Moreover, as used herein, the term "having" etc. is intended to be an open term, which does not limit the elements they modify (for example, "having" A elements can also have B). In addition, the phrase "based on" is intended to mean "based at least in part on", unless explicitly stated otherwise. Furthermore, as used herein, the term "or" when used in a series is intended to be open-ended and may be used interchangeably with "and / or" unless explicitly stated otherwise (eg, if used in conjunction with "either" or "only one of").
Claims
1. A user equipment (UE) for wireless communication, comprising: Memory; and one or more processors coupled to the memory and configured to: sending capability information indicating support for one or more model combinations of machine learning (ML) models, wherein the capability information further indicates one or more performance parameters of an ML model of the ML models relative to a model combination of the one or more model combinations that includes the ML model; and One or more indications to use one or more of the ML models are received based at least in part on the capability information.
2. The UE according to claim 1, wherein the performance parameter in the one or more performance parameters comprises one or more of the following: a hash of the value of the set of performance metrics of the ML model relative to the model combination, an indicator that maps to the value of the set of performance metrics, or The performance category of the ML model relative to the combination of models.
3. The UE of claim 2, wherein the indicator is mapped to the value of the set of performance metrics based at least in part on one or more of: Communication protocol, or Based at least in part on a definition of one or more of the UE or a network node configured to communicate with the UE.
4. The UE according to claim 1, wherein the one or more performance parameters indicate values of one or more of the following: processing resources available for application to the ML model, memory resources available for application to the ML model, quantization of the ML model, or Latency when using the associated ML model.
5. The UE of claim 1 , wherein to send the capability information, the one or more processors are configured to: The capability information is sent via a Radio Resource Control (RRC) message.
6. The UE according to claim 5, wherein the RRC message comprises one or more of the following: an indication of a set of the one or more model combinations; An indication of the model combination, the indication comprising: a first indication of a model parameter associated with the ML model in the combination of models, wherein the ML model is a first ML model, and a second indication of model parameters associated with a second ML model in the combination of models, or An indication of an additional model combination, the indication comprising: a third indication of model parameters associated with a third ML model in the second model combination, and a fourth indication of model parameters associated with a fourth ML model in the additional model combination.
7. The UE according to claim 6, wherein the first ML model and the third ML model are the same ML model, wherein a first indication of the model parameter is associated with a first set of one or more values, wherein the third indication of the model parameter is associated with a second set of one or more values, and Wherein the first set of one or more values is different from the second set of one or more values.
8. The UE according to claim 1, wherein the capability information comprises one or more of the following: UE radio capability information, ML capability information, or Core network capability information.
9. The UE of claim 1, wherein to send the capability information, the one or more processors are configured to: The capability information is sent to the first network node for forwarding to the second network node.
10. The UE of claim 1 , wherein the one or more model combinations of the ML models are based at least in part on one or more of: a first set of said ML models associated with a Radio Access Network (RAN) network node, a second set of said ML models associated with a core network (CN) network node, or A third set of said ML models is associated with an application server.
11. The UE of claim 1 , wherein the one or more processors are further configured to send one or more of: an indication of supported model combinations, or Indication for an unsupported combination of models.
12. A network node for wireless communication, comprising: Memory; and one or more processors coupled to the memory and configured to: receiving capability information indicating support of one or more model combinations of machine learning (ML) models by a user equipment (UE), wherein the capability information further indicates one or more performance parameters of an ML model of the one or more ML models relative to a model combination of the one or more model combinations including the ML model; as well as One or more indications to use one or more of the ML models are sent based at least in part on the capability information.
13. The network node according to claim 12, wherein the performance parameter of the one or more performance parameters comprises one or more of the following: a hash of the values of a set of performance metrics of the ML model relative to the combination of models, an indicator that maps to said value of said set of performance metrics, or The performance category of the ML model relative to the combination of models.
14. The network node of claim 13, wherein the indicator is mapped to the value of the set of performance metrics based at least in part on one or more of: Communication protocol, or Based at least in part on a definition of one or more of the UE or a network node configured to communicate with the UE.
15. The network node of claim 12, wherein the one or more performance parameters indicate values of one or more of the following: processing resources available for application to the ML model, memory resources available for application to the ML model, quantization of the ML model, or Latency when using the associated ML model.
16. The network node of claim 12, wherein to receive the capability information, the one or more processors are configured to: The capability information is sent via a Radio Resource Control (RRC) message.
17. The network node of claim 16, wherein the RRC message comprises one or more of: an indication of a set of the one or more model combinations; An indication of the model combination, the indication comprising: a first indication of a model parameter associated with the ML model in the combination of models, wherein the ML model is a first ML model, and a second indication of model parameters associated with a second ML model in the combination of models, or An indication of an additional model combination, the indication comprising: a third indication of model parameters associated with a third ML model in the second model combination, and a fourth indication of model parameters associated with a fourth ML model in the second combination of models.
18. The network node according to claim 17, wherein the first ML model and the third ML model are the same ML model, wherein a first indication of the model parameter is associated with a first set of one or more values, wherein the third indication of the model parameter is associated with a second set of one or more values, and Wherein the first set of the one or more values is different from the second set of the one or more values.
19. The network node according to claim 12, wherein the capability information comprises one or more of the following: UE radio capability information, ML capability information, or Core network capability information.
20. The network node of claim 12, wherein the one or more processors are further configured to: At least a portion of the capability information is forwarded to an additional network node.
21. The network node of claim 12, wherein the one or more model combinations of the ML models are based at least in part on one or more of: a first set of said ML models associated with a Radio Access Network (RAN) network node, a second set of said ML models associated with a core network (CN) network node, or A third set of said ML models is associated with an application server.
22. The network node of claim 12, wherein the one or more processors are further configured to receive one or more of: an indication of supported model combinations, or Indication for an unsupported combination of models.
23. A method of wireless communication performed by a user equipment (UE), comprising: sending capability information indicating support for one or more model combinations of machine learning (ML) models, wherein the capability information further indicates one or more performance parameters of an ML model of the ML models relative to a model combination of the one or more model combinations that includes the ML model; and One or more indications to use one or more of the ML models are received based at least in part on the capability information.
24. The method of claim 23, wherein the performance parameter of the one or more performance parameters comprises one or more of the following: a hash of the values of a set of performance metrics of the ML model relative to the combination of models, an indicator that maps to said value of said set of performance metrics, or The performance category of the ML model relative to the combination of models.
25. The method of claim 23, wherein sending the capability information comprises: The capability information is sent via a Radio Resource Control (RRC) message.
26. The method of claim 25, wherein the RRC message comprises one or more of: an indication of a set of the one or more model combinations; An indication of the model combination, the indication comprising: a first indication of a model parameter associated with the ML model in the combination of models, wherein the ML model is a first ML model, and a second indication of model parameters associated with a second ML model in the combination of models, or An indication of an additional model combination, the indication comprising: a third indication of model parameters associated with a third ML model in the second model combination, and a fourth indication of model parameters associated with a fourth ML model in the second combination of models.
27. A method of wireless communication performed by a network node, comprising: receiving capability information indicating support of one or more model combinations of machine learning (ML) models by a user equipment (UE), wherein the capability information further indicates one or more performance parameters of an ML model of the one or more ML models relative to a model combination of the one or more model combinations including the ML model; as well as One or more indications to use one or more of the ML models are sent based at least in part on the capability information.
28. The method of claim 27, wherein the performance parameter of the one or more performance parameters comprises one or more of the following: a hash of the value of the set of performance metrics of the ML model relative to the model combination, an indicator that maps to the value of the set of performance metrics, or The performance category of the ML model relative to the combination of models.
29. The method of claim 27, wherein receiving the capability information comprises: The capability information is sent via a Radio Resource Control (RRC) message.
30. The method of claim 29, wherein the RRC message comprises one or more of: an indication of a set of the one or more model combinations; An indication of the model combination, the indication comprising: a first indication of a model parameter associated with the ML model in the combination of models, wherein the ML model is a first ML model, and a second indication of model parameters associated with a second ML model in the combination of models, or An indication of an additional model combination, the indication comprising: a third indication of model parameters associated with a third ML model in the second model combination, and a fourth indication of model parameters associated with a fourth ML model in the second combination of models.
Citation Information
Patent Citations
Performance indicators for combinations of machine learning models
US20230336972A1
