Method of beam report configuration for ai / ML based beam management and device thereof
The method and apparatus for temporal beam prediction using AI/ML techniques address the inefficiencies of exhaustive beam sweeping by reducing overhead and latency, enabling proactive beam selection and improving network performance in mmWave wireless communication systems.
Patent Information
- Application Number
- PCT/CN2025/086538
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-04-03
- Filing Date
- 2025-04-01
- Publication Date
- 2025-10-09
AI Technical Summary
Current beam management approaches in mmWave wireless communication systems face high overhead and latency due to exhaustive beam sweeping, lack proactive adjustment to channel changes, and lack standardized AI/ML-based solutions for temporal prediction.
A method and apparatus for temporal beam prediction using AI/ML techniques, enabling selective beam measurement and proactive selection through standardized configuration and reporting mechanisms.
Reduces measurement overhead, enables proactive beam selection, and improves network performance by predicting future optimal beams based on past measurements.
Smart Images

Figure CN2025086538_09102025_PF_FP_ABST
Abstract
Description
METHOD OF BEAM REPORT CONFIGURATION FOR AI / ML BASED BEAM MANAGEMENT AND DEVICE THEREOF
[0001] CROSS REFERENCE TO RELATED APPLICATION
[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 573,562, filed on April 3rd, 2024. Further, this application claims the benefit of U.S. Provisional Application No. 63 / 573,566, filed on April 3rd, 2024. The contents of these applications are incorporated herein by reference.BACKGROUND OF THE INVENTION
[0003] 1. FIELD OF THE INVENTION
[0004] The present invention relates generally to wireless communications, and more particularly to methods and apparatus for beam management in wireless communication systems. Specifically, the invention relates to temporal prediction of optimal beams using AI / ML techniques in cellular networks operating in mmWave frequency bands.
[0005] 2. DESCRIPTION OF THE PRIOR ART
[0006] In modern wireless communication systems, particularly those operating in millimeter wave (mmWave) frequency bands, beamforming has become essential for achieving reliable communication links and high data throughput. Traditional beam management techniques rely on exhaustive beam sweeping procedures, where a base station and user equipment (UE) systematically test all possible beam combinations to identify optimal beam pairs for communication.
[0007] Current 3GPP specifications define various beam management procedures, including beam measurement, beam reporting, and beam switching. These procedures typically involve periodic or aperiodic channel state information (CSI) measurements and reporting. The UE measures reference signals transmitted by the base station using different beams and reports measurements such as Reference Signal Received Power (RSRP) back to the base station. The base station then selects appropriate transmission beams based on these measurements.
[0008] However, existing beam management approaches face several limitations. While some solutions have proposed incorporating machine learning techniques for beam selection, these approaches typically focus on instantaneous beam prediction without considering temporal evolution of channel conditions. Additionally, current 3GPP specifications do not provide standardized frameworks for implementing artificial intelligence (AI) and / or machine learning (ML) based beam management solutions. Hence, there exists a need for an efficient beam management framework.SUMMARY OF THE INVENTION
[0009] An embodiment provides a method of wireless communication, performed by a user equipment (UE) . The method comprises receiving, from a network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of reference signal (RS) resource indicators, performing measurements on a set of RS resources according to the configuration parameter set, generating, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators, and transmitting, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.
[0010] An embodiment provides a user equipment (UE) for wireless communication. The UE comprises an antenna and a processor coupled to the antenna. The processor is used to receive, via the antenna, from the network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of reference signal (RS) resource indicators, perform measurements on a set of RS resources according to the configuration parameter set, generate, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators, and transmit, via the antenna, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.
[0011] An embodiment provides a non-transitory computer-readable medium storing instructions. When executed by user equipment (UE) , the instructions cause the UE to perform operations comprising receiving, from a network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of reference signal (RS) resource indicators, performing measurements on a set of RS resources according to the configuration parameter set, generating, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators, and transmitting, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.
[0012] In some aspects, support for multiple resource setting types to accommodate different network scenarios is provided. For periodic or semi-persistent RS resource settings, the configuration includes periodicity and offset parameters that determine the timing structure for predictions. The time slot index for predictions can be referenced to various events, including when the UE receives the configuration or when it transmits the report. For aperiodic resource settings, the configuration can include separate prediction periodicity and offset parameters specifically for the predicted time slots.
[0013] In some aspects, flexible reference timing is provided, allowing the prediction framework to reference various system events such as configuration reception, resource transmission, or report transmission.
[0014] To the accomplishment of the foregoing and related ends, certain embodiments comprise the features hereinafter fully described and particularly pointed out in the claims. The following description and accompanying drawings set forth in detail certain illustrative aspects of the embodiments. These aspects are indicative, however, of but a few of the various ways in which the principles of the embodiments may be employed, and the present disclosure is intended to include all such aspects and their equivalents. These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.BRIEF DESCRIPTION OF THE DRAWINGS
[0015] FIG. 1 illustrates an exemplary wireless communications system and access network.
[0016] FIG. 2 illustrates block diagrams of a base station in communication with a UE in a wireless network.
[0017] FIG. 3A illustrates an exemplary logical architecture of a distributed RAN according to the embodiments.
[0018] FIG. 3B illustrates an exemplary physical architecture of a distributed RAN according to the embodiments.
[0019] FIG. 4A illustrates an exemplary DL-centric slot according to the embodiments.
[0020] FIG. 4B illustrates an exemplary UL-centric slot according to the embodiments.
[0021] FIG. 5 illustrates an AI / ML model for spatial and temporal domain beam prediction according to the embodiments.
[0022] FIG. 6 illustrates the temporal aspects of beam measurement and prediction in AI / ML-based beam management according to the embodiments.
[0023] FIG. 7 illustrates AI / ML based temporal beam prediction according to the embodiments.
[0024] FIG. 8 illustrates a signaling exchange between a UE 801 and a network 802 according to the embodiments.
[0025] FIG. 9 illustrates a temporal beam prediction framework using periodic RS resource settings according to the embodiments.
[0026] FIG. 10 illustrates an alternative embodiment of the temporal beam prediction framework.
[0027] FIG. 11 illustrates an alternative embodiment of the temporal beam prediction framework.
[0028] FIG. 12 illustrates a flow diagram presenting a method of wireless communication according to the embodiments.DETAILED DESCRIPTION
[0029] The present invention relates to methods and apparatus for temporal beam prediction and reporting in wireless communication systems, particularly in the context of 3rd Generation Partnership Project (3GPP) based mobile communication systems including to Fifth Generation New Radio (5G NR) and beyond. While the embodiments described herein primarily focus on 3GPP wireless networks, the disclosed techniques may be applied to various wireless multiple access systems. Such systems can include Code Division Multiple Access (CDMA) , Frequency Division Multiple Access (FDMA) , Time Division Multiple Access (TDMA) , Orthogonal Frequency Division Multiple Access (OFDMA) , Single Carrier Frequency Division Multiple Access (SC-FDMA) , and Multi-Carrier Frequency Division Multiple Access (MC-FDMA) systems.
[0030] In modern wireless networks, a Radio Access Network (RAN) can provide communication services through its connection to a Core Network (CN) . The RAN may implement various radio access technologies (RATs) , including Evolved Universal Terrestrial Radio Access Network (E-UTRAN) for Long Term Evolution (LTE) , Next Generation Radio Access Network (NG-RAN) for 5G NR, or multi-RAT deployments. Within the RAN, access nodes may be referred to by various terms depending on the specific network deployment, including NodeBs, evolved NodeBs (eNBs) , next Generation NodeBs (gNBs) , RAN nodes, or Transmission Reception Points (TRPs) .
[0031] This disclosure specifically addresses beam management challenges that arise in advanced antenna systems, particularly for millimeter wave (mmWave) frequencies where beamforming is essential for achieving reliable communication links. Current beam management approaches rely on exhaustive beam sweeping procedures where network and user equipment (UE) must test all possible beam combinations. As antenna arrays become more sophisticated and the number of available beams increases, this exhaustive approach can result in prohibitively high overhead.
[0032] The methods and devices described herein can be understood in the context of 3GPP Technical Specifications (TS) and related documents that predate this disclosure. While the techniques are particularly relevant to mmWave deployments, they may be adapted for future wireless communication standards as the technology evolves.
[0033] As outlined in the 3GPP TS, the UE calculates CSI (Channel State Information) parameters in a hierarchical manner, where Layer Indicator (LI) , CQI (Channel Quality Indicator) , PMI (Precoding Matrix Indicator) , and RI (Rank Indicator) are interdependent. Specifically, LI is determined based on the reported CQI, while CQI is conditioned on PMI, RI, and CRI (CSI-RS Resource Indicator) . These dependencies ensure accurate and consistent CSI reporting, allowing the network to make optimal scheduling and beamforming decisions.
[0034] CSI reporting can be categorized into three types: aperiodic, periodic, and semi-persistent. Aperiodic CSI reporting occurs on PUSCH (Physical Uplink Shared Channel) and is triggered by DCI (Downlink Control Information) . Periodic CSI reporting takes place on PUCCH (Physical Uplink Control Channel) at predefined intervals. Semi-persistent CSI reporting is similar to periodic reporting but can be dynamically activated or deactivated using DCI-triggered PUSCH. The CSI-RS (Channel State Information Reference Signal) resources used for channel measurements can be configured in a periodic, semi-persistent, or aperiodic manner, each with different activation and triggering mechanisms.
[0035] The procedures for L1-RSRP (Reference Signal Received Power) and L1-SINR (Signal-to-Interference-plus-Noise Ratio) reporting is also defined in 3GPP TS. The UE computes these values using CSI-RS or SS / PBCH (Synchronization Signal / Physical Broadcast Channel) resources, providing critical feedback for network optimization. L1-RSRP reporting can use a single 7-bit quantized value or differential reporting when multiple CSI-RS resources are measured. L1-SINR reporting follows a similar structure, quantizing values within a predefined range and enabling interference-aware CSI calculations.
[0036] As noted in the preceding sections, existing beam management approaches face several limitations. First, as the number of beams grows, especially with advanced antenna systems, the overhead of exhaustive beam sweeping becomes excessively high, leading to increased latency and reduced system efficiency. Additionally, current beam management approaches are primarily reactive, adjusting to changes in channel conditions only after they occur. This can result in suboptimal performance, particularly during rapid channel fluctuations in mobile environments or areas with dynamic blockage. The extensive beam sweeping also consumes considerable time and energy, negatively affecting both network performance and UE battery life. Furthermore, existing specifications lack the ability to predict future optimal beams based on past measurements, restricting the system’s capacity to proactively adjust to changing conditions.
[0037] Recent developments have started to explore the integration of artificial intelligence and / or machine learning (AI / ML) into RAN functionalities. However, there is still a lack of defined mechanisms for configuring and reporting AI / ML based beam predictions, particularly for cases involving temporal predictions. To address this gap, there is a clear need for an efficient beam management framework that: reduces measurement overhead by selectively measuring beams; enables proactive beam selection through temporal prediction; provides standardized configuration and reporting mechanisms for AI / ML based solutions; and ensures compatibility with existing 3GPP specifications and procedures. This disclosure presents a solution to these challenges by introducing an innovative beam report configuration framework, specifically designed for temporal AI / ML based beam management. This framework facilitates more efficient and proactive beam selection in modern wireless communication systems.
[0038] FIG. 1 illustrates an exemplary wireless communications system and access network 100. The wireless communication system (also referred to as a wireless wide area network (WWAN) ) includes base stations 102, UEs 104, an Evolved Packet Core (EPC) 160, and an additional core network 190 (such as a 5G Core (5GC) ) . The base stations 102 can be macrocells (high-power cellular base stations) or small cells (low-power cellular base stations) . Macrocells refer to large-scale base stations, while small cells include femtocells, picocells, and microcells.
[0039] The base stations 102 designed for 4G LTE operation (collectively known as Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN) ) may connect to the EPC 160 via backhaul links 132 (e.g., S1 interface) . Similarly, base stations 102 configured for 5G NR (collectively designated as Next Generation RAN (NG-RAN) ) can interface with core network 190 through backhaul links 184. These base stations can perform numerous functions beyond basic connectivity, which may include: user data transfer, radio channel encryption and decryption, integrity protection, header compression, mobility management (e.g., handover, dual connectivity) , interference coordination between cells, connection management, load distribution, handling of non-access stratum (NAS) messages, NAS node selection, network synchronization, RAN resource sharing, multimedia broadcast services (MBMS) , subscriber tracking, RAN information management (RIM) , paging services, position determination, and emergency alert distribution. Additionally, base stations 102 may communicate with each other either directly or indirectly (e.g., through the EPC 160 or core network 190) using backhaul links 134 (e.g., X2 interface) , which can be implemented as either wired or wireless connections.
[0040] The base stations 102 may wirelessly communicate with the UEs 104. Each base station 102 can provide coverage for a specific geographic area 110, and there may be overlapping coverage areas 110. For instance, the small cell 102' may have a coverage area 110' that overlaps with the coverage area 110 of one or more macro base stations 102. A network that includes both small cells and macrocells may be referred to as a heterogeneous network. This type of network may also include Home Evolved Node Bs (HeNBs) , which can serve a restricted group known as a closed subscriber group (CSG) . The communication links 120 between the base stations 102 and the UEs 104 may involve uplink (UL) (also known as reverse link) transmissions from a UE 104 to a base station 102 and / or downlink (DL) (also known as forward link) transmissions from a base station 102 to a UE 104.
[0041] The communication links 120 may use multiple-input and multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming, and / or transmit diversity. The communication links may be through one or more carriers. For each carrier, the base stations 102 and / or UEs 104 may utilize spectrum with bandwidths of various sizes (e.g., 5, 10, 15, 20, 100, 400, etc. MHz) up to 7 MHz per carrier. Through carrier aggregation, these individual carriers can be combined to achieve a total bandwidth of Y×X MHz, where X represents the number of component carriers, enabling higher data rates for transmission in each direction. The carriers may or may not be adjacent to one another. The allocation of carriers can be asymmetric with respect to DL and UL, meaning that more or fewer carriers may be allocated for DL than for UL. The component carriers may include a primary component carrier and one or more secondary component carriers. A primary component carrier may be referred to as a primary cell (PCell) , while a secondary component carrier can be referred to as a secondary cell (SCell) .
[0042] Certain UEs 104 may communicate with each other using a device-to-device (D2D) communication link 158. This D2D communication link 158 can utilize the DL / UL WWAN spectrum and may operate on one or more sidelink channels, such as the physical sidelink broadcast channel (PSBCH) , physical sidelink discovery channel (PSDCH) , physical sidelink shared channel (PSSCH) , and physical sidelink control channel (PSCCH) . D2D communication can occur through various wireless D2D communication systems, including, for example, FlashLinQ, WiMedia, Bluetooth, ZigBee, Wi-Fi based on the IEEE 802.11 standard, LTE, or NR.
[0043] The wireless communications system may also include a Wi-Fi access point (AP) 150, which communicates with Wi-Fi stations (STAs) 152 via communication links 154 in the 5 GHz unlicensed frequency spectrum. When operating in an unlicensed frequency spectrum, the STAs 152 and AP 150 may perform a clear channel assessment (CCA) before communicating to check if the channel is available.
[0044] The small cell 102' may operate in either licensed or unlicensed frequency spectrums, or both. When using an unlicensed frequency spectrum, the small cell 102' can employ NR and utilize the same 5 GHz unlicensed frequency spectrum as the Wi-Fi AP 150. By employing NR in an unlicensed frequency spectrum, the small cell 102' may enhance coverage and / or increase capacity of the access network.
[0045] A base station 102, which may be a small cell 102' or a large cell (e.g., macro base station) , can include an eNB, gNodeB (gNB) , or another type of base station. Some base stations, such as gNB 180, may operate in a traditional sub 6 GHz spectrum, in millimeter wave (mmWave) frequencies, and / or near mmWave frequencies when communicating with the UE 104. When the gNB 180 operates in mmWave or near mmWave frequencies, it can be referred to as an mmWave base station.
[0046] Extremely high frequency (EHF) is part of the RF in the electromagnetic spectrum. EHF has a range of 30 GHz to 300 GHz and a wavelength between 1 millimeter and 10 millimeters. Radio waves in this band may be referred to as millimeter waves. Near mmWave may extend down to a frequency of 3 GHz with a wavelength of 100 millimeters. The super high frequency (SHF) band extends between 3 GHz and 30 GHz, also referred to as centimeter wave.
[0047] Communications using the mmWave and / or near mmWave radio frequency band (e.g., 3 GHz-300 GHz) can experience extremely high path loss and have a short range. The mmWave base station 180 may utilize beamforming 182 with the UE 104 to compensate for these limitations.
[0048] The base station 180 may transmit a beamformed signal to the UE 104 in one or more transmit directions 108a. The UE 104 may receive the beamformed signal from the base station 180 in one or more receive directions 108b. The UE 104 may also transmit a beamformed signal to the base station 180 in one or more transmit directions 108c. The base station 180 may receive the beamformed signal from the UE 104 in one or more receive directions 108c'. This bidirectional beamforming capability is essential for establishing reliable mmWave communications. The base station 180 / UE 104 may perform beam training to determine the best receive and transmit directions for each of the base station 180 / UE 104. The transmit and receive directions for the base station 180 may or may not be the same. The transmit and receive directions for the UE 104 may or may not be the same.
[0049] The 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. The MME 162 can communicate with a Home Subscriber Server (HSS) 174. The MME 162 is the control node that processes signaling between the UEs 104 and the EPC 160. Generally, the MME 162 provides bearer and connection management.
[0050] All user Internet protocol (IP) packets may transfer through the Serving Gateway 166, which itself connects to the PDN Gateway 172. The PDN Gateway 172 can provide UE IP address allocation as well as other functions. The PDN Gateway 172 and the BM-SC 170 may connect to the IP Services 176. The IP Services 176 can include the Internet, an intranet, an IP Multimedia Subsystem (IMS) , a PS Streaming Service, and / or other IP services.
[0051] The BM-SC 170 may provide functions for MBMS user service provisioning and delivery. The BM-SC 170 can serve as an entry point for content provider MBMS transmission, may be used to authorize and initiate MBMS Bearer Services within a public land mobile network (PLMN) , and can be used to schedule MBMS transmissions. The MBMS Gateway 168 may distribute MBMS traffic to the base stations 102 belonging to a Multicast Broadcast Single Frequency Network (MBSFN) area broadcasting a particular service, and can be responsible for session management (start / stop) and for collecting eMBMS related charging information.
[0052] 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 can communicate with a Unified Data Management (UDM) 196. The AMF 192 is the control node that processes signaling between the UEs 104 and the core network 190. Generally, the SMF 194 may provide QoS flow and session management. The UPF 195 can serve as the pathway for all user IP packets. In addition to providing UE IP address allocation, the UPF 195 may perform various other functions. Connectivity between the UPF 195 and IP Services 197 is possible. The IP Services 197 may include the Internet, an intranet, an IP Multimedia Subsystem (IMS) , a PS Streaming Service, and potentially additional IP services.
[0053] The base station may also be known as a gNB, Node B, eNB, an access point, a base transceiver station, a radio base station, a radio transceiver, a transceiver function, a basic service set (BSS) , an extended service set (ESS) , a transmit reception point (TRP) , or other suitable terminology. The base station 102 can function as an access point to the EPC 160 or core network 190 for a UE 104.
[0054] Although this document discusses 5G NR technology, the concepts presented are also relevant to various other wireless communication standards and technologies. These include 4G LTE, LTE-A, CDMA, GSM, and may extend to future generations of wireless and radio access technologies that will evolve from current standards.
[0055] FIG. 2 illustrates block diagrams of a base station 210 in communication with a UE 250 in a wireless network. In the DL, IP packets from the EPC 160 may be provided to a controller / processor 275. The controller / processor 275 can implement layer 3 and layer 2 functionality. Layer 3 includes a radio resource control (RRC) layer, and layer 2 may include a packet data convergence protocol (PDCP) layer, a radio link control (RLC) layer, and a medium access control (MAC) layer.
[0056] The controller / processor 275 can manage RRC layer functions related to broadcasting system information (e.g., MIB, SIBs) , RRC connection control (e.g., RRC connection paging, establishment, modification, and release) , radio access technology (RAT) mobility, and the configuration of measurements for UE measurement reporting. The controller / processor 275 may also handle PDCP layer functions related to header compression / decompression, security (including ciphering, deciphering, integrity protection, and integrity verification) , and handover support; RLC layer functions for transferring upper layer packet data units (PDUs) , error correction through ARQ, concatenation, segmentation, and reassembly of RLC service data units (SDUs) , re-segmentation of RLC data PDUs, and reordering of RLC data PDUs; and MAC layer functions that include mapping logical channels to transport channels, multiplexing MAC SDUs onto transport blocks (TBs) , demultiplexing MAC SDUs from TBs, reporting scheduling information, error correction through HARQ, priority handling, and logical channel prioritization.
[0057] The transmit (TX) processor 216 and the receive (RX) processor 270 may perform layer 1 functions related to various signal processing tasks. Layer 1, which includes the physical (PHY) layer, involves error detection on transport channels, forward error correction (FEC) coding / decoding of transport channels, interleaving, rate matching, mapping onto physical channels, modulation / demodulation of physical channels, and MIMO antenna processing.
[0058] The TX processor 216 handles mapping to signal constellations based on various modulation schemes (e.g., binary phase-shift keying (BPSK) , quadrature phase-shift keying (QPSK) , M-phase-shift keying (M-PSK) , M-quadrature amplitude modulation (M-QAM) ) . The coded and modulated symbols are then divided into parallel streams. Each stream is mapped to an OFDM subcarrier, multiplexed with a reference signal (e.g., pilot) in the time and / or frequency domain, and then combined using an Inverse Fast Fourier Transform (IFFT) to produce a physical channel carrying a time-domain OFDM symbol stream.
[0059] The OFDM stream can be spatially precoded to generate multiple spatial streams. Channel estimates from a channel estimator 274 help determine the coding and modulation scheme, as well as assist in spatial processing. These estimates are derived from a reference signal and / or feedback on channel conditions transmitted by the UE 250. Each spatial stream is sent to a different antenna 220 via a separate transmitter 218TX. Each transmitter 218TX modulates an RF carrier with its respective spatial stream for transmission.
[0060] At the UE 250, each receiver 254RX receives a signal through its respective antenna 252. The receiver 254RX recovers the information modulated onto the RF carrier and passes it to the RX processor 256. The TX processor 268 and RX processor 256 handle layer 1 functionality associated with various signal processing tasks. The RX processor 256 performs spatial processing on the received information to recover any spatial streams intended for the UE 250. If multiple spatial streams are directed to the UE 250, the RX processor 256 combines them into a single OFDM symbol stream. The RX processor 256 then converts the time-domain OFDM symbol stream into the frequency domain using a Fast Fourier Transform (FFT) . The frequency-domain signal consists of a separate OFDM symbol stream for each subcarrier of the OFDM signal.
[0061] The symbols on each subcarrier, and the reference signal, can be recovered and demodulated by determining the most likely signal constellation points transmitted by the base station 210. These soft decisions may be based on channel estimates computed by the channel estimator 258. The soft decisions can then be decoded and deinterleaved to recover the data and control signals that were originally transmitted by the base station 210 on the physical channel. The data and control signals may then be provided to the controller / processor 259, which implements layer 3 and layer 2 functionality.
[0062] The controller / processor 259 may be linked to a memory 260 that holds program code and data, with the memory 260 often referred to as a computer-readable medium. In the UL, the controller / processor 259 handles tasks such as demultiplexing between transport and logical channels, packet reassembly, deciphering, header decompression, and control signal processing to recover IP packets from the EPC 160. Additionally, the controller / processor 259 is responsible for error detection through an ACK and / or NACK protocol to support HARQ operations.
[0063] Similar to the DL transmission functionality provided by the base station 210, the controller / processor 259 handles RRC layer functions related to system information (e.g., MIB, SIBs) acquisition, RRC connections, and measurement reporting; PDCP layer functions for header compression / decompression and security (including ciphering, deciphering, integrity protection, and integrity verification) ; RLC layer functions for transferring upper layer PDUs, error correction through ARQ, concatenation, segmentation, and reassembly of RLC SDUs, re-segmentation of RLC data PDUs, and reordering of RLC data PDUs; and MAC layer functions that include mapping logical channels to transport channels, multiplexing MAC SDUs onto TBs, demultiplexing MAC SDUs from TBs, scheduling information reporting, error correction through HARQ, priority handling, and logical channel prioritization.
[0064] Channel estimates derived by a channel estimator 258 from a reference signal or feedback transmitted by the base station 210 may assist the TX processor 268 in selecting the appropriate coding and modulation schemes, and in facilitating spatial processing. The spatial streams generated by the TX processor 268 can be transmitted to different antennas 252 via separate transmitters 254TX, with each transmitter modulating an RF carrier with its respective spatial stream. The UL transmission is processed at the base station 210 in a similar manner to the receiver function at the UE 250, where each receiver 218RX receives a signal through its respective antenna 220, recovers the modulated information, and sends it to a RX processor 270.
[0065] The controller / processor 275 may be linked to a memory 276 that stores program codes and data, with the memory 276 typically referred to as a computer-readable medium. In the UL, the controller / processor 275 manages demultiplexing between transport and logical channels, packet reassembly, deciphering, header decompression, and control signal processing to recover IP packets from the UE 250. These IP packets are then forwarded to the EPC 160. The controller / processor 275 is also responsible for error detection using an ACK and / or NACK protocol to support HARQ operations.
[0066] New Radio (NR) may refer to radios designed to operate on a new air interface (different from OFDMA-based interfaces) or a fixed transport layer (other than IP) . NR may use OFDM with a cyclic prefix (CP) on both the UL and DL, and supports half-duplex operation using time division duplexing (TDD) . NR may include services like Enhanced Mobile Broadband (eMBB) targeting wide bandwidth (e.g., 80 MHz or more) , mmWave for high carrier frequencies (e.g., 60 GHz) , massive Machine-Type Communications (mMTC) for non-backward compatible MTC techniques, and / or mission-critical communications targeting ultra-reliable low latency communications (URLLC) .
[0067] A single component carrier bandwidth of 100 MHz may be supported. In one example, NR resource blocks (RBs) can span 12 sub-carriers with a sub-carrier bandwidth of 60 kHz over a 0.25 ms duration or a bandwidth of 30 kHz over a 0.5 ms duration (similarly, 50MHz BW for 15kHz SCS over a 1 ms duration) . Each radio frame may consist of 10 subframes (10, 20, 40 or 80 NR slots) with a length of 10 ms. Each slot can indicate a link direction (i.e., DL or UL) for data transmission and the link direction for each slot may be dynamically switched. Each slot can include DL / UL data as well as DL / UL control data. UL and DL slots for NR may be described in more detail below.
[0068] The NR RAN may include a central unit (CU) and distributed units (DUs) . A NR BS (e.g., gNB, 5G Node B, Node B, transmission reception point (TRP) , access point (AP) ) can correspond to one or multiple BSs. NR cells may be configured as access cells (ACells) or data only cells (DCells) . For example, the RAN (e.g., a central unit or distributed unit) can configure the cells. DCells may be cells used for carrier aggregation or dual connectivity and may not be used for initial access, cell selection / reselection, or handover. In some cases DCells can operate without transmitting synchronization signals (SS) , while in other cases they may transmit SS. NR BSs may transmit downlink signals to UEs indicating the cell type. Based on the cell type indication, the UE can communicate with the NR BS. For example, the UE may determine which NR BSs to consider for cell selection, access, handover, and / or measurement based on the indicated cell type.
[0069] FIG. 3A illustrates an exemplary logical architecture of a distributed RAN 300. A 5G access node 306 may include an access node controller (ANC) 302. The ANC can be a 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 can also terminate at the ANC. The ANC may include one or more TRPs 308 (which may also be referred to as BSs, NR BSs, Node Bs, 5G NBs, APs, or some other term) . As described above, a TRP can be used interchangeably with "cell. "
[0070] The TRPs 308 may function as a distributed unit (DU) . The TRPs can connect to one ANC (ANC 302) or more than one ANC (not illustrated) . For example, for RAN sharing, radio as a service (RaaS) , and service specific ANC deployments, the TRP may connect to more than one ANC. A TRP can include one or more antenna ports. The TRPs may be configured to individually (e.g., dynamic selection) or jointly (e.g., joint transmission) serve traffic to a UE.
[0071] The local architecture of the distributed RAN 300 can be used to illustrate fronthaul definition. The architecture may be defined to support fronthauling solutions across different deployment types. For example, the architecture can be based on transmit network capabilities (e.g., bandwidth, latency, and / or jitter) . The architecture may share features and / or components with LTE. According to aspects, the next generation AN (NG-AN) 310 can support dual connectivity with NR. The NG-AN may share a common fronthaul for LTE and NR. The architecture can enable cooperation between and among TRPs 308. For example, cooperation may be preset within a TRP and / or across TRPs via the ANC 302.
[0072] In certain implementations, no inter-TRP interface may be needed or present. In certain implementations, a dynamic configuration of split logical functions can be present within the architecture of the distributed RAN 300. The PDCP, RLC, MAC protocol may be adaptably placed at the ANC or TRP.
[0073] FIG. 3B illustrates an exemplary physical architecture of a distributed RAN 310. A centralized core network unit (C-CU) 312 may host core network functions. The C-CU can be centrally deployed. C-CU functionality may be offloaded (e.g., to advanced wireless services (AWS) ) , in an effort to handle peak capacity. A centralized RAN unit (C-RU) 314 may host one or more ANC functions. Optionally, the C-RU can host core network functions locally. The C-RU may have distributed deployment. The C-RU can be positioned closer to the network edge. A distributed unit (DU) 316 may host one or more TRPs. The DU can be located at edges of the network with radio frequency (RF) functionality.
[0074] FIG. 4A illustrates an exemplary DL-centric slot 400. This DL-centric slot 400 may include a control portion 402, which typically appears in the initial or beginning part of the slot. The control portion 402 can contain various scheduling and / or control information corresponding to different sections of the DL-centric slot. In certain configurations, the control portion 402 may be implemented as a physical DL control channel (PDCCH) .
[0075] The DL-centric slot 400 also includes a DL data portion 404, which is often referred to as the payload of the DL-centric slot. This DL data portion 404 encompasses the communication resources used to transmit DL data from the scheduling entity (e.g., UE or BS) to the subordinate entity (e.g., UE) . In some cases, the DL data portion 404 may be a physical DL shared channel (PDSCH) .
[0076] Additionally, the DL-centric slot 400 may include a common UL portion 406. This portion can be called an UL burst, a common UL burst, or other suitable terms. The common UL portion 406 may carry feedback information related to other sections of the DL-centric slot. For instance, the common UL portion 406 can provide feedback on the control portion 402. Examples of such feedback information include an ACK signal, a NACK signal, a HARQ indicator, or other relevant data. The common UL portion 406 can also carry additional or alternative information, such as data related to random access channel (RACH) procedures, scheduling requests (SRs) , or other suitable types of information.
[0077] As shown in FIG. 4A, the end of the DL data portion 404 may be separated in time from the beginning of the common UL portion 406. This time separation is commonly referred to as a gap, guard period, guard interval, or other suitable terms. This separation allows sufficient time for the switch-over from DL communication (e.g., reception by the subordinate entity, such as the UE) to UL communication (e.g., transmission by the subordinate entity, such as the UE) .
[0078] FIG. 4B illustrates an exemplary UL-centric slot 410. The UL-centric slot 410 may include a control portion 412, which may appear in the initial or beginning part of the UL-centric slot. The control portion 412 in FIG. 4B may be similar to the control portion 402 described above with reference to FIG. 4A. The UL-centric slot 410 may also include an UL data portion 414, which is sometimes referred to as the payload of the UL-centric slot. The UL data portion 414 represents the communication resources used to transmit UL data from the subordinate entity (e.g., UE) to the scheduling entity (e.g., UE or BS) . In some configurations, the control portion 412 may be a physical DL control channel (PDCCH) .
[0079] As illustrated in FIG. 4B, the end of the control portion 412 may be separated in time from the beginning of the UL data portion 414. This time separation is commonly referred to as a gap, guard period, guard interval, or other suitable terms. This separation allows time for the switch-over from DL communication (e.g., reception by the scheduling entity) to UL communication (e.g., transmission by the scheduling entity) .
[0080] The UL-centric slot 410 may also include a common UL portion 416, which in FIG. 4B may be similar to the common UL portion 506 described above with reference to FIG. 4A. The common UL portion 416 can also carry additional or alternative information, such as channel quality indicator (CQI) feedback, sounding reference signals (SRSs) , or other suitable types of information.
[0081] In some cases, two or more subordinate entities (e.g., UEs) may communicate with each other using sidelink signals. Real-world applications of sidelink communications can include public safety, proximity services, UE-to-network relaying, vehicle-to-vehicle (V2V) communications, Internet of Everything (IoE) communications, IoT communications, mission-critical mesh, and other suitable applications. Generally, a sidelink signal refers to a communication sent directly between subordinate entities (e.g., from UE1 to UE2) without routing through the scheduling entity (e.g., UE or BS) , although the scheduling entity may be used for scheduling and / or control purposes. In some cases, sidelink signals may be communicated using a licensed spectrum, unlike wireless local area networks, which typically use an unlicensed spectrum.
[0082] FIG. 5 illustrates an AI / ML (artificial intelligence / machine learning) model for spatial and temporal domain beam prediction. In this example, the base station 502 simultaneously transmits beams 511-534 in various directions via channel 580. After identifying incoming beams, the UE 504 can compute Layer 1 Reference Signal Received Power (L1-RSRP) for each beam. L1-RSRP is the average received power of the resource elements that carry the secondary synchronization signals or channel state information reference signals (CSI-RS) .
[0083] Machine learning algorithms are used to analyze the history of signal strengths from a subset of beams and attempt to find patterns or trends in the data. This helps predict the signal strengths of the remaining unmeasured beams. By identifying patterns in historical data from the subset of beams, the algorithm can predict signal strengths of the other beams even as the UE is moving.
[0084] In this example, the base station 502 is equipped with multiple antennas and is capable of simultaneously radiating 24 different beams 511-534 in various directions. The UE 504, which moves from time to time, may be equipped with its own antenna and periodically measures channel indicators such as RSRP from 4 beams (e.g. beams 515, 516, 529, and 530) selected from the 24 beams radiated by the base station 502. The set of beams (e.g. beams 515, 516, 529, and 530) measured as AI / ML input (sensing beams) is referred to as beam Set B. The set of beams (e.g. the 24 beams) that is being predicted as AI / ML output (usually communication beams) is referred to as beam Set A.
[0085] The measurements collected from these 4 beams are saved over time as historical data. The historical data captures how the channel indicators for the subset of beams change over time, capturing how the UE 504 interacts with those beams. The UE 504 may be configured with a historical data time window 560 during which measurements are stored in the UE 504. In this example, the current time is t_0. The historical data time window 560 spans from time t_ (-3) to t_0. Measurement data for the subset of beams 515, 516, 529, and 530 obtained during the historical data time window 560 are stored in the UE 504 and used as input to the AI / ML model 550 to predict measurements of unmeasured beams at the current time t_0 as well as measurements of all the beams at future times t_1, t_2.
[0086] The historical data serves as input to a machine learning algorithm to predict channel indicators for unmeasured beams, guiding the UE 504 on which beams to focus when it needs to communicate with the base station 502. The algorithm can thus predict channel indicators for all beams based on analyzing patterns and trends in the historical data from the subset of beams.
[0087] Another set of beams (e.g. beams 511, 521, 524, and 534) that are not normally measured under regular circumstances can be sampled periodically and their channel indicators recorded. This can be used to validate if the algorithm’s predictions match actual performance while also updating the AI / ML model. Periodic measurements help improve the algorithm by updating its weights and parameters. As the machine learning algorithm matures, its predictions of the best beams will become increasingly accurate. When the UE 504 initiates communication, it can select the UE transmit or receive beam 570 that is likely to yield superior signal quality (e.g. the best beam) based on the prediction.
[0088] Rather than identifying a single best beam, the method predicts the top-k beams that are likely to have the highest channel indicators. In many cases, focusing on the top-k beams can provide excellent accuracy. The top-k beam prediction is achieved by estimating them based on the top-k channel indicator values. This aligns very well with real-world communication needs, improving system performance.
[0089] The main output of a classification-based AI / ML model includes identifiers (IDs) of the predicted top-k best beams for communication, along with corresponding predicted confidence scores or predicted RSRP for each beam. These beams are determined to be most suitable for communication based on expected signal strength and reliability. For example, if k is set to 5, the model might predict that the five best communication beams in the entire beam set are the beams numbered 532, 530, 534, 528, and 533. This prediction enables the UE 504 to make an informed decision on which beams the base station 502 and UE 504 should use to communicate at any given time, optimizing performance based on signal strength and likelihood of successfully transmitting data.
[0090] Furthermore, the output of a regression-based AI / ML model includes predicted RSRP values for each communication beam in beam Set A. This predicted output directly estimates the expected signal strength of each individual beam in the communication set, allowing the UE 504 to select beams more granularly based on the predicted RSRP values.
[0091] Beamforming, a technique for enhancing data rates and reliability in 5G and beyond wireless communication, especially in millimeter wave (mmWave) frequencies, enables a base station, such as the base station 502, to focus its signal transmission and reception toward a specific user equipment (UE) , such as the UE 504. This targeted approach improves signal quality and reduces interference. To establish an optimal beam connection, the base station 502 and the UE 504 need to identify the best beams to transmit and receive data, a process known as beam management. Traditional beam management often involves exhaustive beam sweeping, where the base station 502 and the UE 504 systematically scan through all available beam directions to find the best one. However, this method becomes inefficient and resource-intensive as the number of antennas and beams increases, leading to significant overhead.
[0092] AI / ML-based beam management, in contrast, offers a more agile and efficient alternative to exhaustive beam sweeping. By using the power of machine learning, this approach predicts the optimal beams for communication based on analyzing patterns in historical data obtained from a subset of beams.
[0093] As illustrated in FIG. 5, rather than measuring all 24 beams, the UE 504 selectively measures the L1-RSRP from a smaller subset of beams, denoted as beam Set B, which serves as input to an AI / ML model 550. By analyzing patterns and trends in the historical data from this subset of beams, the AI / ML model 550 predicts the L1-RSRP values for the remaining unmeasured beams in beam Set A.
[0094] The AI / ML model 550 may take various forms, such as classification-based or regression-based models. A classification model predicts the top-k best beams for communication and provides associated confidence scores. A regression model directly estimates the RSRP values for each communication beam in beam Set A. Regardless of the chosen model, this predictive capability significantly reduces the need for exhaustive measurements, minimizing overhead and enhancing efficiency.
[0095] Temporal beam prediction, a key aspect of AI / ML-based beam management, predicts future optimal beam indices based on historical beam measurements. The UE 504 uses measurements taken over a historical data time window 560, allowing the model 550 to learn temporal patterns and anticipate future beam conditions.
[0096] In essence, AI / ML-based beam management provides a faster and more efficient way to obtain the best beam information, optimizing beam selection for communication between a base station and a UE, especially in dynamic environments where beam conditions may change rapidly.
[0097] FIG. 6 illustrates the temporal aspects of beam measurement and prediction in AI / ML-based beam management. The figure shows two sequences: a measurement sequence 661 and a prediction sequence 662.
[0098] The measurement sequence 661 represents K measurement instances, where the UE 504 conducts beam measurements. These measurements are typically performed on a subset of beams (e.g., beam Set B) , which serves as input to the AI / ML model. These measurements provide the power measurements of the sensing beams, which are used to infer the optimal communication beams.
[0099] The prediction sequence 662 represents F prediction instances, where the AI / ML model predicts the future optimal beam indices. These predictions are based on the beam measurements from the previous time steps, specifically the measurements taken during the K measurement instances.
[0100] This temporal structure aligns with the concept of temporal beam prediction, where the goal is to predict future optimal beam indices using the beam measurements on the sensing beams from previous time steps. The RSRP of one or more beams can be predicted with an input of historical RSRPs.
[0101] That is, the measurement sequence 661 corresponds to the time instances for which the UE 504 reports actual measurements. The prediction sequence 662 represents the future time instances for which the AI / ML model (either at the UE 504 or the base station 502) predicts beam conditions.
[0102] In advanced wireless communication systems, such as 5G and beyond, beamforming may enhance data rates and reliability, particularly in millimeter wave (mmWave) frequencies. The base station 502 and the UE 504 need to identify the optimal beams for communication, a process known as beam management. Traditional beam management often involves exhaustive beam sweeping, which becomes inefficient as the number of available beams increases. To address this challenge, AI / ML-based beam management has been proposed as a more efficient alternative.
[0103] AI / ML-based beam management requires new reporting configurations to accommodate various scenarios and purposes of beam reporting. The reporting format used to carry the reported beam measurement results can differ based on the location of the AI / ML model (either at the network side or the UE side) and the specific requirements of the beam management task.
[0104] When the AI / ML model is located at the network side (e.g., the base station 502) , the UE 504 needs to be configured to report measurements of beams in Set B. The methods for reporting these measurements may vary depending on the specific AI / ML model design. For temporal beam prediction, the UE 504 may need to include a temporal ID in the report to indicate the time of measurement.
[0105] Conversely, when the AI / ML model is located at the UE side, the UE 504 needs to be configured to report the inferred best beams among the communication beams. In this case, the UE 504 may report a variable number of predicted best beams according to the AI / ML model output design. This number is not known beforehand by the network, necessitating a new reporting format for measurements. Similar to the network-side model case, temporal beam prediction may require the inclusion of a temporal ID in the report.
[0106] FIG. 7 illustrates AI / ML based temporal beam prediction according to the embodiments. UE 701 uses AI / ML based models 703 to predict future beams based on set B measurement reports 710, 720, 730 from the network (NW) 702.
[0107] The communication framework begins when the network 702 transmits beam report configuration information to the UE 701, establishing parameters for both measurement and prediction operations. This configuration defines the reference signal (RS) resource indicators that constitute Set B, along with RS resource settings that establish the observation and prediction windows shown in the diagram. The UE 701 utilizes these configuration parameters to determine when to perform measurements and for which future time slots predictions should be generated.
[0108] During the observation window spanning time slots T1, T2, and T3, the UE 701 performs beam measurements 710, 720, and 730 on the Set B RS resources (i.e., Set B beams) received from the network 702. These measurements may follow a specific periodicity and offset defined in the RS resource settings. In another implementation, these measurements may be aperiodic as defined in the RS resource settings. The UE 701 collects these measurements systematically to build a temporal profile of beam performance, which serves as input to the AI / ML model 703. The number of measurement instances in the observation window may be predetermined or configured through the received RS resource settings.
[0109] The AI / ML model 703 processes these historical measurements to generate predictions for optimal beams in future time slots. The model analyzes patterns in the measurement data to identify which RS resource indicators will maximize performance metrics such as reference signal received power (RSRP) , Signal to Interference plus Noise Ratio (SINR) , UE throughput or other performance indicators at specific future time slots (i.e., predicted time slots) . The prediction window, comprising time slots T4 and T5, represents the set of predicted time slots for which the UE generates predictions. The quantity of these predicted time slots can be determined by the configuration received from the network 702.
[0110] Following the prediction process, the UE 701 compiles a beam report including the predicted RS resource indicators for each time slot in the prediction window. This report associates each predicted beam with its corresponding predicted time slot, enabling the network to prepare optimal beam configurations in advance.
[0111] This framework demonstrates an efficient approach to beam management and prediction, where the first set of RS resource indicators (Set B) used for measurements may be fewer than the second set of RS resource indicators (Set A) from which predictions are made. This reduces measurement overhead while maintaining prediction accuracy across the full range of available beams. The configuration parameters may further allow for adaptation based on specific deployment scenarios, supporting different RS resource setting types including prediction periodicity parameters and prediction offset parameters that determine the timing of the predicted time slots.
[0112] FIG. 8 illustrates a signaling exchange between a UE 801 and a network 802 according to the embodiments. The illustrated sequence comprises a plurality of operations executed in a predetermined order to effectuate temporal beam prediction utilizing artificial intelligence / machine learning (AI / ML) methodologies.
[0113] In step 810, the process is initiated when the NW 802 transmits a beam report configuration to the UE 801. The beam report configuration may comprise parameters including: a reportQuantity parameter indicating the type of reporting required; a set of reference signal (RS) resource indicators; a parameter QN specifying the quantity of future time slots for which predictions are to be generated; and timing parameters associated with RS resource settings. The beam report configuration may further comprise periodicity parameters and offset parameters relevant to the prediction framework.
[0114] In step 820, upon reception of the beam report configuration, the UE 801 proceed to determine the future time slots for which beam predictions should be performed. The determination of the time slots is executed in accordance with the received configuration parameters. In certain embodiments, the UE 801 derives the index of the future time slots based on the periodicity and offset parameters included in the RS resource settings. In certain embodiments, the UE 801 derives the index of the future time slots based on the periodicity parameters and offset parameters relevant to the prediction framework.
[0115] Subsequently, in step 830, the UE 801 executes RS measurements and performs beam prediction operations for each of the determined future time slots. The measurement procedure comprises reception and analysis of signals corresponding to the RS resource indicators specified in the configuration. The prediction operation comprises application of AI / ML processing techniques to the measurement data to identify optimal beam configurations for each of the future time slots. The prediction algorithm selects, from among the available RS resource indicators, those indicators that satisfy predetermined performance criteria at each predicted time instance.
[0116] In step 840, the UE 801 transmits a beam report to the NW 802. The beam report comprises the results of the prediction operations, specifically identifying the RS resource indicators that have been predicted to provide optimal performance at each of the determined future time slots. The temporal nature of the predictions enables the NW 802 to preemptively configure beam parameters for future communications, thereby enhancing system efficiency and performance.
[0117] The horizontal lines extending across the lower portion of the diagram represent the future time slots for prediction. These dashed lines illustrate the temporal instances for which the UE 801 generates beam predictions based on the configuration parameters received from the NW 802. The predictions occurs after the completion of step 840, indicates that these time slots represent future communication opportunities that follow the initial measurement and reporting phases. The spacing between said dashed lines may correspond to the periodicity parameter specified in the beam report configuration, thereby providing a visual representation of the prediction window's temporal structure. The essential informational required for the UE 801 to perform the prediction includes: (1) the number of future time slots, represented by the parameter QN; and (2) the index associated with each time slot. These informational elements enable the UE 801 to properly make the temporal boundaries of the prediction window and to ensure accurate correspondence between predicted beams and their respective time slots.
[0118] FIG. 9 illustrates a temporal beam prediction framework using periodic RS resource settings according to the embodiments. The figure depicts a timeline with time slots T0 through T7 showing both measurement and prediction phases within a single system frame.
[0119] At time slot T0, the process begins with a reference point for timing calculations, establishing the initial offset Toffset from the start of the frame. This reference point can be the start of the system time frame, or can represent when the UE (e.g., UE 401) receives beam report configuration from the network (e.g., NW 402) , or another timing reference as specified in the configuration parameters.
[0120] At time slot T1, the network provides Set B RS resource 902, which the UE uses for beam measurements. The time slot T1 may be determined by the periodicity parameter TSI and the offset parameter Toffset. The UE performs measurements on Set B RS resource 902 to gather reference signal data according to the configuration parameters.
[0121] At time slot T2, the network provides Set B RS resource 904, occurring precisely TCSI time units after the previous measurement. The UE performs additional beam measurements on this resource, continuing to build its historical measurement dataset that will be used for prediction.
[0122] At time slot T3, the network provides Set B RS resource 906, again occurring at a periodic interval of TCSI after the previous measurement. The UE performs further beam measurements, adding to its temporal profile of beam performance.
[0123] At time slot T4, the network provides Set B RS resource 908, completing the sequence of measurement resources in this example. This time slot also serves as the reference slot nref for prediction calculations, marking the transition point between the measurement phase and the prediction phase.
[0124] Time slots T5, T6, and T7 represent the predicted time slots for which the UE will determine optimal beams. These time slots can occur at the same periodicity TCSI as the measured Set B RS resource. The UE predicts optimal beam index for specific future time slots based on its analysis of the previous measurements. These predicted time slots are determined by the mathematical relationship:
[0125] Where:
[0126] represents the number of slots per frame in the system. The superscript (frame, μ) indicates that this value depends on both the frame structure and the numerology (μ) being used. Different numerologies in 5G NR have different subcarrier spacing and consequently different numbers of slots per frame.
[0127] nf represents the system frame number (SFN) . It's an integer value that identifies the current radio frame in the system.
[0128] is the slot number within the frame. The superscript μ indicates that this value depends on the numerology (μ) being used, and the subscript s, f shows that it's a slot within frame f.
[0129] Toffset is an offset parameter that provides a timing shift from the beginning of the frame. It allows the network to configure when the periodic events (like RS resources) start relative to the frame boundary.
[0130] TCSI is the periodicity parameter that defines the regular interval between consecutive RS resources or predicted time slots. It determines how frequently the events occur.
[0131] mod TCSI=0 is modulo operation ensures that the specified slot satisfies the periodicity requirement. When a number modulo TCSI equals zero, it means that slot falls on the periodic boundary set by TCSI.
[0132] QN time slots (in this case, 3 time slots corresponding to T5, T6, and T7) are selected for prediction according to the beam report configuration. The prediction results will identify optimal RS resource indicators for each of these predicted time slots, enabling the network to prepare appropriate beam configurations in advance.
[0133] As shown by the above mathematical relationship, the time slot number must satisfy both a periodicity condition and a range constraint relative to the reference slot nref (e.g., time slot T4) In other words, the time slot for measurement or prediction should satisfy:
[0134] This formula creates a series of evenly spaced slots throughout the radio frames, with the spacing determined by TCSI and the initial position determined by Toffset. Only time slots that align with this periodic pattern can be selected for measurement or prediction purposes, ensuring consistent timing for beam management operations.
[0135] In addition, the predicted time slot should satisfy:
[0136] In practical terms, this formula selects time slots that are in the future relative to the reference slot, align with the periodic structure defined by TCSI, and fall within the first QN periods after the reference slot.
[0137] These mathematical relationships ensure that both measurement and prediction occur at consistent intervals aligned with the RS resource setting periodicity, maintaining a structured approach to temporal beam prediction across the frame.
[0138] In certain embodiments, the temporal beam prediction framework operates through a structured procedure involving both the network and UE. The process begins when the network configures periodic, semi-persistent, or aperiodic reports for the UE to communicate model inference results. This configuration includes several essential parameters: a reportQuantity value set to predicted RS Resource Indicator, which establishes the type of information to be reported; a periodic or semi-persistent RS resource settings for Set B with defined periodicity (TCSI) and offset (Toffset) values; a Nslot parameter specifying the quantity QN of future time slots for which predictions are to be generated; and the Set A RS resource indicators.
[0139] Upon receiving this report configuration, the UE determines the exact future time slots for which predictions should be generated. The determination relies on the periodicity and offset parameters of the associated resource settings, in conjunction with a reference time point. The reference time point may be established in one of three ways: based on the time slot of the scheduled Set B resources, the time slot when the UE receives the beam report configuration, or the time slot when the UE is required to transmit the report.
[0140] With the prediction timeline established, the UE proceeds to perform measurements on the associated RS resources and executes beam prediction calculations for each identified future time slot. Finally, the UE transmits the comprehensive beam report containing the prediction results back to the network, completing the temporal prediction cycle and enabling the network to prepare optimal beam configurations for future communication.
[0141] FIG. 10 illustrates an alternative embodiment of the temporal beam prediction framework using different timing parameters for prediction slots. The figure depicts a timeline with time slots T0 through T4 showing how prediction slots are determined relative to a reference time point within a system frame.
[0142] At time slot T0, the process begins with the initial slot in the frame, providing a reference for absolute slot numbering. The slot is identified by its frame-level index where nfrepresents the system frame number (SFN) .
[0143] At time slot T1, the system establishes a reference time point nref for prediction calculations. This reference time point could represent various events depending on the configuration, such as the time when the UE receives a beam report configuration, when the RS resources for measurement is transmitted, or when it is scheduled to transmit a report. This slot serves as the anchor point from which future prediction slots are determined.
[0144] A timing offset Toffset2 is applied after the reference slot, creating a gap before the prediction slots begin. This offset can provide flexibility in positioning the prediction window relative to the reference point, allowing the system to accommodate various processing delays or timing requirements.
[0145] Time slots T2, T3, and T4 represent the predicted time slots for which the UE will determine optimal beams. These slots occur at regular intervals defined by the periodicity parameter TCSI2. The UE predicts optimal beam indices specifically for these future time slots based on its analysis of previous measurements. The figure illustrates that QN time slots (in this case, 3 time slots corresponding to T2, T3, and T4) are selected for prediction according to the beam report configuration.
[0146] The slot number of the predicted time slots should satisfy the mathematical relationship:
[0147] The formula establishes which slots align with the periodic pattern and specified time offset.
[0148] The additional constraint ensures that only slots that are between 1 and QN periods ahead of the reference slot are included in the prediction set.
[0149] The embodiment uses dedicated prediction timing parameters TCSI2 and Toffset2 that may differ from the measurement resource timing parameters. The parameters TCSI2 and Toffset2 serve as critical timing controls specifically designed for the prediction phase of the beam management framework. TCSI2 defines the fundamental periodicity between consecutive prediction slots, establishing regular intervals at which the UE will generate future beam predictions. This parameter allows the network to configure prediction intervals independently from measurement intervals, enabling optimization based on mobility patterns, processing capabilities, or channel coherence time. On the other hand, Toffset2 provides a precise timing offset from the reference time point to the first prediction slot, creating a configurable delay that accommodates various system needs such as processing time, reporting latencies, or alignment with other network functions. Together, these parameters form what the specification terms "predictionPeriodicityandOffset, " a dedicated timing mechanism that gives network operators fine-grained control over when predictions occur. This separation from measurement timing parameters (which might use different periodicity values) offers significant flexibility, allowing deployments to maintain frequent measurements while generating predictions at appropriate intervals, or to align prediction slots with specific transmission opportunities where beam information would be most valuable. The mathematical relationship between these parameters ensures that prediction slots follow a consistent, predictable pattern that both the network and UE can synchronize to, regardless of the specific RS resource settings used for measurements. This can provide additional flexibility for scenarios where measurement and prediction cycles need different timing characteristics, such as in cases where measurement resources follow one periodicity while prediction requirements follow another.
[0150] In certain embodiments, the temporal beam prediction framework operates through a coordinated sequence of operations between the network and UE. The process is initiated when the network configures aperiodic reports (i.e., report configuration) for the UE to communicate AI / ML model inference results. The report configuration includes several critical parameters: a reportQuantity field set to predicted RS Resource Indicator to specify the type of information being reported; a RS resource settings for Set B; a Nslot parameter specifying the quantity QN of future time slots for which predictions are to be generated; ; the Set A RS resource indicators that identify potential beam candidates; and a predictionPeriodicityandOffset parameter that establishes the timing structure for prediction slots with periodicity TCSI2 and / or offset Toffset.
[0151] Upon receiving this comprehensive configuration, the UE may proceed to determine the specific future time slots for which predictions will be generated. This determination follows the predictionPeriodicityandOffset parameters in conjunction with a reference time point, which may be established in one of three ways: based on the time slot of scheduled RS resources for measurement (i.e., Set B) , the time slot when the UE receives the beam report configuration, or the time slot when the UE is required to transmit the report.
[0152] With the prediction timeline established, the UE conducts measurements on the configured Set B RS resources and employs its AI / ML model to generate beam predictions for each identified future time slot. Finally, the UE compiles these predictions into a beam report and transmits it to the network, providing advance notice of optimal beam configurations for upcoming communication opportunities and enabling more efficient resource allocation and performance optimization.
[0153] The embodiments illustrated in FIG. 9 and FIG. 10 represent two distinct approaches to temporal beam prediction, each optimized for different network configurations and operational scenarios. The embodiment in FIG. 9 depicts a framework where the prediction slots are intrinsically linked to the measurement resource timing parameters, creating a unified timing structure across both measurement and prediction phases. In this configuration, the Set B RS resources 902–908 appear at regular intervals determined by periodicity TCSI and offset Toffset, and the prediction slots follow this same periodic structure. This approach is particularly suitable for scenarios where measurement and prediction activities should maintain synchronization with the same underlying resource grid, such as in cases involving periodic or semi-persistent RS resource settings. The mathematical condition governing the slot selection relies on a direct relationship with the measurement resource timing parameters, expressed as:
[0154] followed by a range constraint:
[0155] In contrast, the embodiment in FIG. 10 introduces a more flexible approach where prediction timing can be decoupled from measurement timing through dedicated prediction timing parameters TCSI2 and Toffset2. This configuration allows the reference point nref to establish an anchor for prediction calculations that is independent of the measurement resource pattern. The formula explicitly incorporates the reference time point into the calculation, with TCSI2 potentially differing from the measurement periodicity. This embodiment offers greater flexibility for scenarios with aperiodic resource settings or where prediction requirements operate on a different timescale than measurements, enabling the system to optimize both measurement efficiency and prediction horizon independently.
[0156] FIG. 11 illustrates an alternative embodiment of the temporal beam prediction framework where the offset parameter equals the periodicity parameter (Toffset2 = TCSI2) . This unique configuration creates a simplified timing structure for prediction slots within the system frame.
[0157] At time slot T0, the process begins with the initial slot in the frame, providing a reference for absolute slot numbering. The slot is identified by its frame-level index where nfrepresents the system frame number (SFN) .
[0158] At time slot T1, the system establishes the reference time point nref for prediction calculations. This reference slot serves as the anchor from which future prediction slots are determined. The special condition, in which Toffset2 is equal to (or approximately equal to) TCSI2, may lead to a unique alignment, causing the first prediction slot to occur exactly one periodicity interval after the reference slot.
[0159] Time slots T2, T3, and T4 represent the prediction time slots for which the UE will determine optimal beams. These slots occur at regular intervals of exactly TCSI2 after the reference slot. Because Toffset2 is equal to (or approximately equal to) TCSI2, prediction slots begin exactly one period after the reference slot and continue for QN consecutive periods (in this case, three periods corresponding to slots T2, T3, and T4) .
[0160] In this case, the slot number of the predicted time slots should satisfy:
[0161] This simplified formula eliminates the need for the separate Toffset2 parameter, as its function is effectively incorporated through the direct relationship with the reference slot nref. The first condition ensures that prediction slots align with periodicity boundaries defined by TCSI2, while the second condition constrains the prediction window to include only slots that fall between one and QN periods after the reference time slot.
[0162] This special configuration offers implementation advantages, as it simplifies the timing calculations and creates a more intuitive relationship between the reference slot and prediction slots. With Toffset2 = TCSI, the system creates a straightforward sequence where prediction slots start immediately after one full period from the reference point and continue at regular intervals. This configuration can reduce computational complexity and provide more predictable timing relationships, which may be particularly valuable in scenarios with limited processing capabilities or strict timing requirements.
[0163] FIG. 12 illustrates a flow diagram presenting a method 1200 of wireless communication according to the embodiments. The method includes the following steps performed by a UE:
[0164] S1202: Receive, from a network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of RS resource indicators ;
[0165] S1204: Performing measurements on a set of RS resources according to the configuration parameter set;
[0166] S1206: Generate, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators; and
[0167] S1208: Transmit, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.
[0168] In step S1202, the UE receives a beam report configuration from the network. This report configuration contains critical parameters that govern the entire temporal prediction process. The configuration parameter set includes timing specifications such as periodicity values (TCSI or TCSI2) , offset parameters (Toffset or Toffset2) , and the number of prediction slots (QN) . It also defines the one or more sets of RS resource indicators that will be used for both measurement and prediction purposes. This set may include both Set B RS resources (used for measurements) and Set A RS resources (candidates for prediction) . The configuration establishes the rules by which the UE will determine when to perform measurements and for which future time slots it should generate predictions.
[0169] In step S1204, the UE performs beam measurements according to the configuration parameter set received from the base station. For periodic or semi-persistent RS resource settings, the UE identifies measurement opportunities based on the periodicity and offset parameters. The UE conducts measurements on the specified RS resources, collecting signal quality metrics such as RSRP. These measurements can be performed systematically to build the temporal dataset needed for prediction. The specific timing of these measurements may be influenced by when the UE received the beam report configuration or when it is scheduled to transmit the report, as specified in the configuration parameter set. The measurement process focuses particularly on the first set of RS resource indicators designated for sensing purposes.
[0170] In step S1206, based on the beam measurements, the UE can generate a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with these predicted indicators. This prediction process may employ an AI / ML based model for temporal prediction. The UE determines which RS resource indicators will meet a predetermined performance metric for each time slot in the set of predicted time slots, up to the quantity indicated by the QN. The time slot indices for prediction are determined according to the configuration, which may use either the periodicity and offset of linked RS resource settings or dedicated prediction periodicity and offset parameters. The UE can, for example, select RS resource indicators that maximize RSRP, SINR or other performance indicators for each time slot in the predicted set, ensuring optimal beam selection for future communication.
[0171] In step S1208, the UE transmits to the network a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots. This report provides the network with advance knowledge of which beams are expected to perform best at specific future time slots. The report format follows the reportQuantity specification in the configuration, focusing on the predicted RS resource indicators that meet the predetermined performance metrics. This process enables the network (more specifically, the base station) to prepare appropriate beam configurations in advance, improving system efficiency and performance. The timing of this report transmission may itself serve as a reference point for determining the indices of predicted time slots in certain configurations, creating a dynamic relationship between reporting and prediction timing.
[0172] In essence, this invention represents a significant advancement in wireless communication technology by introducing a structured framework for AI / ML based temporal beam prediction. By establishing a standardized configuration and reporting mechanism for future beam states, the system overcomes a critical gap in current 3GPP specifications while significantly reducing overhead compared to conventional exhaustive beam sweeping. The flexible timing parameters enable precise control over prediction windows, allowing networks to optimize beam selection based on specific deployment scenarios and mobility patterns. The two-set approach, which utilizes a smaller set of sensing beams to predict the optimal communication beams, achieves significant efficiency improvements while preserving performance. More significantly, this invention creates a standardized foundation for integrating AI / ML capabilities into beam management, enabling cellular networks to transition from reactive to proactive beam selection. This forward-looking approach enhances reliability in challenging environments, improves handover performance, reduces latency, and ultimately delivers more consistent user experiences, particularly in high-frequency deployments where beam management is essential for maintaining robust connections.
[0173] For clarity in this specification, certain terminological conventions are observed. The singular forms "a" , "an" , and "the" are intended to encompass plural forms as well, unless the context clearly indicates otherwise. The term "and / or" refers to and encompasses any and all possible combinations of one or more of the associated listed items. The terms "includes, " "including, " "comprises, " and "comprising" specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0174] The term "exemplary" is used herein to mean "serving as an example, instance, or illustration" and should not be construed as necessarily preferred or advantageous over other aspects or designs.
[0175] The use of ordinal designators like "first, " "second, " and so forth in the specification and claims serves to differentiate between multiple instances of similarly named elements. These designators do not imply any inherent sequence, priority, or chronological order in the manufacturing process or functional relationship between elements. Rather, they are employed solely as a means of uniquely identifying and distinguishing between separate instances of elements that share a common name or description.
[0176] Unless specifically stated otherwise, the term "some" refers to one or more. Various combinations using "at least one of" or "one or more of" followed by a list (e.g., A, B, or C) should be interpreted to include any combination of the listed items, including individual items and multiple items.
[0177] Terms such as "coupled, " "connected, " "connecting, " and "electrically connected" are used synonymously to describe a state of being electrically or electronically linked. When an entity is described as being in "communication" with another entity or entities, it implies the capability of sending and / or receiving electrical signals, which may contain image / voice or data / control information, regardless of whether these signals are analog or digital in nature.
[0178] As may be used throughout this specification and the appended claims, terms of approximation and degree such as "substantially, " "approximately, " "generally, " "essentially, " "nearly, " "about, " and similar expressions are used to account for variations in precision, manufacturing tolerances, measurement accuracy, environmental conditions, and inherent material properties that may affect the described features or characteristics. Such variations may range from ±20%in broader applications to progressively tighter tolerances of ±10%, ±5%, ±3%, ±2%, ±1%, or ±0.5%in more precise implementations. The specific degree of variation encompassed by these terms of approximation in any given context is informed by the nature of the component, relationship, or parameter being described, the technical requirements of the particular embodiment, and the understanding of one skilled in the relevant art.
[0179] In the context of this patent specification, the term "user equipment" (UE) encompasses a broad range of devices possessing radio communication capabilities. This definition includes, but is not limited to, smartphones (specifically, handheld touchscreen mobile computing devices capable of connecting to one or more cellular networks) , Personal Data Assistants (PDAs) , pagers, laptop computers, desktop computers, wireless handsets, and any computing device equipped with a wireless communications interface. User equipment may also be referred to by various alternative terms, including but not limited to: client, mobile device, mobile terminal, user terminal, mobile unit, mobile station, mobile user, subscriber, user, remote station, access agent, user agent, receiver, radio equipment, reconfigurable radio equipment, or reconfigurable mobile device. These terms should be considered interchangeable within the context of this document.
[0180] The scope of UEs also extends to Internet of Things (IoT) devices. IoT UEs are characterized by a network access layer specifically designed for low-power IoT applications that typically involve short-lived UE connections. These IoT UEs may employ various technologies for data exchange, including Machine-to-Machine (M2M) , Machine Type Communication (MTC) , or massive MTC (mMTC) . Such data exchanges may occur with an MTC server or device via a Public Land Mobile Network (PLMN) , with other UEs using Proximity Services (ProSe) or Device-to-Device (D2D) communications, or through sensor networks or IoT networks. It is noteworthy that M2M or MTC data exchanges are often initiated by the machine itself rather than by human intervention.
[0181] An IoT network, as referenced in the above section, describes an interconnected system of IoT UEs. These UEs may include uniquely identifiable embedded computing devices integrated within the broader Internet infrastructure. IoT UEs may execute background applications, such as keep-alive messages or status updates, to maintain and facilitate the connections within the IoT network.
[0182] UEs are configured to establish communicative coupling with Radio Access Networks (RANs) through a radio interface. This radio interface is a physical communication interface or layer designed to operate with various cellular communication protocols. These protocols may include, but are not limited to, GSM protocol, CDMA network protocol, Push-to-Talk (PTT) protocol, PTT over Cellular (POC) protocol, UMTS protocol, 3GPP LTE protocol, 5G protocol, and New Radio (NR) protocol.
[0183] As a specific example, a UE and a RAN may utilize a Uu interface (such as an LTE-Uu interface) to exchange control plane data. This exchange occurs via a protocol stack comprising multiple layers: a Physical (PHY) layer, a Medium Access Control (MAC) layer, a Radio Link Control (RLC) layer, a Packet Data Convergence Protocol (PDCP) layer, and a Radio Resource Control (RRC) layer. In this context, a Downlink (DL) transmission refers to data sent from the RAN to the UE, while an Uplink (UL) transmission refers to data sent from the UE to the RAN.
[0184] Furthermore, UEs may employ a sidelink for direct communication with other UEs, facilitating D2D, Peer-to-Peer (P2P) , and / or ProSe communication. A ProSe interface, for instance, may incorporate one or more logical channels. These channels include, but are not limited to, a Physical Sidelink Control Channel (PSCCH) , a Physical Sidelink Shared Channel (PSSCH) , a Physical Sidelink Discovery Channel (PSDCH) , and a Physical Sidelink Broadcast Channel (PSBCH) .
[0185] The various aspects described herein may be implemented using a variety of hardware and software components. These may include processors, Digital Signal Processors (DSPs) , Application Specific Integrated Circuits (ASICs) , Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic, or discrete hardware components. A processor in this context may be a microprocessor, but could also be any conventional processor, controller, microcontroller, or state machine. Processors may also be implemented as combinations of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
[0186] The aspects described in this specification can be implemented through both hardware and software instructions. These instructions may be stored on various types of computer-readable media, including but not limited to Random Access Memory (RAM) , flash memory, Read Only Memory (ROM) , Electrically Programmable ROM (EPROM) , Electrically Erasable Programmable ROM (EEPROM) , registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the art. In a typical configuration, the storage medium is connected to a processor, enabling the processor to read information from and write information to the medium. In some configurations, the storage medium may be integral to the processor itself.
[0187] In some embodiments, the computing instructions may be executed by an operating system, which may include but is not limited to Microsoft Windows, Apple Mac OS X, macOS or iOS, various distributions of the Linux operating system, or Google Android operating system.
[0188] Some embodiments may involve computers on a distributed computing network, such as a network with multiple clients and / or servers. In such embodiments, clients may run software implementing client-side portions of the described systems and methods, while servers handle requests from these clients. Communication between clients and servers may occur via one or more electronic networks, which may include the Internet, wide area networks, mobile telephone networks, wireless networks (e.g., Wi-Fi, 5G) , or local area networks, implemented using any known network protocols.
[0189] In implementations where the systems described in this specification collect user information, provisions may be made to protect user privacy and data. Specifically, users may be afforded the opportunity to opt in or out of programs or features that collect personal information, such as data related to user preferences or smart device usage patterns. Furthermore, in certain embodiments, data protection measures may be implemented to anonymize collected information prior to storage or utilization. For instance, a user's identity may be anonymized to prevent the determination or association of personally identifiable information with that specific user. Additionally, user preferences and interaction data may be generalized, potentially based on broader demographic categories, rather than being linked to individual users.
[0190] It should be noted that the operational steps described in any exemplary aspects within this specification are provided as examples and for discussion purposes. These operations may be performed in numerous different sequences other than the illustrated sequences. Furthermore, operations described in a single step may actually be performed as multiple distinct steps, and multiple steps may be combined into a single operational step. The steps in the appended figures may be subject to numerous modifications as will be apparent to those skilled in the art.
[0191] Some embodiments may incorporate all explicitly disclosed features as well as additional features that, while not specifically described herein, are compatible with and enhance the core invention. Conversely, other embodiments may selectively omit certain non-disclosed elements, either partially or in their entirety, while still falling within the scope of the invention. This flexibility in feature inclusion or exclusion allows for a range of implementations tailored to specific applications or requirements, without departing from the fundamental principles of the invention.
[0192] The logical stages illustrated in the drawings may be reordered, combined, or broken out if they are not order-dependent. The ordering and groupings presented in this specification are not exhaustive, and other arrangements will be apparent to those skilled in the art. These stages may be implemented in hardware, firmware, software, or any combination thereof.
[0193] The drawings and descriptions provided in this specification offer detailed illustrations of various embodiments of the invention. However, it should be understood by those skilled in the art that these embodiments can be implemented without necessarily adhering to every specific detail provided herein. In some instances, well-established methods, procedures, components, and circuits have been mentioned without elaborate explanations to avoid obscuring the key aspects of the embodiments. It is important to note that the figures presented in this specification, including any component diagrams, are intended for illustrative purposes and may not be drawn to scale. This allows for a clear presentation of the inventive concepts while leaving room for variations and adaptations within the scope of the invention.
[0194] While the invention has been described in connection with certain embodiments, it will be understood by those skilled in the art that various modifications and adaptations can be made without departing from the scope of the invention. The specific embodiments presented are intended to illustrate the invention and not to limit its application or construction. Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.
Claims
1.A method of wireless communication, performed by a user equipment (UE) , comprising:receiving, from a network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of reference signal (RS) resource indicators;performing measurements on a set of RS resources according to the configuration parameter set;generating, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators; andtransmitting, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.2.The method of claim 1, wherein generating the set of predicted RS resource indicators and the set of predicted time slots is based on an artificial intelligence (AI) or machine learning (ML) model for temporal prediction.3.The method of claim 1, wherein the configuration parameter set comprises a reportQuantity set to predicted RS resource indicators.4.The method of claim 1, wherein the configuration parameter set comprises a parameter indicating a quantity of the predicted time slots.5.The method of claim 4, wherein generating the set of predicted RS resource indicators and the set of predicted time slots comprises generating the set of predicted RS resource indicators that meet a predetermined performance metric for each time slot in the set of predicted time slots up to the indicated quantity of the predicted time slots.6.The method of claim 1, wherein the beam report configuration is associated with a periodic RS resource setting or semi-persistent RS resource setting having a periodicity parameter and an offset parameter.7.The method of claim 6, wherein time slot index of the set of predicted time slots are determined based on the periodicity parameter and the offset parameter of the associated periodic RS resource setting or semi-persistent resource setting.8.The method of claim 7, wherein the time slot index are further determined based on a time when the UE receives the beam report configuration.9.The method of claim 7, wherein the time slot index are further determined based on a time when the UE transmits the beam report.10.The method of claim 1, wherein the beam report configuration is associated with a periodic RS resource setting, semi-persistent RS resource setting, or aperiodic resource setting.11.The method of claim 10, wherein the configuration parameter set further comprises a prediction periodicity parameter and / or a prediction offset parameter for the set of predicted time slots.12.The method of claim 11, wherein time slot index of the set of predicted time slots are determined based on the prediction periodicity parameter.13.The method of claim 11, wherein the time slot index are further determined based on the prediction offset parameter.14.The method of claim 11, wherein the time slot index are further determined based on a time when the UE receives the beam report configuration.15.The method of claim 11, wherein the time slot index are further determined based on a time when the UE transmits the beam report.16.The method of claim 1, wherein generating the set of predicted RS resource indicators and the set of predicted time slots comprises selecting RS resource indicators that maximize reference signal received power (RSRP) , Signal to Interference plus Noise Ratio (SINR) and / or UE throughput for each time slot in the set of predicted time slots.17.The method of claim 1, wherein the beam report configuration further comprises a parameter indicating RS resource indicators of a first set used for the measurements on the set of RS resources and RS resource indicators of a second set used for prediction.18.The method of claim 17, wherein a quantity of RS resource indicators in the first set is less than or equal to a quantity of RS resource indicators in the second set.19.A user equipment (UE) for wireless communication, comprising:an antenna; anda processor coupled to the antenna, configured to:receive, via the antenna, from the network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of reference signal (RS) resource indicators;perform measurements on a set of RS resources according to the configuration parameter set;generate, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators; andtransmit, via the antenna, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.20.A non-transitory computer-readable medium storing instructions that, when executed by a user equipment (UE) , cause the UE to perform operations comprising:receiving, from a network, a beam report configuration comprising a configuration parameter set for temporal beam prediction and one or more sets of reference signal (RS) resource indicators;performing measurements on a set of RS resources according to the configuration parameter set;generating, based on the measurements, a set of predicted RS resource indicators selected from the one or more sets of RS resource indicators, and a set of predicted time slots associated with the predicted RS resource indicators; andtransmitting, to the network, a beam report comprising the set of predicted RS resource indicators and the associated set of predicted time slots.
Citation Information
Patent Citations
Method and appratus for beam group reporting in mobile communications
US20230156486A1
Methods, devices, and medium for communication
WO2023245581A1
Support of UE centric ai based temporal beam prediction
WO2024036605A1
Methods and apparatuses for reporting CSI prediction for a set of beams
WO2024065372A1