Beam reporting and user equipment capability configuration for artificial intelligence / machine learning base beam management

Abstract

According to an aspect of the present disclosure, a method, a computer readable medium, and an apparatus are provided. The device may be a user device. In one embodiment, a user equipment receives, from a base station, a beam reporting configuration for artificial intelligence / machine learning (AI / ML) base beam management, where the beam reporting configuration includes a set of configuration parameters that enable reporting for the AI / ML base beam management. The user equipment determines a set of beams satisfying a predetermined performance indicator based on the beam reporting configuration. The user equipment reports the set of beams to the base station.

Description

[0001] Cross-references

[0002] This application claims the benefits of U.S. Provisional Application Serial No. 63 / 516,177, entitled “Method and apparatus for beam reporting configuration for AI / ML basic beam management,” filed July 28, 2023, and U.S. Provisional Application Serial No. 63 / 517,620, entitled “Method and apparatus for user equipment capability reporting for AI / ML basic beam,” filed August 4, 2023, both of which are expressly incorporated herein by reference in their entirety. Technical Field

[0003] This disclosure generally relates to wireless communications, and more specifically, to beam reporting techniques for basic beam management of artificial intelligence / machine learning (AI / ML) in wireless communication systems. Background Technology

[0004] The statements in this section provide only background information in relation to this disclosure and may not constitute prior art.

[0005] Wireless communication systems are widely deployed to provide a variety of telecommunications services, such as telephone, video, data, messaging, and broadcasting. Typical wireless communication systems may employ multiple access technologies to support communication with multiple users by sharing available system resources. Examples of these multiple access technologies include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency division multiple access (FDMA) systems, orthogonal frequency division multiple access (OFDMA) systems, single-carrier frequency division multiple access (SC-FDMA) systems, and time division synchronous code division multiple access (TD-SCDMA) systems.

[0006] These multiple access technologies have been adopted in various telecommunications standards to provide a universal protocol enabling different wireless devices to communicate at the city, national, regional, and even global levels. One example of a telecommunications standard is 5G New Radio (NR). 5G NR is part of the ongoing evolution of mobile broadband driven by the Third Generation Partnership Project (3GPP) to meet new requirements related to latency, reliability, security, scalability (e.g., the Internet of Things, IoT), and other requirements. Some forms of 5G NR may be based on the 4G Long Term Evolution (LTE) standard. There is a need for further improvements to 5G NR technology. These improvements may also apply to other multiple access technologies and telecommunications standards that employ them. Summary of the Invention

[0007] The following provides a simplified summary of one or more patterns to provide a basic understanding of them. This summary is not a comprehensive overview of all hypothetical patterns, nor is it intended to identify key or important elements of all patterns, nor to define the scope of any or all patterns. Its sole purpose is to present some concepts of one or more patterns in a simplified form as a prelude to a more detailed description thereafter.

[0008] In one embodiment of this disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus may be a user equipment (UE). The UE receives from a base station a beam reporting configuration for artificial intelligence / machine learning (AI / ML) basic beam management, wherein the beam reporting configuration includes a set of configuration parameters enabling reporting for the AI / ML basic beam management. The UE determines a beam set that meets predetermined performance metrics based on the beam reporting configuration. The UE reports the beam set to the base station.

[0009] To achieve the foregoing and related objectives, the one or more specimens include the features described in detail below and specifically pointed out in the claims. The following description and drawings detail certain illustrative features of the one or more specimens. However, these features merely indicate a few ways in which the principles of the various specimens can be applied, and this description is intended to include all such specimens and their equivalents. Attached Figure Description

[0010] Figure 1 This is a schematic diagram illustrating an example of a wireless communication system and access network.

[0011] Figure 2 It is a schematic diagram illustrating the communication between a base station and user equipment in the access network.

[0012] Figure 3 This section describes an example logical architecture for a distributed access network.

[0013] Figure 4 This describes an example physical architecture for a distributed access network.

[0014] Figure 5 This is a schematic diagram showing an example of a downlink (DL) center slot.

[0015] Figure 6 This is a schematic diagram showing an example of an uplink (UL) center timeslot.

[0016] Figure 7(A) is a schematic diagram illustrating an artificial intelligence / machine learning (AI / ML) model used for spatial and temporal beam prediction.

[0017] Figure 7(B) is a schematic diagram illustrating the temporal state of beam measurement and prediction in AI / ML basic beam management.

[0018] Figure 8(A) illustrates a flowchart 800 for beam reporting in the AI / ML basic beam management system.

[0019] Figure 8(B) presents a flowchart 830 outlining another beam reporting process, which has additional steps that focus on user equipment capability reporting.

[0020] Figure 8(C) presents a flowchart 850 outlining the further beam reporting process, which also includes additional steps focusing on user equipment capability reporting.

[0021] Figure 9 This diagram illustrates a CSI report configuration structure. Detailed Implementation

[0022] The detailed description below, taken with reference to the accompanying drawings, is intended to describe various configurations and is not intended to represent the only configurations in which the concepts described herein can be practiced. The detailed description includes specific details to provide a thorough understanding of the various concepts. However, those skilled in the art will appreciate that these concepts can be practiced without these specific details. In some cases, well-known structures and components are shown in block diagram form to avoid obscuring these concepts.

[0023] Several forms of telecommunications systems will now be described with reference to various devices and methods. These devices and methods will be described in detail below and illustrated in the accompanying drawings by various modules, components, circuits, processes, algorithms, etc. (collectively, "elements"). These elements can be implemented using electronic hardware, computer software, or any combination of both. Whether these elements are implemented as hardware or software depends on the specific application and design constraints imposed on the overall system.

[0024] For example, an element, any part of an element, or any combination of elements can be implemented as a "processing system" containing one or more processors. Examples of processors include microprocessors, microcontrollers, graphics processing units (GPUs), central processing units (CPUs), application processors, digital signal processors (DSPs), reduced instruction set computing (RISC) processors, systems on a chip (SoC), baseband processors, field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other hardware suitable for performing the various functions described in this disclosure. One or more processors in a processing system can execute software. Software should be understood broadly as instructions, instruction sets, code, code segments, program code, programs, subroutines, software components, application programs, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, etc., regardless of whether it is called software, firmware, middleware, microcode, hardware description language, or something else.

[0025] Therefore, in one or more example states, the described functionality can be implemented in hardware, software, or any combination of both. If implemented in software, these functions can be stored or encoded as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media. Storage media can be any available medium that is accessible to a computer. For example, and not limited to, such computer-readable media can include random-access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), optical disk storage, magnetic disk storage, other magnetic storage devices, combinations of computer-readable media of the types described above, or any other medium that can be used to store computer-executable code in the form of instructions or data structures and is accessible to a computer.

[0026] Figure 1 This is a schematic diagram of an example wireless communication system and access network 100. The wireless communication system (also known as a wireless wide area network (WWAN)) includes base station 102, user equipment 104, an evolved packet core (EPC) 160, and another core network 190 (e.g., a 5G core network (5GC)). Base station 102 may include macrocells (high-power cellular base stations) and / or small cells (low-power cellular base stations). Macrocells include base stations. Small cells include home cells, microcells, and picocells.

[0027] Base station 102 configured as 4G LTE (collectively referred to as Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN)) can interface with EPC 160 via backhaul link 132 (e.g., SI interface). Base station 102 configured as 5G NR (collectively referred to as Next Generation RAN (NG-RAN)) can interface with core network 190 via backhaul link 184. In addition to other functions, base station 102 may perform one or more of the following functions: user data transmission, radio channel encryption and decryption, integrity protection, header compression, mobility control functions (e.g., handover, dual connectivity), inter-cell interference coordination, connection setup and release, load balancing, distribution of non-access stratum (NAS) messages, NAS node selection, synchronization, radio access network (RAN) sharing, multimedia broadcast multicast service (MBMS), user and device tracking, RAN information management (RIM), paging, location, and warning message delivery. Base station 102 may communicate with each other directly or indirectly (e.g., via EPC 160 or core network 190) via backhaul link 134 (e.g., X2 interface). Backhaul link 134 may be wired or wireless.

[0028] Base station 102 can wirelessly communicate with user equipment 104. Each base station 102 can provide communication coverage for a corresponding geographic coverage area 110. Overlapping geographic coverage areas 110 may exist. For example, a small cell 102' may have a coverage area 110' that overlaps with the coverage areas 110 of one or more macro base stations 102. A network containing small cells and macro cells may be referred to as a heterogeneous network. Heterogeneous networks may also include Home Evolved Node Bs (HeNBs), which can provide service for a restricted group called a closed subscriber group (CSG). The communication link 120 between base station 102 and user equipment 104 may include uplink (UL) (also known as reverse link) transmission from user equipment 104 to base station 102 and / or downlink (DL) (also known as forward link) transmission from base station 102 to user equipment 104. Communication link 120 may use multiple-input multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming, and / or transmit diversity. The communication link may be carried out via one or more carriers. Base station 102 / user equipment 104 may use each carrier with a spectrum bandwidth up to 7 MHz (e.g., 5, 10, 15, 20, 100, 400 MHz, etc.) to allocate a total of up to Yx MHz (x component carriers) for transmission in each direction in carrier aggregation. Carriers may be adjacent or non-adjacent. Carrier allocation may be asymmetrical between DL and UL patterns (e.g., more or fewer carriers may be allocated to DL than to UL). Component carriers may include one primary component carrier and one or more secondary component carriers. The primary component carrier may be referred to as the primary cell (PCell), and the secondary component carriers may be referred to as secondary cells (SCells).

[0029] Some user equipment 104 can communicate using device-to-device (D2D) communication links 158. D2D communication links 158 can use DL / UL WWAN spectrum. D2D communication links 158 can use one or more sidelink channels, such as physical sidelink broadcast channels (PSBCH), physical sidelink discovery channels (PSDCH), physical sidelink shared channels (PSSCH), and physical sidelink control channels (PSCCH). D2D communication can be performed through various wireless D2D communication systems, such as FlashLinQ, WiMedia, Bluetooth, ZigBee, Wi-Fi based on the IEEE 802.11 standard, LTE, or NR.

[0030] The wireless communication system may also include a Wi-Fi access point (AP) 150 that communicates with Wi-Fi stations (STAs) 152 via a communication link 154 in the 5 GHz unlicensed spectrum. When communicating in the unlicensed spectrum, STAs 152 / AP 150 may perform a clear channel assessment (CCA) before communication to determine if the channel is available.

[0031] Small Cell 102' can operate in licensed and / or unlicensed spectrum. When operating in unlicensed spectrum, Small Cell 102' can employ NR and use the same 5 GHz unlicensed spectrum as Wi-Fi AP 150. Employing NR in unlicensed spectrum can enhance coverage and / or increase the capacity of the access network.

[0032] Base station 102, whether a small cell 102' or a large cell (e.g., a macro base station), may include an eNB, a gNodeB (gNB), or other types of base stations. Some base stations, such as gNB 180, can communicate with user equipment 104 in conventional sub-6 GHz spectrum, millimeter wave (mmW) frequencies, and / or near-millimeter wave frequencies. When gNB 180 operates in millimeter wave or near-millimeter wave frequencies, gNB 180 may be referred to as a millimeter wave base station. Extremely high frequency (EHF) is a radio frequency component of the electromagnetic spectrum. EHF ranges from 30 GHz to 300 GHz, with wavelengths between 1 mm and 10 mm. Radio waves in this band may be referred to as millimeter waves. Near-millimeter waves can extend down to frequencies of 3 GHz with wavelengths of 100 mm. The ultra-high frequency (SHF) band extends between 3 GHz and 30 GHz and is also known as centimeter waves. Communication using millimeter-wave / near-millimeter-wave radio frequency bands (e.g., 3 GHz - 300 GHz) suffers from extremely high path loss and short range. Millimeter-wave base station 180 can utilize beamforming 182 to communicate with user equipment 104 to compensate for the extremely high path loss and short range.

[0033] Base station 180 may transmit beamforming signals to user equipment 104 in one or more transmission directions 108a. User equipment 104 may receive beamforming signals from base station 180 in one or more reception directions 108b. User equipment 104 may also transmit beamforming signals to base station 180 in one or more transmission directions. Base station 180 may receive beamforming signals from user equipment 104 in one or more reception directions. Base station 180 / user equipment 104 may perform beam training to determine the optimal reception and transmission directions for each base station 180 / user equipment 104. The transmission and reception directions of base station 180 may be the same or different. The transmission and reception directions of user equipment 104 may be the same or different.

[0034] EPC 160 may include a Mobility Management Entity (MME) 162, other MMEs 164, a Serving Gateway 166, a Multimedia Broadcast Multicast Service (MBMS) Gateway 168, a Broadcast Multicast Service Center (BM-SC) 170, and a Packet Data Network (PDN) Gateway 172. MME 162 can communicate with the Home Subscriber Server (HSS) 174. MME 162 is the control node that handles signaling between UE 104 and EPC 160. Typically, MME 162 provides bearer and connection management. All user Internet Protocol (IP) packets are transmitted through Serving Gateway 166, which is itself connected to PDN Gateway 172. PDN Gateway 172 provides UE IP address allocation and other functions. PDN Gateway 172 and BM-SC 170 are connected to IP Service 176. IP service 176 may include the Internet, intranet, IP Multimedia Subsystem (IMS), PS streaming service, and / or other IP services. BM-SC 170 can provide MBMS user service provisioning and delivery functions. BM-SC 170 can serve as an entry point for content provider MBMS transmission, can be used to authorize and initiate MBMS bearer services within a public landmobile network (PLMN), and can be used to schedule MBMS transmissions. MBMS gateway 168 can be used to distribute MBMS traffic to base station 102 belonging to a Multicast Broadcast Single Frequency Network (MBSFN) area that broadcasts specific services, and may be responsible for session management (start / stop) and collecting billing information related to eMBMS.

[0035] The core network 190 may include an Access and Mobility Management Function (AMF) 192, other AMFs 193, a location management function (LMF) 198, a session management function (SMF) 194, and a User Plane Function (UPF) 195. The AMF 192 may communicate with the Unified Data Management (UDM) 196. The AMF 192 is the control node that handles signaling between the UE 104 and the core network 190. Typically, the SMF 194 provides QoS streaming and session management. All user Internet Protocol (IP) packets are transmitted through the UPF 195. The UPF 195 provides UE IP address allocation and other functions. The UPF 195 connects to an IP service 197. The IP service 197 may include the Internet, an intranet, an IP Multimedia Subsystem (IMS), PS streaming services, and / or other IP services.

[0036] A base station may also be referred to as a gNB, Node B, evolved Node B (eNB), access point, base transceiver station, wireless base station, wireless transceiver station, transceiver station function, basic service set (BSS), extended service set (ESS), transmit reception point (TRP), or other suitable terms. Base station 102 provides UE 104 with access to EPC 160 or core network 190. Examples of UE 104 include cellular phones, smartphones, session initiation protocol (SIP) phones, laptops, personal digital assistants (PDAs), satellite broadcasting, global positioning systems, multimedia devices, video devices, digital audio players (e.g., MP3 players), cameras, game consoles, tablets, smart devices, wearable devices, vehicles, electricity meters, gas pumps, large or small kitchen appliances, medical devices, implants, sensors / actuators, displays, or any other similarly functional device. Some UE 104 devices may be referred to as Internet of Things (IoT) devices (e.g., parking meters, gas pumps, toasters, vehicles, heart monitors, etc.). UE 104 may also be referred to as a station, mobile station, user station, mobile unit, user unit, radio unit, remote unit, mobile device, radio device, wireless communication device, remote device, mobile user station, access terminal, mobile terminal, radio terminal, remote terminal, mobile phone, user agent, mobile client, client, or other suitable terms.

[0037] Although this disclosure may refer to 5G New Radio (NR), it may apply to other similar fields, such as LTE, LTE-Advanced (LTE-A), Code Division Multiple Access (CDMA), Global System for Mobile communications (GSM), or other wireless / wireless access technologies.

[0038] Figure 2This is a block diagram illustrating communication between a base station 210 and a user equipment (UE) 250 in an access network. In the downlink, IP packets from the evolved packet core (EPC) 160 can be provided to the controller / processor 275. The controller / processor 275 implements Layer 3 and Layer 2 functions. Layer 3 includes the Radio Resource Control (RRC) layer, and Layer 2 includes the Packet Data Convergence Protocol (PDCP) layer, Radio Link Control (RLC) layer, and Medium Access Control (MAC) layer. The controller / processor 275 provides RRC layer functions related to system information (e.g., Master Information Block (MIB), System Information Blocks (SIBs)) broadcasting, RRC connection control (e.g., RRC connection paging, RRC connection establishment, RRC connection modification, and RRC connection release), inter-radio access technology (RAT) mobility, and measurement configuration for UE measurement reporting; PDCP layer functions related to header compression / decompression, security (encryption, decryption, integrity protection, integrity verification), and handover support functions; RLC layer functions related to upper-layer packet data units (PDUs) transmission, error correction via Automatic Repeat Request (ARQ), RLC service data units (SDUs) connection, segmentation and reassembly, RLC data PDU resegmentation, and RLC data PDU reordering; and mapping between logical channels and transport channels, multiplexing of MAC SDUs to transport blocks (TBs), and MAC from TBs. MAC layer functions related to SDU demultiplexing, scheduling information reporting, error correction via Hybrid Automatic Repeat Request (HARQ), priority handling, and logical channel prioritization.

[0039] The transmit (TX) processor 216 and receive (RX) processor 270 implement Layer 1 functions related to various signal processing functions. Layer 1, including the physical (PHY) layer, may include error detection of the transport channel, forward error correction (FEC) encoding / decoding of the transport channel, interleaving, rate matching, mapping to the physical channel, modulation / demodulation of the physical channel, and multiple-input multiple-output (MIMO) antenna processing. The TX processor 216 processes the mapping to the signal constellation according to various modulation schemes (e.g., binary phase-shift keying (BPSK), quadrature phase-shift keying (QPSK), M-phase-shift keying (M-PSK), and M-quadrature amplitude modulation (M-QAM)). The encoded and modulated symbols can then be split into parallel streams. Each stream can then be mapped to an Orthogonal Frequency Division Multiplexing (OFDM) subcarrier, multiplexed in the time and / or frequency domains with a reference signal (e.g., a pilot), and then combined using an Inverse Fast Fourier Transform (IFFT) to generate a physical channel carrying a time-domain OFDM symbol stream. The OFDM streams are spatially precoded to generate multiple spatial streams. Channel estimates from channel estimator 274 can be used to determine coding and modulation schemes, as well as for spatial processing. The channel estimates can be derived from the reference signal and / or channel condition feedback transmitted by UE 250. Each spatial stream can then be provided to a different antenna 220 via a separate transmitter 218TX. Each transmitter 218TX can modulate a radio frequency (RF) carrier for transmission with the corresponding spatial stream.

[0040] At UE 250, each receiver 254RX receives signals through its corresponding antenna 252. Each receiver 254RX recovers the information modulated onto the RF carrier and provides the information to the receive (RX) processor 256. The TX processor 268 and RX processor 256 implement Layer 1 functions related to various signal processing functions. The RX processor 256 can perform spatial processing on the information to recover any spatial stream oriented towards UE 250. If multiple spatial streams are oriented towards UE 250, they can be combined by the RX processor 256 into a single OFDM symbol stream. The RX processor 256 then uses a Fast Fourier Transform (FFT) to transform the OFDM symbol stream from the time domain to the frequency domain. The frequency domain signal consists of a separate OFDM symbol stream for each subcarrier of the OFDM signal. The symbols and reference signals on each subcarrier are recovered and demodulated by determining the most probable signal constellation points transmitted by base station 210. These soft decisions can be based on channel estimates calculated by channel estimator 258. The soft decision is then decoded and deinterleaved to recover the data and control signals originally transmitted by base station 210 on the physical channel. The data and control signals are then provided to controller / processor 259, which implements Layer 3 and Layer 2 functions.

[0041] Controller / processor 259 may be associated with memory 260, which stores program code and data. Memory 260 may be referred to as a computer-readable medium. In the uplink, controller / processor 259 provides demultiplexing, packet reassembly, decryption, header decompression, and control signal processing between the transmission and logical channels to recover IP packets from EPC 160. Controller / processor 259 is also responsible for error detection using acknowledgment (ACK) and / or denial (NACK) protocols to support HARQ operation.

[0042] Similar to the functions related to downlink transmission of base station 210, controller / processor 259 provides RRC layer functions related to system information (e.g., MIB, SIBs) acquisition, RRC connection, and measurement reporting; PDCP layer functions related to header compression / decompression and security (encryption, decryption, integrity protection, integrity verification); RLC layer functions related to upper-layer PDU transmission, error correction via ARQ, connection, segmentation and reassembly of RLC SDUs, resegmentation of RLC data PDUs, and reordering of RLC data PDUs; and MAC layer functions related to mapping between logical channels and transport channels, multiplexing of MAC SDUs to TBs, demultiplexing of MAC SDUs from TBs, scheduling information reporting, error correction via HARQ, priority processing, and logical channel priority.

[0043] The channel estimate derived by the channel estimator 258 from the reference signal or feedback transmitted from the base station 210 can be used by the TX processor 268 to select an appropriate coding and modulation scheme and facilitate spatial processing. The spatial stream generated by the TX processor 268 can be provided to different antennas 252 via individual transmitters 254TX. Each transmitter 254TX can modulate the RF carrier into a corresponding spatial stream for transmission. Uplink transmission is processed in the base station 210 in a manner similar to the reception function on the user equipment 250. Each receiver 218RX receives the signal through its corresponding antenna 220. Each receiver 218RX recovers the information modulated onto the RF carrier and provides the information to the RX processor 270.

[0044] Controller / processor 275 may be associated with memory 276, which stores program code and data. Memory 276 may be referred to as a computer-readable medium. In the uplink, controller / processor 275 provides demultiplexing, packet reassembly, decryption, header decompression, and control signal processing between transport and logical channels to recover IP packets from user equipment 250. IP packets from controller / processor 275 may be provided to EPC 160. Controller / processor 275 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operation.

[0045] New Radio (NR) can refer to radio operating under a new air interface (e.g., a non-Orthogonal Frequency Divisional Multiple Access (OFDMA) underlying air interface) or a fixed transport layer (e.g., a non-Internet Protocol (IP)). NR can use Orthogonal Frequency Division Multiplexing (OFDM) with a cyclic prefix (CP) in both uplink and downlink, and may include support for half-duplex operation using Time Division Duplex (TDD). NR may include Enhanced Mobile Broadband (eMBB) services targeting wide bandwidth (e.g., above 80 MHz), millimeter wave (mmW) services targeting high carrier frequencies (e.g., 60 GHz), massive Machine Type Communication (mMTC) services targeting non-backward-compatible MTC technologies, and / or mission-critical Ultra-Reliable Low Latency Communications (URLLC) services.

[0046] A single component carrier bandwidth can support 100 MHz. In one example, NR resource blocks (RBs) can span 12 subcarriers with a subcarrier bandwidth of 60 kHz and a duration of 0.25 ms, or a bandwidth of 30 kHz and a duration of 0.5 ms (similarly, a 15 kHz subcarrier spacing (SCS) supports 50 MHz bandwidth within a 1 ms duration). Each radio frame can consist of 10 subframes (10, 20, 40, or 80 NR slots) with a length of 10 ms. Each slot can indicate the link direction of data transmission (i.e., downlink or uplink), and the link direction of each slot can be dynamically switched. Each slot can include downlink / uplink data and downlink / uplink control data. NR uplink and downlink slots can be configured as follows: Figure 5 and Figure 6 A more detailed description is provided below.

[0047] A Radio Access Network (RAN) may include a Central Unit (CU) and Distributed Units (DUs). An NR Base Station (BS, e.g., gNB, 5G Node B, Node B, Transmission Reception Point (TRP), Access Point (AP)) may correspond to one or more base stations. NR cells can be configured as Access Cells (ACells) or Data Only Cells (DCells). These cells can be configured within the RAN (e.g., the Central Unit or Distributed Unit). DCells can be used for carrier aggregation or dual connectivity and may not be used for initial access, cell selection / reselection, or handover. In some cases, DCells may not transmit synchronization signals (SS); in others, they may transmit SS. NR base stations may transmit downlink signals to user equipment to indicate the cell type. Based on the cell type indication, user equipment can communicate with NR base stations. For example, user equipment can determine the NR base station based on the indicated cell type for cell selection, access, handover, and / or measurement.

[0048] Figure 3 An example logical architecture of a distributed RAN 300 according to the present disclosure is described. A 5G access node 306 may include an Access Node Controller (ANC) 302. The ANC may be the central unit (CU) of the distributed RAN. The backhaul interface to the Next Generation Core Network (NG-CN) 304 may terminate at the ANC. The backhaul interface to neighboring Next Generation Access Nodes (NG-ANs) 310 may terminate at the ANC. The ANC may include one or more TRPs 308 (also referred to as base stations, NR base stations, Node Bs, 5G NBs, APs, or other terms). As mentioned above, TRP can be used interchangeably with "cell".

[0049] TRPs 308 may be Distributed Units (DUs). TRPs may connect to one ANC (ANC 302) or multiple ANCs (not shown). For example, for RAN sharing, Radio as a Service (RaaS), and service-specific ANC deployments, TRPs may connect to multiple ANCs. A TRP may include one or more antenna ports. TRPs may be configured to provide traffic services to user equipment individually (e.g., dynamically selected) or jointly (e.g., jointly transmitted).

[0050] The local architecture of the distributed RAN 300 can be used to illustrate the fronthaul definition. The architecture can be defined to support fronthaul solutions across different deployment types. For example, the architecture can be based on transport network capabilities (e.g., bandwidth, latency, and / or jitter). The architecture can share features and / or components with LTE. Depending on the configuration, the Next Generation Access Node (NG-AN) 310 can support dual connectivity with NR. The NG-AN can share a common fronthaul for both LTE and NR.

[0051] This architecture enables collaboration between and within TRPs 308. For example, collaboration can be established within a TRP and / or across TRPs via ANC 302. Depending on the configuration, inter-TRP interfaces may not be required or may not exist.

[0052] Depending on certain patterns, the architecture of a distributed RAN 300 may involve dynamic configuration of segmentation logic functions. PDCP, RLC, and MAC protocols can be adaptively placed in the ANC or TRP.

[0053] Figure 4An example physical architecture of a distributed RAN 400 according to certain embodiments of this disclosure is shown. A centralized core network unit (C-CU) 402 can carry core network functions. The C-CU can be centrally deployed. C-CU functions can be offloaded (e.g., to advanced wireless services (AWS)) to handle peak capacity. A centralized RAN unit (C-RU) 404 can carry one or more ANC functions. Optionally, the C-RU can carry core network functions locally. The C-RU can be distributed. The C-RU can be closer to the network edge. A distributed unit (DU) 406 can carry one or more TRPs. The DU can be located at the network edge and has radio frequency (RF) functions.

[0054] Figure 5 This is a schematic diagram 500 showing an example of a downlink (DL)-centric timeslot. The DL-centric timeslot may include a control section 502. The control section 502 may exist in the initial or beginning portion of the DL-centric timeslot. The control section 502 may include various scheduling and / or control information corresponding to different portions of the DL-centric timeslot. In some configurations, the control section 502 may be a physical downlink control channel (PDCCH), such as... Figure 5 As shown. The DL-centered time slot may also include a DL data portion 504. The DL data portion 504 may sometimes be referred to as the payload of the DL-centered time slot. The DL data portion 504 may include communication resources for communicating DL data from a scheduling entity (e.g., a user equipment (UE) or a base station (BS)) to a subordinate entity (e.g., a UE). In some configurations, the DL data portion 504 may be a physical downlink shared channel (PDSCH).

[0055] The DL-centric time slot may also include a common uplink (UL) section 506. The common UL section 506 may sometimes be referred to as a UL burst, common UL burst, and / or other appropriate terms. The common UL section 506 may include feedback information corresponding to other sections of the DL-centric time slot. For example, the common UL section 506 may include feedback information corresponding to the control section 502. Examples of non-limiting feedback information may include acknowledgment (ACK) signals, denial (NACK) signals, hybrid automatic repeat request (HARQ) indicators, and / or other appropriate types of information. The common UL section 506 may include additional or alternative information, such as information related to random access channel (RACH) procedures, scheduling requests (SRs), and other appropriate types of information.

[0056] like Figure 5 As shown, the end of DL data section 504 may be time-separated from the start of common UL section 506. This time separation may sometimes be referred to as a gap, guard period, guard interval, and / or other appropriate terms. This separation provides time for switching between DL communication (e.g., reception operation of a subordinate entity (e.g., UE)) and UL communication (e.g., transmission of a subordinate entity (e.g., UE)). Those skilled in the art will understand that the above is merely an example of a DL-centered time slot, and alternative structures with similar characteristics may exist without departing from the pattern described herein.

[0057] Figure 6 This is a schematic diagram 600 showing an example of an uplink (UL) centered timeslot. The UL-centered timeslot may include a control section 602. The control section 602 may be present in the initial or beginning portion of the UL-centered timeslot. Figure 6 The control section 602 in the above reference can be similar to the one described above. Figure 5 The control section 502 is described. The UL-centered time slot may also include a UL data section 604. The UL data section 604 may sometimes be referred to as the payload of the UL-centered time slot. The UL section may refer to communication resources used to communicate UL data from a subordinate entity (e.g., UE) to a scheduling entity (e.g., UE or BS). In some configurations, the control section 602 may be a physical downlink control channel (PDCCH).

[0058] like Figure 6As shown, the end of control section 602 may be time-separated from the start of UL data section 604. This time separation may sometimes be referred to as a gap, protection period, protection interval, and / or other appropriate terms. This separation provides time for switching between DL communication (e.g., receiving operations of a scheduling entity) and UL communication (e.g., transmissions of a scheduling entity). The UL-centric time slot may also include a common UL section 606. Figure 6 The public UL section 606 in the above reference can be similar to the above reference. Figure 5 The public UL section 506 is described. Public UL section 606 may additionally or alternatively include information relating to channel quality indicators (CQI), sounding reference signals (SRSs), and other suitable types of information. Those skilled in the art will understand that the above is merely an example of a UL-centered timeslot, and alternative structures with similar characteristics may exist without departing from the pattern described herein.

[0059] In some cases, two or more dependent entities (e.g., user equipment) may use sidechain signaling to communicate. Practical applications of such sidechain communication may include public safety, proximity services, user equipment-to-network relay, vehicle-to-vehicle (V2V) communication, Internet of Everything (IoE) communication, Internet of Things (IoT) communication, mission-critical meshes, and / or various other applicable applications. Typically, sidechain signaling may refer to signaling communication from one dependent entity (e.g., user equipment 1) to another dependent entity (e.g., user equipment 2) without relaying the communication through a scheduling entity (e.g., user equipment or base station), even if the scheduling entity may be used for scheduling and / or control purposes. In some examples, sidechain signaling may use licensed spectrum for communication (unlike wireless LANs that typically use unlicensed spectrum).

[0060] Figure 7(A) is a figure 700 illustrating an artificial intelligence / machine learning (AI / ML) model for spatial and temporal beam prediction. In this example, base station 702 transmits beams 711-734 simultaneously in multiple directions via channel 780. After identifying the incident beams, user equipment 704 can calculate the Layer 1 Reference Signal Received Power (L1-RSRP) for each beam. L1-RSRP is the average received power of resource elements carrying subsynchronous signals or channel state information reference signals (CSI-RS).

[0061] Machine learning algorithms are used to analyze the historical signal strength of a subset of beams and attempt to find patterns or trends in the data. This helps predict the signal strength of the remaining unmeasured beams. By identifying patterns in the historical data of the beam subset, the algorithm can predict the signal strength of other beams, even when the user equipment is moving.

[0062] In this example, base station 702 is equipped with multiple antennas capable of simultaneously radiating 24 different beams 711-734 in multiple directions. User equipment 704 is frequently mobile and may be equipped with its own antennas, periodically measuring channel metrics, such as RSRP, of four beams selected from the 24 beams radiated from base station 702 (e.g., beams 715, 716, 729, and 730). The set of beams measured as AI / ML inputs (sensing beams) (e.g., beams 715, 716, 729, and 730) is called beam set B. The set of beams predicted as AI / ML outputs (typically communication beams) (e.g., 24 beams) is called beam set A.

[0063] Measurement data collected from these four beams is stored as historical data over time. This historical data captures how channel metrics of a subset of beams change over time, thus capturing how user equipment 704 interacts with these beams. User equipment 704 may be configured with a historical data time window 760 during which measurement data is stored in user equipment 704. In this example, the current time is... Historical data time window 760 from time arrive Measurement data of beam subsets 715, 716, 729, and 730 acquired during the historical data time window 760 are stored in user equipment 704 and used as input to AI / ML model 750 to predict the current time. Measurement data of unmeasured beams and future time Measurement data for all beams.

[0064] Historical data is used as input to a machine learning algorithm to predict the channel metrics of unmeasured beams, guiding user equipment 704 to focus on which beams when it needs to communicate with base station 702. Therefore, the algorithm can predict the channel metrics of all beams by analyzing patterns and trends in a subset of historical beam data.

[0065] Another set of beams (e.g., beams 711, 721, 724, and 734) is not measured under normal circumstances, but its channel metrics can be sampled and recorded periodically. This can be used to verify whether the algorithm's predictions match actual performance, while updating the AI / ML model. Periodic measurements help improve the algorithm by updating weights and parameters. As machine learning algorithms mature, their predictions of the optimal beam will become increasingly accurate. When user equipment 704 initiates communication, it can select whether to transmit or receive beam 770, which may produce superior signal quality (e.g., the optimal beam) based on predictions.

[0066] Instead of identifying a single best beam, this method predicts the front beams that are likely to have the highest channel performance. One beam. In many cases, focus on the front. Each beam can provide excellent accuracy. By using a front-based... Estimating these channel index values, achieving the previous... Beam prediction. This closely matches actual communication needs and improves system performance.

[0067] The main outputs of classification-based AI / ML models include pre-prediction data used for communication. Identifiers (IDs) for each optimal beam, along with a corresponding predicted confidence score or predicted RSRP for each beam. These beams are determined as the most suitable beams for communication based on expected signal strength and reliability. For example, if... With a setting of 5, the model may predict that the five optimal communication beams in the entire beam set are beams numbered 732, 730, 734, 728, and 733. This prediction enables user equipment 704 to make informed decisions at any given time about which beams base station 702 and user equipment 704 should use for communication, optimizing performance based on signal strength and the likelihood of successful data transmission.

[0068] Furthermore, the output of the regression-based AI / ML model includes the predicted RSRP value for each communication beam in beamset A. This prediction output directly estimates the expected signal strength of each individual beam in the communication set, enabling user equipment 704 to select beams more precisely based on the predicted RSRP values.

[0069] Beamforming is a technique for enhancing data rates and reliability in 5G and higher versions of wireless communication, particularly in the millimeter-wave (mmWave) band, enabling a base station (e.g., base station 702) to focus its signal transmission and reception on a specific user equipment (UE) (e.g., UE 704). This directional approach improves signal quality and reduces interference. To establish optimal beamforming, base station 702 and UE 704 need to identify the best beam for transmitting and receiving data; this process is called beam management. Traditional beam management typically involves a comprehensive beam scan, where base station 702 and UE 704 systematically scan all available beam directions to find the optimal one. However, as the number of antennas and beams increases, this approach becomes inefficient and resource-intensive, leading to significant overhead.

[0070] In contrast, AI / ML-based beam management offers a more flexible and efficient alternative to full beam scanning. By leveraging the power of machine learning, this approach predicts the optimal beam for communication based on historical data analysis obtained from a subset of beams.

[0071] As shown in Figure 7(A), UE 704 does not measure all 24 beams, but selectively measures the L1-RSRP from a smaller subset of beams, referred to as beam set B, which serves as input to AI / ML model 750. By analyzing patterns and trends in historical data from this beam subset, AI / ML model 750 predicts the L1-RSRP values ​​of the remaining unmeasured beams in beam set A.

[0072] AI / ML models 750 can take many forms, such as classification-based or regression-based models. Classification models predict the best for communication. The model identifies the beams and provides associated confidence scores. The regression model directly estimates the RSRP value for each communication beam in beam set A. Regardless of the model chosen, this predictive capability significantly reduces the need for comprehensive measurements, minimizes overhead, and improves efficiency.

[0073] Temporal beam prediction is a key feature of AI / ML-based beam management, which predicts the optimal future beam index based on historical beam measurements. UE 704 uses measurements taken within a historical data time window 760, enabling model 750 to learn temporal patterns and predict future beam conditions.

[0074] Essentially, AI / ML-based beam management provides a faster and more efficient way to obtain optimal beam information and optimize the communication beam selection between the base station and the UE, especially in dynamic environments where beam conditions may change rapidly.

[0075] Figure 7(B) is a figure 760 illustrating the temporal pattern of beam measurement and prediction in AI / ML-based beam management. The figure shows two sequences: measurement sequence 761 and prediction sequence 762.

[0076] Measurement sequence 761 indicates Measurement examples, where UE 704 performs beam measurements. These measurements are typically performed on a subset of beams (e.g., beam set B) as input to an AI / ML model. These measurements provide power measurements of the sensed beams, used to infer the optimal communication beam.

[0077] Predicted sequence 762 indicates Predictive instances where AI / ML models predict the optimal future beam index. These predictions are based on beam measurements from the previous time step, particularly... Measurements performed during the measurement instance.

[0078] This temporal structure aligns with the concept of temporal beam prediction, which aims to predict the optimal future beam index using beam measurements from sensed beams from previous time steps. The RSRP of one or more beams can be predicted by inputting historical RSRPs.

[0079] In other words, measurement sequence 761 corresponds to the time instance of the actual measurement reported by UE 704. Prediction sequence 762 represents the future time instance of beam conditions predicted by the AI / ML model (whether in UE 704 or base station 702).

[0080] In advanced wireless communication systems, such as 5G and later, beamforming can improve data rates and reliability, especially in the millimeter-wave (mmWave) band. Base station 702 and UE 704 need to identify the optimal beam for communication; this process is known as beam management. Traditional beam management typically involves a full beam scan, which becomes inefficient as the number of available beams increases. To address this challenge, AI / ML-based beam management is proposed as a more efficient alternative.

[0081] AI / ML-based beam management requires new reporting configurations to accommodate various beam reporting scenarios and purposes. The report format used to deliver beam measurement results may vary depending on the location of the AI / ML model (whether on the network side or the UE side) and the specific requirements of the beam management task.

[0082] When the AI / ML model is located on the network side (e.g., base station 702), UE 704 needs to be configured to report measurement results for beam set B. The method for reporting these measurement results may vary depending on the specific AI / ML model design. For temporal beam prediction, UE 704 may need to include a time ID in the report to indicate the measurement time.

[0083] Conversely, when the AI / ML model is located on the UE side, UE 704 needs to be configured to report the inferred optimal beam from the communication beams. In this case, UE 704 may output a variable number of predicted optimal beams in the design report based on the AI / ML model. The network does not know this number in advance, thus requiring a new measurement report format. Similar to the network-side model case, temporal beam prediction may need to include a time ID in the report.

[0084] For network-side data collection purposes, UE 704 needs to be configured to report measurement results for a set of beams, which may include beam set A, beam set B, the union of beam sets A and B, or other combinations. UE 704 may output a variable number of optimal measurement beams as labels in the design report based on the AI / ML model, which needs to be configured by the network. Furthermore, UE 704 may input a variable number of measurement beams in the design report based on the AI / ML model, which may or may not be known to the network in advance.

[0085] To accommodate these different scenarios and requirements, new Radio Resource Control (RRC) parameters need to be created in the CSI report configuration. These parameters will allow the network to differentiate between different situations and enable User Equipment 704 to send measurement reports in the appropriate beam reporting format.

[0086] To address the challenges of traditional beam management and leverage the advantages of AI / ML-based approaches, new Radio Resource Control (RRC) parameters can be introduced into the reporting configuration. These parameters are designed to enable efficient beam reporting for AI / ML-based beam management, adaptable to various scenarios and purposes.

[0087] The new set of configuration parameters in the report configuration may include several elements. First, the mode parameters indicate the AI ​​beam management (BM) mode, such as spatial (BM Case 1) and temporal beam prediction (BM Case 2). For temporal beam prediction, additional time information needs to be added to the beam report. This may include parameters indicating the periodicity of the predicted time instances, the number of predicted time instances, and the number of observed time instances.

[0088] Secondly, the receive (Rx) parameter specifies the type of Rx measurement. These parameters are particularly useful for network-side models because they rely on the user equipment 704 providing accurate beam measurements as input. Different AI / ML models may be trained with different assumptions based on how the user equipment 704 measures RSRP / SINR. These assumptions may involve formulating AI / ML model inputs using all available Rx beams, a specific Rx beam, or a near-optimal Rx beam. Examples of near-optimal Rx beams might include previously used Rx beams or Rx beams obtained by scanning a specific Tx beam using the user equipment's Rx beams.

[0089] The Rx type parameter is designed to address potential measurement issues that may affect the accuracy of AI / ML model predictions. Two scenarios that could lead to poor L1 RSRP measurements are: (1) when user equipment 704 selects an Rx beam pointing in a suboptimal direction, resulting in inaccurate measurements; and (2) when user equipment 704 is far from base station 702, measurements may still be poor even with a good beam orientation. While AI / ML models can handle the second scenario through training, the Rx type parameter helps mitigate the first scenario by guiding user equipment 704 to select the optimal Rx beam orientation.

[0090] Third, the model output parameter indication report is the inference AI / ML model output report used for the user equipment side model. These parameters may include sub-parameters specifying the number of inference beams to be reported. This allows the network to configure user equipment 704 to report a specific number of optimal predicted beams based on the output of the user equipment side AI / ML model.

[0091] Fourth, the model input parameter report indicates that the AI / ML model input is used for the network-side model. These parameters may include sub-parameters specifying the report format, which can vary depending on the input design of the AI / ML model. This flexibility allows the network to configure user equipment 704 to report measurement results in a format that conforms to the requirements of the network-side AI / ML model.

[0092] Fifth, the data collection parameters indicate that the report is for data collection purposes, which is very useful for AI / ML model training. These parameters may include sub-parameters specifying the report format (designed based on the AI / ML model input) and label size. The network can also be designed to indicate whether to include labels in the report based on the AI / ML model output. For data collection, user equipment 704 needs to report both model input (beam measurement) and ground truth (actual optimal beam) to enable the AI / ML model to learn from real-world data.

[0093] Finally, the model monitoring parameter indication report is used for model monitoring purposes. These parameters can be used to specify various monitoring reporting purposes, such as reporting events, reporting monitoring performance metrics, or reporting beam measurements for a specific beam set. This allows the network to monitor the performance and accuracy of the AI / ML underlying beam management system over time.

[0094] By introducing these new RRC parameters, reporting configurations can be customized to support a variety of AI / ML-based beam management scenarios. This approach allows for more efficient and flexible beam reporting, enabling systems to leverage the power of AI / ML technologies while minimizing overhead and maximizing performance in 5G and later wireless communication systems.

[0095] Figure 8(A) illustrates a flowchart 800 of a beam reporting process in an AI / ML basic beam management system. This process involves the interaction between the network (NW) and user equipment (e.g., user equipment 704) through a base station (e.g., base station 702).

[0096] In module 802, the network determines the beam reporting configuration for AI / ML basic beam management. The beam reporting configuration is then communicated between the network and user equipment 704. For example, user equipment 704 receives the configuration transmitted from the network. This configuration may optimize the beam management process because it defines the parameters and instructions used by user equipment 704 for its reporting tasks. The beam reporting configuration may cover various formats, including the number of beams to be reported, the types of measurements to be included in the report, the specific report format to be used, and the set of configuration parameters that enable reporting for AI / ML basic beam management. The set of configuration parameters may include several elements discussed in previous sections. These may include AI beam management (BM) modes (e.g., spatial beam prediction or temporal beam prediction), measurement Rx types, parameters for user equipment-side model output reports, parameters for network-side model input reports, and parameters for data collection reports.

[0097] In module 804, user equipment 704 determines a set of beams that meet predetermined performance indicators based on the beam report configuration. User equipment 704 may determine a subset (set C) of beams that meet specified performance indicators. These indicators may include factors such as signal strength and reliability. User equipment 704 may infer a set of sensed beams (set B) using measurement and / or AI / ML models. In module 806, user equipment 704 reports the beam set to the base station.

[0098] If the AI / ML model is on the user equipment side, it processes measurements from set B to predict the optimal beam in set A. If the model is on the network side, the user equipment 704 reports the measurement results from set B, enabling the network to make predictions.

[0099] Once a subset of beams that meets the performance specifications is determined, user equipment 704 reports these measurements based on the received beam reports. This may involve reporting the predicted best beam, its corresponding reference signal received power (RSRP) value, or other specified specifications.

[0100] Figure 8(B) shows a flowchart 830 that outlines an alternative beam reporting process, adding a step focused on user equipment capability reporting. Modules 832, 834, and 836 of this process are similar to modules 802, 804, and 806 of the process shown in Figure 8(A), respectively.

[0101] Furthermore, in module 838, user equipment 704 reports a set of capability parameters. This set of parameters indicates whether user equipment 704 supports reporting for artificial intelligence / machine learning (AI / ML) basic beam management. The set of capability parameters may include various elements describing the user equipment's capabilities in AI / ML basic beam management. These may include support for different AI beam management modes (e.g., spatial or temporal beam prediction), supported receive (Rx) usage types, capabilities related to user equipment-side model output reports (e.g., the number of the top K beams supported for reporting), capabilities related to network-side model input reports (e.g., supported data filtering methods), and capabilities related to data collection reports (e.g., supported label sizes). In the embodiment shown in Figure 8(B), user equipment 704 first reports its capabilities, and then the network determines a beam reporting configuration based on user equipment 704's capability report.

[0102] Figure 8(C) shows a flowchart 850 that outlines the further beam reporting process, adding a step for reporting user equipment capabilities. Modules 852, 854, 856, and 858 of this process are similar to modules 832, 834, 836, and 838 of the process shown in Figure 8(B), respectively.

[0103] However, unlike the embodiment shown in Figure 8(B), where module 838 occurs before module 832, in the embodiment shown in Figure 8(C), module 858 may occur after module 852. Upon receiving the beam reporting configuration, in module 860, user equipment 704 determines whether it supports the configuration. If supported, the user equipment performs its reporting duties according to the received configuration. Otherwise, the process continues to module 858 and then returns to module 852 to take further action. That is, the network initially determines a beam reporting configuration. If user equipment 704 is able to support the configuration, it performs its reporting duties. If not, user equipment 704 reports its capabilities to the network, and the network may then determine a new beam reporting configuration based on that capability report.

[0104] By reporting these capabilities, User Equipment 704 provides information to the network, making beam management more efficient and effective. The network can then use this information to optimize its configurations and requests to match the specific capabilities of each User Equipment.

[0105] Figure 9 Figure 910 illustrates the CSI report configuration structure, CSI-ReportConfig, according to this disclosure. This configuration structure includes information elements (IEs) designed to support AI / ML-based beam management in advanced wireless communication systems.

[0106] The configuration structure includes an `enableAIreporting` information element, which serves as a flag indicating whether AI reporting is enabled. When this information element is empty, user equipment 704 assumes that conventional beam reporting is used.

[0107] The AI_BM mode information element specifies the AI / ML beam management mode used for reporting. It provides two options: BM-Case1 and BM-Case2. BM-Case1 corresponds to spatial beam prediction, where user equipment 704 assumes the report is for spatial beam management. BM-Case2 involves temporal beam prediction, where time information is included in the beam report. For BM-Case2, two sub-parameters are defined: Predict_PeriodicityNSize and Observe_Size. Predict_PeriodicityNSize indicates the number and periodicity of predicted time instances per time instance in the prediction window. Observe_Size specifies the number of previous measurements of a set of beams used as model input.

[0108] The Rx_usage_type information element is used to report measurements of a set of beams for AI / ML beam management. When this information element is configured, user equipment 704 assumes that the measurements are derived from configured Rx usage behavior. This information element addresses potential measurement issues that may affect the prediction accuracy of AI / ML models, particularly when user equipment 704 may select suboptimal Rx beam orientations.

[0109] The `Output_reporting_on_UEside_model` information element is used when the AI / ML model is located on the user equipment side. When user equipment 704 is configured with this information element, it prepares information related to the best predicted beam in the beam report. This information element includes a sub-parameter called `Top k Reports`, indicating the possible number of best predicted beams for a prediction time instance. The value of `Top k Reports` ranges from n0 to nK, where nk represents the top K beams that user equipment 704 reports for a prediction time instance. The value n0 allows user equipment 704 to adaptively select the K value for reporting. If BM-Case2 is configured, this value is applied to all time instances indicated by `Predict_PeriodicityNSize`.

[0110] The `Input_reporting_on_NWside_model` information element is used when the AI / ML model is located on the network side. When this information element is configured, User Equipment 704 is prepared to measure a set of reference signal (RS) resources configured in the beam report. The value selection of this information element indicates the reporting format type used to report the measurements. This information element represents a significant change from conventional reporting methods. In conventional systems, when the network configures a set of RS resources (e.g., 24 beams) for User Equipment 704 to measure, User Equipment 704 typically only reports the first four best RSRP measurements. However, with AI / ML-based beam management, when the network configures a set of RS resources (e.g., 4 or 8 beams, depending on the size of set B), User Equipment 704 is expected to report all measurements it obtains, as specified in the network configuration.

[0111] The Data_collection_reporting information element (IE) is used for data collection purposes, specifically for training artificial intelligence / machine learning (AI / ML) models. When the User Equipment (UE) 704 is configured with this information element, it prepares two types of information in the beam report: measurement results of the first set of Reference Signal (RS) resources and an indicator of the best beam measured in the second set of RS resources. The value selection of this information element indicates the reporting format type used to report the measurement results and indicators. This information element includes a sub-parameter named Label Size, indicating the number of best beam indicators of the second set of RS resources that the UE 704 includes in the beam report for each time instance.

[0112] These new information elements provide a flexible and comprehensive framework for configuring AI / ML-based beam management reports. They allow networks to specify detailed requirements for beam reports, adapting to various scenarios such as user equipment-side models, network-side models, and data collection for model training. By introducing these parameters, the system can leverage the advantages of AI / ML technologies in beam management while maintaining flexibility to adapt to different network configurations and user equipment capabilities.

[0113] The introduction of these new information elements addresses the limitations of traditional beam reporting methods, especially in the context of AI / ML-based beam management. They enable more efficient and targeted reporting, allowing for better utilization of network resources and potentially improving the overall system performance of 5G and higher-level wireless communication systems.

[0114] In the first example utilizing the CSI report configuration structure, the network (NW) can configure user equipment (UE) 704 to report the top k predicted best beams from the UE-side model for beam management (BM) case 1. This example illustrates the practical application of the aforementioned AI / ML-based beam management reporting configuration.

[0115] The network configures two parameters in the CSI report configuration via base station 702: "AI_BM mode" and "Output report of user equipment side model". These parameters guide UE 704 on how to perform and report its beam prediction.

[0116] For the "AI_BM mode" parameter, the network can configure it to BM-Case1, indicating spatial beam prediction. This configuration tells UE 704 to perform spatial beam prediction instead of temporal beam prediction. In spatial beam prediction, UE 704 predicts the optimal beam based on the spatial distribution of the current signal strength, without considering time-series data.

[0117] Alternatively, if temporal sampling is required, the network can be configured with BM-Case2. In this case, additional sub-parameters will be set, such as Predict_PeriodicityNSize and Observe_Size. For example, these can be set to 1, indicating a single prediction instance and a single observation instance.

[0118] The “Output Report of User Equipment Side Model” parameter is configured to specify the top k reports. This parameter includes a sub-parameter that determines how many top beams the UE 704 should report. In this example, the network configures this value to n4, instructing the UE 704 to report the top-4 predicted best beams. The content of this report is further specified in the information element reportQuantity, and may include the beam identifier and the predicted Reference Signal Received Power (RSRP) value.

[0119] Upon receiving this configuration, when UE 704 is triggered to generate a report, it follows a specific process. First, UE 704 measures the configured RS resource sets, which correspond to set B in the AI / ML model. These measurements can be performed in real time, or, to reduce processing overhead, UE 704 can use the most recent previous measurement (if available).

[0120] Next, UE 704 applies its user equipment-side AI / ML model to these measurements. This model has been trained on historical data, processing inputs from set B to infer the top-4 predicted best beam. The model considers various factors, such as signal strength patterns and historical performance, to make these predictions.

[0121] Finally, UE 704 compiles a report based on the model's output and configuration parameters. This report includes resource indicators (beam IDs) for the top-4 predicted best beams. If the UE's AI model is able to predict RSRP values ​​(depending on the specific model implementation), UE 704 will also include these predicted RSRP values ​​in the report. UE 704 then transmits this report back to the network via base station 702.

[0122] This example demonstrates how the new CSI report configuration structure enables more efficient beam management. By allowing the UE704 to report only the predicted top beam based on its AI / ML model, rather than exhaustive measurements, the system reduces signaling overhead while still providing the network with critical information for optimizing communications. This approach leverages the UE's local processing capabilities and its ability to quickly adapt to changing signal conditions, potentially leading to overall system performance improvements in 5G and higher-level wireless communication systems.

[0123] In the second example utilizing the CSI report configuration structure, the network (NW) can configure user equipment (UE) 704 to report the top k predicted best beams from the UE-side model, as in Case 2 of beam management (BM). This example illustrates the application of AI / ML-based beam management report configuration in temporal beam prediction.

[0124] The network configures two main parameters in the CSI report configuration via base station 702: "AI_BM mode" and "Output report of user equipment side model". These parameters guide UE 704 on how to perform and report its beam prediction, especially for time beam prediction.

[0125] For the "AI_BM mode" parameter, the network configures it to BM-Case2, indicating temporal beam prediction. This configuration tells UE 704 to perform temporal beam prediction, taking time-series data into account in the prediction. In temporal beam prediction, UE 704 predicts the optimal beam not only based on the spatial distribution of the current signal strength, but also based on how these signal strengths change over time.

[0126] When configuring BM-Case2, the network also sets two important sub-parameters: Predict_PeriodicityNSize and Observe_Size. In this example, Predict_PeriodicityNSize is set to {20ms, size=2}. This means that UE 704 is instructed to make predictions for two future time slots, each with a periodicity of 20 milliseconds. Observe_Size is set to 4, indicating that UE 704 should use measurements from four previous time slots as input to its predictive model.

[0127] The “Output_reporting_on_UEside_model” parameter is configured to specify the top k reports. In this example, the network configures this value to n1, instructing UE704 to report the top-1 predicted best beam for each prediction time slot. The content of this report is further specified in the IE reportQuantity and may include details such as the beam identifier and the predicted RSRP value.

[0128] Upon receiving this configuration, when user equipment 704 is triggered to generate a report, it follows a specific process. First, user equipment 704 measures the configured RS resource set, which corresponds to set B in the AI / ML model, within four time periods specified by the observation size parameter. These measurements can be performed in real time, or, to reduce processing overhead, user equipment 704 can use the most recent previous measurement (if available).

[0129] Next, user equipment 704 applies its user equipment-side AI / ML model to these measurements. This model, trained on historical time-series data, processes input from set B to infer the Top-1 predicted best beam for each pair of future time slots. The model considers various factors, such as signal strength patterns, historical performance, and time trends, to make these predictions.

[0130] Finally, User Equipment 704 compiles a report based on the model's output and configuration parameters. This report includes resource indicators (beam IDs) for the top-1 predicted best beams for two future time slots at 20-millisecond intervals. If the User Equipment's AI model is able to predict RSRP values ​​(depending on the specific model implementation), User Equipment 704 will also include the predicted RSRP values ​​for each prediction time slot in the report.

[0131] This example demonstrates the flexibility and robustness of the new CSI reporting configuration structure in supporting advanced AI / ML-based beam management technologies. By allowing user equipment 704 to report predictions of future time slots based on time patterns, the system can predict and adapt to changing signal conditions more effectively than traditional methods. This approach leverages the local processing power of user equipment and its ability to learn from historical data, potentially leading to improved overall system performance and more efficient resource utilization, suitable for 5G and more advanced wireless communication systems.

[0132] In the third example utilizing the CSI report configuration structure, the network (NW) configures user equipment 704 to report measurements as input to the network-side AI / ML model. In this scenario, base station 702 configures several key parameters in the CSI report configuration: “AI_BM Mode”, “Rx_usage_type”, and “Input_reporting_on_NWside_model”. These parameters instruct user equipment 704 how to perform measurements and report them for network-side model inference.

[0133] For the "AI_BM Mode" parameter, base station 702 configures it to BM-Case1, indicating spatial beam prediction. This configuration tells user equipment 704 that spatial beam prediction should be performed instead of temporal beam prediction. In spatial beam prediction, user equipment 704 reports measurements based on the spatial distribution of the current signal strength, without considering time-series data.

[0134] The “Rx_usage_type” parameter is set to “BestRx”. This configuration is crucial because it defines how the user equipment 704 should perform its measurements. When set to “BestRx”, for each Tx beam of the model input, the user equipment 704 will perform multiple measurements using different Rx beams. It will then determine the best measurement among these measurements and include this best measurement in the measurement report. This method aims to provide the most accurate signal strength information for each Tx beam, which is essential for accurate predictions by the network-side AI / ML model.

[0135] The “Input_reporting_on_NWside_model” parameter is configured to specify the reporting method. If the user device 704 is configured with a specific container, it will use the corresponding container method (reporting format) to report measurements. This parameter allows for flexibility in how measurements are reported and can be customized according to the specific requirements of the network-side AI / ML model.

[0136] Upon receiving this configuration and being triggered to generate a report, user equipment 704 follows a specific procedure. First, it measures the configured RS resource sets, which correspond to set B in the AI / ML model. These measurements are performed using Rx beam scanning, as specified in the "BestRx" configuration. This means that for each Tx beam in set B, user equipment 704 will try multiple Rx beams and select the best measurement.

[0137] If the most recent previous measurements of the configured RS resource set are available and still valid, user equipment 704 can use these measurements to reduce processing overhead. This method is particularly useful when channel conditions are relatively stable over a short period of time.

[0138] Finally, User Equipment 704 compiles a report based on these measurement and configuration parameters. For each RS resource (Tx beam) configured in Set B, User Equipment 704 reports the optimally measured RSRP to Base Station 702. This report follows the specified report format configured in the "Input_reporting_on_NWside_model" parameter.

[0139] In the fourth example utilizing the CSI report configuration structure, the network (NW) configuration sets the parameters for network-side data collection in Beam Management (BM) Case 1. This example demonstrates how the new CSI report configuration can be used to collect data for training and improving AI / ML models on the network side.

[0140] Base station 702 has several key parameters configured in its CSI reporting configuration: “AI_BM Mode”, “Rx_usage_type”, and “Data_collection_reporting_for_NWside_model”. These parameters instruct user equipment 704 how to perform measurements and report them for data collection.

[0141] For the "AI_BM Mode" parameter, base station 702 configures it to BM-Case1, indicating spatial beam prediction. This configuration tells user equipment 704 that spatial beam prediction should be performed instead of temporal beam prediction. In spatial beam prediction, user equipment 704 reports measurements based on the spatial distribution of the current signal strength, without considering time-series data.

[0142] The “Rx_usage_type” parameter is set to “QuasiType1”. This configuration is crucial because it defines how user equipment 704 should perform its measurements. When set to “QuasiType1”, for each transmit (Tx) beam of the model input, user equipment 704 will perform a measurement using the optimal receive (Rx) beam derived from previous measurements and include the corresponding measurement results in the report. This approach aims to balance measurement accuracy and efficiency by utilizing information from previous measurements to guide the current measurement process.

[0143] The “Data_collection_reporting_for_NWside_model” parameter is configured to specify the reporting method and label size. If the user device 704 is configured with a specific container, it will use the corresponding container method (reporting format) to report the measurement results. This parameter allows for flexibility in how measurement results are reported and can be customized according to the specific requirements of the network-side AI / ML model training process.

[0144] In addition, base station 702 is configured with a label size parameter. In this example, it is set to n8, meaning that for each time slot of the prediction window, user equipment 704 should include an indicator of the optimal 8 beams in its report. This configuration is particularly important for data collection because it provides the ground truth labels needed to train AI / ML models.

[0145] Upon receiving this configuration and being triggered to generate a report, user equipment 704 follows a specific procedure. First, it uses the first set of RS resources (corresponding to set B) previously used as the Rx beam measurement configuration for receiving Rx. This measurement provides input data for the AI / ML model.

[0146] Next, user equipment 704 uses a second set of RS resources (corresponding to set A) previously used as the Rx beam measurement configuration for receiving Rx. After these measurements, user equipment 704 derives the indicator for the first 8 beams from the second set of RS resources by comparing the measurement results. This step provides ground truth labels for AI / ML model training.

[0147] Finally, user equipment 704 compiles a report that includes two types of information: the measurement results of the first group (set B) and the indicators of the first eight beams in the second group (set A). This report is sent to base station 702 using a specified container method.

[0148] Furthermore, a framework for user equipment capability reporting needs to be established. This framework allows user equipment 704 to inform base station 702 of its specific capabilities related to AI / ML basic beam management functions. By providing this information, the network can optimize its configuration and requests to align with the specific capabilities of each user equipment.

[0149] The user equipment capability report for AI / ML basic beam management includes several feature indicators and their corresponding components, as shown in Table 1 below.

[0150] Table 1: List of features and components in user equipment capabilities

[0151]

[0152] The AI ​​reporting indicator tells User Equipment 704 whether it supports AI / ML basic beam management reporting. User Equipment 704 reports a Boolean value (True / False) to indicate its capability. This allows the network to determine whether User Equipment 704 can participate in the AI / ML basic beam management process.

[0153] The AI ​​BM mode indicator tells the User Equipment (UE) 704 which artificial intelligence / machine learning (AI / ML) beam management use cases it supports. The UE 704 reports a range of supported modes, which may include BM-Case 1 (spatial beam prediction) and BM-Case 2 (temporal beam prediction). For BM-Case 2, the UE 704 also reports the supported UE-side model prediction periodicity (e.g., 20ms, 40ms) and the number of supported future prediction time slots (e.g., 1, 2). This detailed reporting enables the network to understand the UE 704's temporal prediction capabilities, allowing for more effective configuration of temporal beam prediction tasks.

[0154] The Receive (Rx) Usage Type Indicator indicates which Rx beam assumptions the UE 704 supports for measuring signals configured in the network. The UE 704 reports a list of supported Rx usage types. Supported Rx usage types may include BestRx. The UE 704 scans all available Rx beams and uses the best for measurement. Supported Rx usage types may include QuasiRxType1. The UE 704 finds the best Rx beam for a specific signal configuration and then uses this beam for other signals. Supported Rx usage types may include QuasiRxType1. The UE 704 uses the previously best Rx beam for measurement. This information allows the network to understand how the UE 704 performs measurements, which is crucial for the accuracy of AI / ML model inputs and predictions. The UE 704 can also send indicators indicating other possible Rx beam assumptions.

[0155] The UE-side model output report allows UE 704 to report the output of its AI / ML model to base station 702, providing information for optimizing beam selection and overall network performance.

[0156] The ability of the UE-side model to output reports is represented by several parameters: 1. Support for UE-side model output reporting: This is a binary indicator (yes / no) that informs base station 702 and UE704 whether AI / ML-based beam prediction and reporting can be performed. If UE 704 does not implement any AI / ML model, it reports that it does not support this function.

[0157] 2. Top k Reports: This parameter indicates the number of top beams that the UE 704 can report for its AI / ML model output. It is represented as a series of enumerated values ​​(n0, n1, n2, n3, n4, ..., nK), where each value corresponds to a specific number of beams. The special value n0 allows the UE 704 to adaptively determine how many beams to report in each reporting instance, providing flexibility under varying channel conditions. Furthermore, this parameter can define two cases: (1) the UE 704 reports a number, indicating that it only supports reporting this specific number of beams; (2) the UE 704 reports a number, indicating that it supports reporting any number of beams less than or equal to the reported number.

[0158] 3. Report Content: This parameter specifies the types of information that UE 704 can include in its AI / ML model output. Options include: Predicted_best_beam_IDs: Identifiers of the best-predicted beams.

[0159] Predicted_best_beam_RSRPs: The predicted reference signal received power (RSRP) value for the best beam.

[0160] Predicted_best_beam_SINRs: The predicted signal-to-interference-plus-noise ratio (SINR) values ​​for the best beam.

[0161] This allows UE 704 to report not only the beam identifier but also the expected signal quality, enabling more informed decision-making at base station 702.

[0162] 4. Reporting Container Method: This parameter indicates the beam reporting format supported by UE 704 for measurement reporting. Options include: Container_method 1 (Traditional Container): The current beam reporting format.

[0163] Container_method 2 (Beam Reporting Without RI): Current format, but only reports signal strength measurements (RSRP / RSRQ / SINR), not CRI / SSBRI.

[0164] Container_method 3 (Two-part CSI Beam Report): A format that divides the beam report into two parts, similar to the two-part CSI used for reporting PMI.

[0165] Container_method 4 (Append Time Information): A format for appending time information to a traditional beam report.

[0166] The flexibility of the reporting container approach allows for efficient reporting based on different network configurations and requirements. For example, a two-part CSI beamforming report may be advantageous in scenarios where the network needs to process beaming information in stages, while the format of additional timing information may be crucial for timing beam prediction in BM-Case2 scenarios.

[0167] By reporting these capabilities, UE 704 provides base station 702 with a comprehensive understanding of its AI / ML-based beam prediction capabilities. This information enables base station 702 to optimize its configuration requests to align with the specific capabilities of each UE 704 in the network.

[0168] For example, if UE 704 reports that it can predict the beam IDs and RSRPs of the first four beams, base station 702 can be configured to report this information. This detailed prediction can help base station 702 make more informed decisions regarding beam selection and resource allocation, potentially improving overall network performance.

[0169] Furthermore, the UE 704's ability to adaptively determine how many beams to report (n0 in the first k reports) can be useful in dynamic environments. In scenarios where channel conditions change rapidly, the UE 704 may adjust the number of beams reported based on the current channel state, reporting more beams when the channel is complex and fewer when the channel is stable.

[0170] The network-side model's input reporting capability allows UE 704 to report measurements as inputs to the network-side AI / ML model, typically located at base station 702.

[0171] UE 704 reports its ability to report network-side model inputs through several parameters: 1. Support for network-side model input reporting: This is a binary indicator (yes / no) that tells base station 702 and UE704 whether measurements suitable for network-side AI / ML model input can be performed and reported.

[0172] 2. Data Filtering Method: This parameter indicates the methods supported by UE 704 for filtering measurement data before reporting. Supported methods may include: No_filter: UE 704 reports all measurements without filtering.

[0173] Filter_15dB_belowthebest: UE 704 only reports measurements with signal strength higher than (best measured signal strength - 15dB).

[0174] Filter_10dB_belowthebest: User equipment 704 only reports measurement results with signal strength higher than (best measured signal strength - 10dB).

[0175] Filter_best4: User equipment 704 only reports the top 4 measurements with the best signal strength.

[0176] Filter_best8: User equipment 704 only reports the top 8 measurements with the best signal strength.

[0177] 3. Report Container Method: This parameter specifies the beam reporting format supported by user equipment 704 for measurement reporting. Options may include: Container_method1 (legacy container): Current beam report format.

[0178] Container_method2 (beam_report_without_RI): Current format, but only reports signal strength measurements (RSRP / RSRQ / SINR), not CRI / SSBRI.

[0179] Container_method3 (two_part_CSI_beam_report): A format that divides the beam report into two parts, similar to the two-part CSI used to report PMI.

[0180] Container_method4 (time_info_attached): A format for attaching time information to a traditional beam report.

[0181] Data filtering can reduce uplink control information (UCI) overhead. In cases where the network-side AI / ML model may not require very low L1 RSRP measurements, base station 702 can instruct user equipment 704 to omit reporting these low measurements. This filtering reduces the amount of data user equipment 704 needs to report, thereby reducing UCI overhead.

[0182] For example, if base station 702 is configured to use the Filter_15dB_belowthebest method for user equipment 704, user equipment 704 will only report measurement results within 15dB of the strongest measurement signal. This method can effectively eliminate weak signals that may be useless to AI / ML models, while still providing a comprehensive view of stronger signals in the environment.

[0183] The flexibility of the reporting container approach allows for efficient reporting based on different network configurations and requirements. For example, a two-part CSI beam reporting approach may be advantageous when the network needs to process beam information in stages, while the format of the additional timing information may be crucial for timing beam prediction in BM-Case2 scenarios.

[0184] The ability to report data collection for AI / ML foundational beam management allows user devices to report measurement results and real-world data that can be used to train and improve network-side AI / ML models. This feature is crucial for the continuous enhancement of AI / ML foundational beam management systems.

[0185] User equipment 704's ability to report its data collection reports via several parameters: 1. Support for AI BM data collection reporting: This is a binary indicator (yes / no) that informs base station 702 and user equipment 704 whether they are able to perform and report measurement results suitable for AI / ML basic beam management purposes.

[0186] 2. Data Filtering Method: This parameter indicates the methods supported by user equipment 704 for filtering measurement data before reporting. Supported methods may include: No_filter: User equipment 704 reports all measurement results without filtering.

[0187] Filter_15dB_belowthebest: User equipment 704 only reports measurement results with signal strength higher than (best measured signal strength - 15dB).

[0188] Filter_10dB_belowthebest: User equipment 704 only reports measurement results with signal strength higher than (best measured signal strength - 10dB).

[0189] Filter_best4: User equipment 704 only reports the top 4 measurements with the best signal strength.

[0190] Filter_best8: User equipment 704 only reports the top 8 measurements with the best signal strength.

[0191] These filtering methods can help reduce uplink control information (UCI) overhead by limiting the amount of data reported, while still providing valuable information for model training.

[0192] 3. Report Container Method: This parameter specifies the beam reporting format supported by user equipment 704 for measurement reporting. Options may include: Container_method1 (legacy container): Current beam report format.

[0193] Container_method2 (beam_report_without_RI): Current format, but only reports signal strength measurements (RSRP / RSRQ / SINR), not CRI / SSBRI.

[0194] Container_method3 (two_part_CSI_beam_report): A format that divides the beam report into two parts, similar to the two-part CSI used to report PMI.

[0195] Container_method4 (time_info_attached): A format for attaching time information to a traditional beam report.

[0196] These different reporting container methods provide flexibility in data reporting, allowing for efficient information transfer based on different network configurations and AI / ML model requirements.

[0197] 4. Label Size: This parameter indicates the number of labels (true values ​​of the model output) that the user device 704 can report. It is represented as a series of enumerated values ​​(n1, n2, n3, n4, ..., nK), where each value corresponds to the number of specific labels that the user device 704 can report.

[0198] The data collection reporting capability allows the network to collect real-world data to train and optimize its models. When configured for data collection, user device 704 reports two types of information: 1. Measurement results from the first set of RS resources (corresponding to set B), serving as input features for the AI / ML model. 2. An indicator of the best beam measured in the second set of RS resources (corresponding to set A), serving as a ground truth label for the AI / ML model.

[0199] For example, if base station 702 configures user equipment 704 with a tag size of n8, user equipment 704 will report the indicators of the best 8 beams in set A in each reporting instance. This provides a rich real-world dataset for training AI / ML models.

[0200] The flexibility of data filtering and reporting container methods allows networks to strike a balance between the quantity and quality of collected data and the associated UCI overhead. For example, using the Filter_15dB_belowthebest method can help focus on the most important measurements while reducing the amount of data transmitted.

[0201] The collected data can be used to train new models, fine-tune existing models, and validate the performance of AI / ML algorithms in real-world scenarios. This capability helps to enhance beam management overall in 5G and more advanced wireless communication systems, potentially leading to improvements in spectral efficiency and user experience.

[0202] It is understood that the specific order or hierarchy of modules in the disclosed process / flowchart is an illustration of exemplary methods. Depending on design preferences, the specific order or hierarchy of modules in the process / flowchart can be rearranged. Furthermore, certain modules can be combined or omitted. The accompanying method claims present the elements of the various modules in an illustrative order and are not intended to limit one to the specific order or hierarchy presented.

[0203] The foregoing description is intended to enable any person skilled in the art to practice the various embodiments described herein. Various modifications to these embodiments will be apparent to a person skilled in the art, and the general principles defined herein can be applied to other embodiments. Therefore, the claims are not intended to be limited to the embodiments shown herein, but should be consistent with the full scope of the language claims, where reference to a single element does not mean “only one” unless otherwise stated, but rather “one or more.” The term “exemplary” as used herein means “as an example, instance, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as superior to other embodiments. Unless otherwise stated, the term “some” refers to one or more. Phrases such as “at least one A, B, or C,” “one or more A, B, or C,” “at least one A, B, and C,” “one or more A, B, and C,” and “A, B, C, or any combination thereof” include any combination of A, B, and / or C, and may include multiple A, multiple B, or multiple C. Specifically, phrases such as "at least one A, B, or C," "one or more A, B, or C," "at least one A, B, and C," "one or more A, B, and C," and "A, B, C, or any combination thereof" can be A only, B only, C only, A and B, A and C, B and C, or A and B and C, wherein any such combination may include one or more members of A, B, or C. All structural and functional equivalents of the elements of the various forms described herein, whether known or subsequently known to those skilled in the art, are expressly incorporated herein by reference and are intended to be included in the claims. Furthermore, the disclosures herein are not intended to be made public, whether such disclosures are expressly recited in the claims. Terms such as "module," "mechanism," "element," and "device" may not be substitutes for the term "means." Therefore, unless an element expressly uses the term "means," no claim element should be construed as means plus function.

Claims

1. A method of wireless communication for a user equipment, comprising: receiving, from a base station, a beam reporting configuration for artificial intelligence / machine learning (AI / ML) -based beam management, wherein the beam reporting configuration comprises a set of configuration parameters that enables reporting for the AI / ML-based beam management; determining, based on the beam reporting configuration, a set of beams that satisfy a predetermined performance indicator; and reporting, to the base station, the set of beams. 2.The method of claim 1, wherein the set of configuration parameters comprises a mode parameter that indicates spatial beam prediction or temporal beam prediction. 3.The method of claim 2, wherein when the mode parameter indicates temporal beam prediction, the set of configuration parameters further comprises: a parameter that indicates a periodicity of prediction time instances and a number of the prediction time instances; and a parameter that indicates a number of observation time instances. 4.The method of claim 1, wherein the set of configuration parameters comprises a receive (Rx) usage type parameter that indicates an Rx usage type for measurement. 5.The method of claim 1, wherein the set of configuration parameters comprises a model output parameter that indicates that the reporting for the AI / ML-based beam management is an inference model output reporting for a user equipment-side AI / ML model. 6.The method of claim 5, wherein the model output parameter comprises a parameter that indicates a number of inference beams to be reported. 7.The method of claim 1, wherein the set of configuration parameters comprises a model input parameter that indicates that the reporting for the AI / ML-based beam management is an AI / ML model input for a network-side model. 8.The method of claim 7, wherein the model input parameter comprises a parameter that indicates a reporting format. 9.The method of claim 1, wherein the set of configuration parameters comprises a data collection parameter that indicates that the reporting for the AI / ML-based beam management is for data collection. 10.The method of claim 9, wherein the data collection parameter comprises a parameter that indicates a reporting format and a parameter that indicates a label size. 11.The method of claim 1, further comprising: reporting, to the base station, a set of capability parameters, wherein the set of capability parameters indicates whether the user equipment supports the reporting for the AI / ML-based beam management. 12.The method of claim 11, wherein the set of capability parameters comprises a mode feature that indicates which AI / ML-based beam management use cases are supported by the user equipment. 13.The method of claim 12, wherein when the mode feature indicates that the user equipment supports temporal beam prediction, the set of capability parameters further comprises: a component that indicates a periodicity of prediction time slots and a number of the prediction time slots supported by the user equipment. ​ ​ 14. The method of claim 11, wherein the capability parameter set comprises a receive (Rx) usage type feature indicating which Rx beam hypotheses are supported by the user equipment.

15. The method of claim 11, wherein the capability parameter set comprises an output reporting feature defining support for model output reporting of a user equipment side AI / ML model.

16. The method of claim 15, wherein the output reporting feature comprises: a component indicating whether the user equipment supports model output reporting of the user equipment side AI / ML model; a component indicating the user equipment supports reporting a number or a maximum number of beams output by the user equipment side AI / ML model; a component indicating the user equipment supports reporting content output by the user equipment side AI / ML model; and a component indicating the user equipment supports a beam reporting format for the AI / ML based beam management reporting.

17. The method of claim 11, wherein the capability parameter set comprises an input reporting feature indicating whether the user equipment supports model input reporting of a network side AI / ML model.

18. The method of claim 17, wherein the input reporting feature comprises: a component indicating whether the user equipment supports model input reporting of the network side AI / ML model; a component indicating a measurement data filtering method supported by the user equipment for reporting; and a component indicating a beam reporting format supported by the user equipment for measurement reporting.

19. The method of claim 11, wherein the capability parameter set comprises a data collection reporting feature indicating whether the user equipment supports data collection reporting for AI / ML models used for AI / ML based beam management.

20. The method of claim 19, wherein the data collection reporting feature comprises: a component indicating whether the user equipment supports data collection reporting; a component indicating a measurement data filtering method supported by the user equipment for reporting; a component indicating a beam reporting format supported by the user equipment for measurement reporting; and a component indicating a number of labels supported by the user equipment for ground truth model output reporting. ​ ​ ​

Images