Beam alignment for wireless networks based on pre-trained machine learning model and angle of arrival

Abstract

A method includes determining signal measurements for received signals for a plurality of first beams; determining a beam index and an estimated angle of arrival for transmitting a second beam for the one or more network nodes based on the signal measurements for the set of one or more first beams and the positioning information for the user equipment; determining that the associated user equipment receives a second beam for one or more of the network nodes transmitting the second beam, wherein the associated user equipment receiving the second beam is at least one of: a beam from a codebook having an angle of arrival most closely matching the estimated angle of arrival of the second beam transmitted by the network node, or a beam having a beam weight calculated based on the estimated angle of arrival of the second beam transmitted by the network node; and controlling to transmit a beam index for transmitting the second beam for the one or more network nodes.

Description

Technical Field

[0001] This specification relates to wireless communications. Background Art

[0002] A communication system may be a facility that enables communication between two or more nodes or devices, such as fixed or mobile communication devices. Signals may be carried on wired or wireless carriers.

[0003] An example of a cellular communication system is an architecture standardized by the Third Generation Partnership Project (3GPP). Recent developments in the field are generally referred to as the Long Term Evolution (LTE) of the Universal Mobile Telecommunications System (UMTS) radio access technology. Evolved UMTS Terrestrial Radio Access (E-UTRA) is the air interface of the 3GPP's Long Term Evolution (LTE) upgrade path for mobile networks. In LTE, a base station or access point (AP), referred to as an enhanced node AP (eNB), provides wireless access within a coverage area or cell. In LTE, a mobile device or mobile station is referred to as a user equipment (UE). LTE has included multiple improvements or developments. Various aspects of LTE continue to improve.

[0004] 5G New Radio (NR) is developed as part of the ongoing mobile broadband evolution process to meet the requirements of 5G, similar to the early evolution of 3G and 4G wireless networks. In addition, 5G targets emerging use cases beyond mobile broadband. The goal of 5G is to provide significant improvements in wireless performance, which may include new levels of data rate, latency, reliability, and security. 5G NR can also be scaled to efficiently connect large-scale Internet of Things (IoT) and can provide new types of mission-critical services. For example, ultra-reliable low-latency communication (URLLC) devices may require high reliability and very low latency. Summary of the invention

[0005] According to an example embodiment, a method may include: determining, by a user equipment, a signal measurement for a signal for a plurality of first beams received from a network node; determining, by the user equipment, a beam index and an estimated angle of arrival for one or more network nodes to send a second beam based on the signal measurement for a set of one or more first beams and positioning information for the user equipment; determining, by the user equipment, for one or more of the second beams sent by the network node, that an associated user equipment receives the second beam, wherein the second beam received by the associated user equipment is at least one beam from a codebook having an angle of arrival that most closely matches the estimated angle of arrival for the network node to send the second beam, or a beam having a beam weight calculated based on the estimated angle of arrival for the network node to send the second beam; and controlling, by the user equipment, to send to the network node a beam index for one or more network nodes to send the second beam.

[0006] According to an example embodiment, a method may include: controlling, by a network node, sending a signal to a user equipment via multiple first beams; controlling, by the network node, receiving signal measurements for a set of one or more first beams from the user equipment; controlling, by the network node, receiving positioning information for the user equipment; determining, by the network node, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurements for the set of one or more first beams and the positioning information for the user equipment; and controlling, by the network node, sending to the user equipment information indicating the estimated angle of arrival for transmitting the second beam by one or more network nodes.

[0007] Other example embodiments are provided or described for each of the example methods, including: a component for performing any of the example methods; a non-transitory computer-readable storage medium including instructions stored thereon, which, when executed by at least one processor, are configured to cause a computing system to perform any of the example methods; an apparatus including at least one processor, and at least one memory including computer program code, the at least one memory and the computer program code being configured to, together with the at least one processor, cause the apparatus to at least perform any of the example methods.

[0008] The details of one or more example embodiments are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is a block diagram of a wireless network according to an example embodiment.

[0010] Figure 2 is a flow chart illustrating the operation of a user equipment or UE.

[0011] Figure 3 is a flow chart illustrating the operation of a gNB or network node.

[0012] Figure 4 is a diagram illustrating the training of a machine learning (ML) model by a network node or gNB or other network entity.

[0013] Figure 5 is a block diagram illustrating the operation of a user equipment (or UE).

[0014] Figure 6 is a diagram illustrating the operation of a gNB or a network node.

[0015] Figure 7 is a diagram illustrating the training of a machine learning (ML) model.

[0016] Figure 8 is a diagram showing gNB beams and UE beams.

[0017] Fig. 9 is a diagram illustrating the ML model inference stage at a UE or user equipment.

[0018] Fig.10 2 is a diagram showing representation of the gNB and UE codebooks for the second beam scanning phase.

[0019] Fig.11 is a block diagram of a wireless station or node (eg, a network node, a user node or UE, a relay node, or other node). DETAILED DESCRIPTION

[0020] Figure 1 is a block diagram of a wireless network 130 according to an example embodiment. Figure 1 In a wireless network 130, user equipment 131, user equipment 132, user equipment 133, and user equipment 135 (which may also be referred to as mobile stations (MS) or user equipment (UE)) may be connected (and communicate) with a base station (BS) 134 (which may also be referred to as an access point (AP), an enhanced Node B (eNB), a gNB, or a network node). The terms user equipment and user equipment (UE) may be used interchangeably. The BS may also include or be referred to as a radio access network (RAN) node, and may include a portion of the BS or a portion of the RAN node, such as a centralized unit (CU) and / or a distributed unit (DU) in the case of a separate BS or a separate gNB. At least a portion of the functionality of a BS (e.g., an access point (AP), a base station (BS), or an (e) Node B (eNB), a gNB, a RAN node) may also be implemented by any node, server, or host operably coupled to a transceiver (such as a remote radio head). BS (or AP) 134 provides wireless coverage within cell 136, including providing wireless coverage to user equipment (or UE) 131, user equipment 132, user equipment 133, and user equipment 135. Although only four user equipments (or UEs) are shown as connected or attached to BS 134, any number of user equipments may be provided. BS 134 is also connected to core network 150 via S1 interface 151. This is just a simple example of a wireless network, and other examples may be used.

[0021] A base station (e.g., such as BS 134) is an example of a radio access network (RAN) node within a wireless network. A BS (or RAN node) may be or may include (or may alternatively be referred to as) an access point (AP), a gNB, an eNB, or a portion thereof (e.g., a centralized unit (CU) and / or a distributed unit (DU) in the case of a split BS or a split gNB), or other network nodes.

[0022] According to an illustrative example, a BS node (e.g., BS, eNB, gNB, CU / DU, ...) or a radio access network (RAN) may be part of a mobile telecommunication system. The radio access network (RAN) may include one or more BSs or RAN nodes that implement a radio access technology (e.g., to allow one or more UEs to have access to a network or core network). Thus, for example, a RAN (RAN node, such as a BS or gNB) may reside between one or more user equipment or UEs and a core network. According to an example embodiment, each RAN node (e.g., BS, eNB, gNB, CU / DU, ...) or BS may provide one or more wireless communication services for one or more UEs or user equipment, for example, to allow the UE to have wireless access to a network via a RAN node. Each RAN node or BS may perform or provide wireless communication services, such as allowing a UE or user equipment to establish a wireless connection to a RAN node, and to send data to one or more UEs and / or receive data from one or more UEs. For example, after establishing a connection to a UE, a RAN node or a network node (e.g., BS, eNB, gNB, CU / DU, ...) may forward data received from a network or core network to a UE, and / or forward data received from a UE to a network or core network. A RAN node or network node (e.g., BS, eNB, gNB, CU / DU, ...) may perform various other wireless functions or services, such as, for example, broadcasting control information (e.g., system information or on-demand system information) to UEs, paging UEs when there is data to be delivered to the UEs, assisting in handing over UEs between cells, scheduling resources for uplink data transmission from (multiple) UEs and downlink data transmission to (multiple) UEs, sending control information to configure one or more UEs, etc. These are a few examples of one or more functions that a RAN node or BS may perform.

[0023] A user device or user node (user terminal, user equipment (UE), mobile terminal, handheld wireless device, etc.) may refer to a portable computing device including a wireless mobile communication device operating with or without a user identification module (SIM), including but not limited to the following types of devices: mobile station (MS), mobile phone, cellular phone, smart phone, personal digital assistant (PDA), handheld device, device using a wireless modem (alarm or measurement device), for example, laptop and / or touch screen computer, tablet computer, tablet phone, game console, notebook computer, transportation, sensor and multimedia device, as an example, or any other wireless device. It should be understood that a user device may also be (or may include) a device that is almost exclusively uplink, an example of which is a camera or video camera that loads images or video clips to a network. Moreover, a user node may include a user equipment (UE), a terminal device, a user terminal, a mobile terminal, a mobile station, a mobile node, a subscriber device, a subscriber node, a subscriber terminal, or other user nodes. For example, a user node may be used for wireless communications with one or more network nodes (e.g., gNB, eNB, BS, AP, CU, DU, CU / DU) and / or with one or more other user nodes, regardless of the technology or radio access technology (RAT). In LTE (as an illustrative example), the core network 150 may be referred to as an evolved packet core (EPC), which may include: a mobility management entity (MME) that may handle or assist in the mobility / handover of user equipment between BSs; one or more gateways that may forward data and control signals between BSs and a packet data network or the Internet; and other control functions or blocks. Other types of wireless networks, such as 5G (which may be referred to as new radio (NR)), may also include a core network.

[0024] In addition, the techniques described herein may be applied to various types of user equipment or data service types, or may be applied to user equipment that may have multiple applications running thereon, where the multiple applications may belong to different data service types. New radio (5G) development may support multiple different applications or multiple different data service types, such as: machine type communication (MTC), enhanced machine type communication (eMTC), Internet of Things (IoT), and / or narrowband IoT user equipment, enhanced mobile broadband (eMBB), and ultra-reliable low latency communication (URLLC). Many of these new 5G (NR) related applications may require higher performance than previous wireless networks.

[0025] IoT may refer to the growing group of objects that may have internet or network connectivity, such that these objects may send information to and receive information from other network devices. For example, many sensor type applications or devices may monitor physical conditions or situations, and may, for example, send reports to a server or other network device when an event occurs. Machine type communications (MTC, or machine-to-machine communications) may, for example, be characterized by fully automatic data generation, exchange, processing, and actuation between intelligent machines, with or without human intervention. Enhanced mobile broadband (eMBB) may support much higher data rates than currently available in LTE.

[0026] Ultra-Reliable Low Latency Communication (URLLC) is a new data service type or new use case that can be supported for new radio (5G) systems. By way of example, this can enable emerging new applications and services such as industrial automation, autonomous driving, vehicle safety, e-health services, etc. 3GPP aims to provide a 5G network with a 5G-capable architecture corresponding to 10G. -5 The invention provides connectivity with reliability of block error rate (BLER) of 1000ms and user / data plane (U-plane) latency of up to 1 ms. Thus, for example, a URLLC user equipment / UE may require significantly lower block error rate and low latency than other types of user equipment / UE (regardless of whether there is a requirement for simultaneous high reliability). Thus, for example, a URLLC UE (or a URLLC application on the UE) may require much shorter latency than an eMBB UE (or an eMBB application running on the UE).

[0027] The techniques described herein may be applied to various wireless technologies or wireless networks, such as 5G (New Radio (NR)), centimeter wave, and / or millimeter wave band networks, IoT, MTC, eMTC, eMBB, URLLC, 6G, etc., or any other wireless network or wireless technology. These example networks, technologies, or data service types are provided only as illustrative examples.

[0028] Beamforming can be used to transmit and / or receive signals. By adjusting the weights (e.g., amplitude and / or phase) of each antenna element of an antenna panel or antenna system, a node (e.g., a network node and / or a UE) can provide directivity, where the transmission power can be directed in a specific direction for transmitting signals via beamforming. Thus, beamforming can allow a node to provide transmitter-side directivity, where a transmitting node (e.g., a gNB or network node, or a UE or user equipment) can apply a weight or a set of weights to an antenna element to form a beam for transmitting a signal. Similarly, beamforming can also be used to provide receiver-side directivity, where a receiving node can apply a weight or a set of weights to an antenna element to form a receive beam. Because each beam can typically cover only a limited area or direction, multiple beams (e.g., where each beam points to a different direction) are typically required to cover the full range of directions. Example beams can include synchronization signal block reference signal (SSB) beams and channel state information reference signal (CSI-RS) beams, where each reference signal is associated with a different beam because it can point to a different direction. The gNB or network node may utilize multiple beams to cover the entire service area, and one or a subset of these beams may be associated with the UE (e.g., may be pointed in a direction toward the UE and / or may be the strongest beam (or best beam, such as a beam with the highest RSRP measurement) for communicating with the UE).

[0029] A wide beam may be wider (have a wider or larger angle to cover a larger range of angular directions) than a narrow beam. A wide beam may include, for example, an SSB beam or a CSI-RS beam, while a narrow beam may typically include a CSI-RS beam.

[0030] 3GPP beam management may include processes P1, P2 and P3, which are briefly summarized as follows:

[0031] P1 (gNB wide beam scanning): The gNB scans through a set of wide angle beams assigned to different SSB / CSI-RS resources while transmitting signals (e.g., SSB or CSI-RS signals) on the resources associated with each beam. Scanning may refer to the node generating a sequence of beams across a range of directions or possibly covering all directions. The UE performs signal measurements on different SSBs (or for different SSB beams) (e.g., the UE measures the reference signal received power (RSRP)), and after the measurements, the UE requests access to the gNB by sending (e.g., by sending a random access preamble) in the time-frequency position corresponding to the SSB of the best beam.

[0032] P2 (gNB narrow beam scanning): The gNB performs beam scanning with a set of narrow beams assigned to different CSI-RS resources that cover the wide angular space of the SSB beams employed in P1 (or indicated by the UE as the best wide beam). The UE performs signal measurements (e.g., measuring the RSRP of the signals sent by the gNB via the narrow beams) and reports the measurements of one or more best or strongest gNBs (or network nodes) transmitting narrow beams to the gNB using CSI reports. The gNB can select the best gNB to transmit narrow beam based on the RSRP measurements.

[0033] P3 (UE beam scanning): The gNB uses the best or optimal gNB transmit narrow beam selected from P2 to send multiple CSI-RS (multiple narrow beams or CSI-RS beams), and the UE scans through a set of UE receive narrow beams to refine the UE beam direction. The UE can select the best UE receive narrow beam based on RSRP measurement and communicate the result to the gNB.

[0034] Therefore, during data transmission, the gNB (or network node) uses the best or strongest narrow beam found in P2 (e.g., the beam or reference signal with the highest RSRP), and the UE uses the best or strongest beam found in P3 (e.g., the beam or reference signal with the highest RSRP).

[0035] These three processes (P1, P2, and P3) are inefficient for two main reasons. On the one hand, scanning all beams in each of the processes P1, P2, and P3 is time consuming. This significantly increases the latency of the system during the initial access operation. Second, the above process significantly increases measurement reports, which increases signaling overhead. This reduces system throughput and spectral efficiency. For example, data transmission opportunities may be reduced due to the resources used for signaling and obtaining beam measurements.

[0036] Figure 2 and / or Figure 3 The techniques of the method can be used, for example, to replace process P2 and / or P3 with a new beam management process that uses machine learning (ML) driven outputs to identify or at least narrow the number of signal (e.g., CSI-RS) measurements used to find the best receive (Rx) and transmit (Tx) beams. These techniques can, for example, speed up the process of beam alignment, and can reduce measurement and signaling overhead and / or enable more accurate selection of transmit and / or receive beams, and can also be applied to UEs with different beam grid configurations.

[0037] Figure 22 is a flow chart illustrating the operation of a user equipment or UE. Operation 210 includes determining, by a user equipment (e.g., UE), a signal measurement (e.g., measuring reference signal received power (RSRP)) for signals (e.g., for reference signals such as synchronization signal blocks (SSBs) or channel state information reference signals (CSI-RS) signals) received from a network node for multiple first beams (e.g., for multiple wide beams or SSB beams). Operation 220 includes determining, by the user equipment, a beam index and an estimated angle of arrival (AoA) for transmitting a second beam for one or more network nodes (e.g., transmitting a narrow beam for one or more network nodes or gNBs or transmitting a CSI-RS beam for a gNB) based on the signal measurement for a set of one or more first beams and positioning information for the user equipment. Operation 230 includes: determining, by the user equipment, that the associated user equipment receives the second beam (e.g., the associated UE receives a narrow beam or the associated UE receives a CSI-RS beam) for one or more of the network nodes transmitting the second beam, wherein the associated user equipment receives the second beam is at least one of the following: a beam from a codebook having an arrival angle that most closely matches the estimated arrival angle of the network node transmitting the second beam, or a beam having a beam weight calculated based on the estimated arrival angle for the network node transmitting the second beam. And, operation 240 includes controlling, by the user equipment, to send a beam index for one or more network nodes transmitting the second beam to the network node. Therefore, for example, the associated UE receiving the narrow beam or the associated user equipment receiving the second beam can be determined in two ways, for example: 1) a beam of the codebook having an arrival angle that most closely matches the estimated AoA of the network node transmitting the second beam; and / or 2) by calculating the beam weight based on the estimated arrival angle for the network node transmitting the second beam.

[0038] about Figure 2 A method, wherein determining a beam index and an estimated angle of arrival may include: providing, by a user device, signal measurements for a set of one or more first beams and positioning information for the user device as input to a pre-trained machine learning model enabled (e.g., installed and running or executed) on the user device (or UE) (e.g., which may be a general machine learning model trained by a network node (or gNB) or another entity); and controlling, by the user device, receiving, as output from the pre-trained machine learning model, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes (e.g., a narrow beam for transmitting one or more gNBs or a CSI-RS beam for transmitting a gNB) as a label indicating or identifying the beam index and the estimated AoA for transmitting a second beam for one or more network nodes.

[0039] about Figure 2 The method may also include: controlling, by a user equipment, to receive a pre-trained machine learning model from a network node; selecting, by the user equipment, a set of one or more first beams based on signal measurements performed by the user equipment for multiple first beams (e.g., selecting one or more first beams with the highest RSRP (e.g., SSB beams or wide beams)); and determining, by the user equipment, positioning information for the user equipment (e.g., the user equipment or UE may determine its positioning information (which identifies the location or position of the user equipment or UE) by receiving its positioning information from a location management function (LMF) or other node or entity, or by calculating its own positioning information based on the received signal and / or the received information).

[0040] about Figure 2 The method may also include: determining, by the user equipment, the best or selected network node to transmit the second beam based on signal measurements of one or more network nodes transmitting the second beam received from the network node; and controlling, by the user equipment, to send a beam index of the best or selected network node to transmit the second beam to the network node.

[0041] about Figure 2 A method in which a user equipment determines the best or selected network node to transmit a second beam based on a signal measurement of one or more network nodes transmitting a second beam received from the network node, which may include: performing the following items for each of the one or more network nodes transmitting a second beam: receiving a signal from the network node via a beam pair controlled by the user equipment, the beam pair including the network node transmitting the second beam and the associated user equipment receiving the second beam; measuring a signal parameter (e.g., RSRP) of the received signal by the user equipment; and selecting, by the user equipment based on the measurement, the best or selected network node (e.g., one with the highest RSRP or other signal parameters) to transmit the second beam from a plurality of network nodes transmitting second beams (e.g., a gNB transmitting a narrow beam or a gNB transmitting a CSI-RS beam) .

[0042] about Figure 2 The method may also include: controlling, by the user equipment, receiving data from the network node via a beam pair, the beam pair including a best or selected network node transmitting a second beam and an associated user equipment receiving the second beam.

[0043] about Figure 2 In the method, the pre-trained machine learning model may not be specific to the user device, and / or may not be specific to the receiving beam configuration of the user device, but may be common to several user devices, and / or may be general or applicable to different receiving beam configurations of multiple user devices, and is trained based on information provided by multiple user devices.

[0044] about Figure 2 The method may include an arrival angle including an arrival azimuth angle and an arrival elevation angle. The method may further include: converting, by the user equipment, the arrival angle from the first coordinate system to the second coordinate system based on a transformation matrix.

[0045] about Figure 2 The method may also include: controlling, by the user equipment, to send information to the network node so that the network node can perform pre-training of the machine learning model.

[0046] about Figure 2 In the method, the first beam may include a wide beam; and / or the second beam may include a narrow beam narrower than the wide beam.

[0047] about Figure 2 In the method, the first beam may include a wide beam; the network node transmitting the second beam may include a narrow beam; and the user equipment receiving the second beam may include a narrow beam; wherein the wide beam has a beam width wider than the narrow beam.

[0048] about Figure 2 In the method, the first beam may include a synchronization signal block (SSB) beam or a channel state information reference signal (CSI-RS) beam; the network node may transmit a second beam which may include a channel state information reference signal (CSI-RS) beam; and the user equipment may receive the second beam which may include a channel state information reference signal (CSI-RS) beam.

[0049] about Figure 2 In the method, the signal measurement may include a reference signal received power (RSRP) measurement of a reference signal.

[0050] Figure 3 3 is a flow chart illustrating the operation of a network node (e.g., a gNB). Operation 310 includes controlling, by the network node, sending a signal to a user equipment (e.g., a UE) via a plurality of first beams (e.g., a plurality of wide beams or SSB beams). Operation 320 includes controlling, by the network node, receiving a signal measurement (e.g., an RSRP measurement or other signal measurement) for a set of one or more first beams from the user equipment. Operation 330 includes controlling, by the network node, receiving positioning information for the user equipment. Operation 340 includes determining, by the network node, a beam index and an estimated angle of arrival for transmitting a second beam (e.g., a gNB transmitting a narrow beam or a gNB transmitting a CSI-RS beam) for one or more network nodes based on the signal measurement for the set of one or more first beams and the positioning information for the user equipment. And, operation 350 includes controlling, by the network node, sending to the user equipment information indicating an estimated angle of arrival for transmitting the second beam by the one or more network nodes.

[0051] Figure 3 The method may further include: controlling, by the network node, to receive, from a user equipment (or UE), a beam index of a best network node transmitting the second beam among one or more network nodes transmitting the second beam.

[0052] about Figure 3 The method for determining the beam index and the estimated angle of arrival may include: providing, by a network node, signal measurements for a set of one or more first beams and positioning information for a user equipment as input to a trained machine learning model enabled on the network node; and controlling, by the network node, receiving, as output from the trained machine learning model, the beam index and the estimated angle of arrival for transmitting a second beam for one or more network nodes.

[0053] about Figure 3 The method may also include: controlling, by the network node, one or more network nodes indicated by a pre-trained machine learning model to transmit a second beam to send a signal; and controlling, by the network node, receiving, from a user device, a beam index of the best or selected network node transmitting the second beam among the one or more network nodes transmitting the second beam.

[0054] about Figure 3 The method may also include: controlling, by the network node, to send data to the user equipment via a beam pair, wherein the beam pair includes an optimal or selected network node transmitting a second beam and an associated user equipment receiving the second beam.

[0055] about Figure 3 In the method, the pre-trained machine learning model may not be specific to the user device, or may not be specific to the receiving beam configuration of the user device, but may be common to several user devices, and / or may be trained based on information provided by multiple user devices.

[0056] about Figure 3 In the method, the first beam may include a wide beam; and / or the second beam may include a narrow beam narrower than the wide beam.

[0057] about Figure 3 In the method, the first beam may include a wide beam; the network node transmitting the second beam may include a narrow beam; and the user equipment receiving the second beam may include a narrow beam; wherein the wide beam has a beam width wider than the narrow beam.

[0058] about Figure 3In the method, the first beam may include a synchronization signal block (SSB) beam or a channel state information reference signal (CSI-RS) beam; the network node may transmit a second beam which may include a channel state information reference signal (CSI-RS) beam; and the user equipment may receive the second beam which may include a channel state information reference signal (CSI-RS) beam.

[0059] about Figure 3 In the method, the signal measurement may include a reference signal received power (RSRP) measurement of a reference signal.

[0060] about Figure 3 The method may also include training a machine learning model based on information received from multiple user devices to obtain a trained machine learning model.

[0061] about Figure 3 The method may also include: receiving, by a network node, information from a plurality of user devices, the information including signal measurements for a set of one or more first beams and positioning information for the user devices; and training a machine learning model based on the information received from the plurality of user devices to obtain a pre-trained machine learning model.

[0062] about Figure 3 The method, the angle of arrival (AoA) may include both an azimuth angle of arrival and an elevation angle of arrival, and the method may further include: converting the angle of arrival from a first coordinate system to a second coordinate system based on a transformation matrix by the network node.

[0063] Figures 4 to 11 and the associated text show and / or describe one or more further features and / or details that may be found in Figure 2 and / or Figure 3 within a method or as Figure 2 and / or Figure 3 A portion of a method is provided or included.

[0064] Model training at gNB

[0065] Figure 4 is a diagram illustrating the training of a machine learning (ML) model by a network node or gNB or other network entity. Figure 4As shown, the gNB 410 can communicate with one or more UEs, including UE 412. The UE can provide information to the gNB 410, and the gNB can train the ML model to be used on or by one or more UEs. Training input data: RSRP measurements of the N best SSB (or wide) beams measured by the UE at P1, and positioning information of the UE. Training of the ML model can use a supervised learning approach: input data representing the values ​​of the variables (RSRP measurements of the N best SSB or wide beams, and UE positioning information) and labels (as outputs to be trained) to train the ML model parameters. Label: An index representing the AoA estimated for each of the best gNB transmit (TX) beams and those beams, typically reported by the UE. ML model training can be performed by the gNB based on information received from multiple (or more) UEs.

[0066] Step A: UE 412 performs the P1 procedure for initial access to gNB 410. During P1, the UE measures the RSRP from the SSB beams for beam selection. At the end of P1, UE 412 may request access to gNB 410 by transmitting in the time-frequency location corresponding to the best SSB beam. gNB 410 may then configure UE 412 to send back the indices of the best N SSB beams for data collection and their associated N RSRPs. N may be set to a value that depends on the UE capabilities.

[0067] Step B: gNB 410 requests UE location (UE positioning information) from UE or using a network-based positioning method. The gNB may obtain UE positioning information of UE 412 from UE 412, from the location management function (LMF) of the network, or other techniques.

[0068] Step C: gNB 410 performs P2 and configures the UE to measure CSI-RS (or measure RSRP of CSI-RS beams or reference signals) while scanning a small set of narrow (e.g., CSI-RS) beams around the best (e.g., SSB or wide) beam of P1. The UE sends a CSI-RS (e.g., narrow beam) measurement report back to the gNB, which identifies the best narrow (or CSI-RS) beam (e.g., the best gNB transmits the narrow beam) that achieves the highest RSRP, e.g., from the list of CSI-RS measurements.

[0069] Step D: The gNB performs P3 and uses the best narrow beam (best gNB transmit narrow beam) of P2 to transmit several CSI-RS, while the UE scans the UE receive narrow (e.g., CSI-RS) beams to identify the best UE receive narrow (e.g., best UE receive CSI-RS) beam that has or achieves the best (highest) RSRP. During P3, the UE may estimate the AoA for each Rx beam using beam scanning, and report the RSRP measured on the best Rx beam and the associated AoA information to the gNB. The AoA may be reported in a global coordinate system or a local coordinate system as follows: If the UE reports the AoA in the global coordinate system, the UE may transform the AoA from the local coordinate system to the global coordinate system using a transformation matrix signaled or provided by the gNB.

[0070] If the UE reports AoA in the local coordinate system, then in addition to the UE device orientation information, the UE may report the AoA expressed in the local coordinate system to the gNB. The gNB may transform the AoA to the global coordinate system by using the device orientation to compute the transformation matrix. For UEs supporting vertical beamforming, both AoA in azimuth and elevation directions may be signaled or provided to the best beam pair.

[0071] Step E: Supervised learning methods or other techniques may be employed to train the ML model taking into account (or based on) data collected from multiple UEs at the gNB. For each UE, RSRP beam (e.g., SSB) measurements and UE positioning information (UE location) from P1 may be used as input data samples for the ML model. The best transmit beam index in P2 and the associated estimated AoA in P3 (e.g., which may be converted to a predefined grid) may be used as output data samples (labels) for each set of inputs (RSRP beam measurements and UE positioning information for SSB beams or wide beams). Other UEs (such as UE 420) may perform similar or identical processes.

[0072] Figure 5 is a block diagram illustrating the operation of a user equipment (or UE). Steps 1-6 are Figure 5 Shown in.

[0073] Model Inference at UE

[0074] Step 1: A new UE 512 (which may or may not provide data or information to be used for ML model training) performs a P1 procedure for initial access to the gNB 410. Therefore, it is not necessary for a UE to receive and use the ML model to provide information or data for training the ML model, because the ML model is generic and can be used by many or all UEs.

[0075] During P1, the UE measures the RSRP of the SSB (or wide) beams and then requests access to the gNB 410 by sending a random access (RACH) preamble (to establish a connection with the gNB 410) in the time-frequency position corresponding to the best (e.g., with the highest RSRP, as measured by the UE 512) SSB beam. After connecting to the gNB, the gNB may configure the UE to perform measurements and report them according to the measurement configuration. In the measurement report, there is an option to include SSB-based measurement results. The UE may then send an SSB measurement report to the gNB including the RSRP of the N best SSB (or wide) beams measured by the UE.

[0076] Step 2: A pre-trained ML model 520 is transmitted or sent from the gNB 410 to the UE 512. Pre-trained may indicate that the ML model has been previously trained, for example, by another entity (such as, by the gNB). The ML model may be downloaded as a file using a dedicated AI / ML DL signaling message, for example, using the best SSB beam identified during step 1. The ML model may be installed and enabled (where the ML model is running or executing, or being executed by a processor or other circuitry) on the UE 512.

[0077] Step 3: Step 3 may include step 3A and step 3B. Step 3A may include ML model inference and may include the UE obtaining or determining ML model outputs (beam index and estimated or predicted AoA) for one or more gNB transmit narrow beams based on SSB beam signal measurements and UE positioning information. Step 3B may include the UE determining the UE receive narrow beam (e.g., CSI-RS) for each of the one or more gNB transmit narrow beams (beam index and AoA) determined at step 3A (output by ML model 520).

[0078] For step 3A, at the UE, a set of N RSRP measurements from the SSB beams recorded during step 1 is combined with UE location (positioning or location) information such as latitude and longitude coordinates (obtained using a GNSS-based method, such as a Global Positioning System method or any other positioning method on the UE side) and a vertical component obtained from UE device sensors (if available). The aggregated data is then used as input to the ML model. Thus, the UE device performs the ML model inference and saves the results. Input to the ML model: N RSRP measurements of the N best SSB beams and UE positioning information; the output of the ML model is the beam indices of the K (e.g., K = 1, 2, ..., or 5 (or other number)) best transmit beams, and the estimated (or ML model predicted) AoA for each of those beams.

[0079] The AoA obtained by the UE 512 from the ML model 520 output may be expressed in a global coordinate system, and the UE may require a transformation matrix signaled by the gNB to convert the AoA from the global coordinate system to the local coordinate system. Alternatively, the gNB 512 may train the ML model to predict the AoA in the UE local coordinate system, and in this case the AoA may be used directly.

[0080] In step 3B, for each of the K best beams (beam index and estimated AoA) output by the ML model 520, the UE selects a UE receive narrow (e.g., CSI-RS) beam in the codebook that has the closest estimated AoA. Thus, for each of the K best beams (beam index and estimated AoA for each of the K best beams) output by the ML model 520, the UE 512 may determine or select a UE receive narrow beam from a beam codebook (a set of beams, each with its own AoA), wherein the selected UE receive narrow beam (e.g., CSI-RS) beam is the beam in the codebook (or from the codebook) that has the AoA that most closely matches the estimated AoA of the beam output from the ML model 520 (or predicted by the ML model 520). For each AoA output or predicted by the ML model 520, there may be a different UE receive narrow beam selected or determined by the UE 512 (e.g., based on a codebook). Thus, for example, the UE 512 may select a beam of the codebook based on the minimum (or least) angular separation between the predicted AoA (output by the ML model 520) and the angular direction (or AoA) of each beam within the UE beam codebook. These selected (e.g., from the codebook, based on the minimum angular separation) UE receive narrow beams are the UE receive narrow beams associated with each of the K best gNB transmit narrow beams output by the ML model 520.

[0081] The gNB and / or UE may later determine which of the K gNBs transmitting narrow beams (e.g., CSI-RS) is the best, since the UE and gNB may typically use only the best beam pair for communication (e.g., the best gNB of the K gNBs transmitting narrow beams transmits a narrow beam, and the associated UE receives the narrow (e.g., CSI-RS) beam). However, the gNB and UE may then determine which of the K gNBs transmitting narrow beams (e.g., CSI-RS) is the best, according to the following steps and Figure 6 , further steps are performed to determine which of these K gNB transmit narrow beams is the best beam (e.g., the beam with the highest RSRP measured by the UE).

[0082] The UE 512 may determine the beam weights in two ways: First, the UE 512 may calculate the beam weights based on the estimated AoA (predicted by the ML model 520), e.g., the UE may determine or directly calculate the antenna weights to be applied to the antenna system to form a beam in the direction of the estimated AoA. Alternatively, if a discrete or finite set of beams is used by the UE (e.g., a codebook or a set of specific beams, each beam having a different set of AoA and antenna weights), the UE 512 may select a beam from the codebook having an AoA that most closely matches the AoA estimated or predicted by the ML model 520 (e.g., the UE 512 selects 1 of the M beams in the codebook that is closest to the AoA output by the ML model 520).

[0083] Step 4: UE 512 may send an ML model feedback message to gNB 410 indicating the K best CSI-RS beam indices obtained from the ML model output (without sending AoA to gNB). The message at step 4 may include a list of K beam (CSI-RS) indices, i.e. It requires beam refinement throughout the CSI-RS transmission (or by gNB narrow beam scanning at step 5).

[0084] Step 5: During the second beam scanning phase, the gNB scans through a set of K best gNB transmit narrow beams (indicated in the message sent at step 4, and CSI-RS / beam is sent for each of the K best gNB narrow transmit beams indicated by the message received from the UE at step 4). The UE receives a reference signal (e.g., CSI-RS) via each of the K gNB narrow transmit beams, using an associated UE receive narrow beam for each of the K gNB narrow transmit beams. As described above, the associated UE receive narrow beam may be determined as the beam from the codebook having an AoA that most closely matches the estimated AoA output by the ML model 520 for that gNB transmit narrow beam. For each beam in the sequence of the K best gNB transmit narrow beams sent by the gNB, the UE adapts (changes or adjusts) the UE receive narrow beam direction according to the AoA of the selected or associated UE receive narrow beam (e.g., of the selected codebook beam). UE 512 measures (or performs other signal measurements) the RSRP of each transmit / receive beam pair (where, for the K best gNB transmit narrow beams output by ML model 520 and sent by gNB 410, a beam pair includes a gNB narrow transmit beam and an associated UE receive narrow beam). At the end of this stage, the UE identifies the maximum or highest RSRP value from the list of K measurements to select the best gNB transmit narrow beam among the K gNB transmit narrow beams. Thus, the UE can retrieve and store the AoA of the best gNB transmit narrow beam (and the associated UE receive narrow beam). Also at step 5, as Figure 5 As shown, UE 512 sends a beam measurement report to gNB 410 that identifies the best (having the highest RSRP as measured by the UE) gNB transmitting a beam (e.g., indicating a beam index) among the narrow (e.g., CSI-RS) beams.

[0085] Step 6: gNB 410 initiates data transmission (sends data to UE 512) in DL (downlink) (or may receive data from UE in UL (uplink)) using the best gNB transmit narrow beam indicated to the gNB in ​​step 5. The UE may receive the data from the gNB using the UE receive narrow beam associated with the best gNB transmit narrow beam (which is stored by UE 512). These two beams form a beam pair that may be used for data transmission (for uplink and / or downlink transmission) between the gNB and the UE.

[0086] Figure 6 is a diagram illustrating the operation of a gNB or network node. Figure 6 In the present invention, the ML model is enabled on the gNB (e.g., installed and running, or executed by the gNB through a processor or other circuit system).

[0087] Model inference on gNB side

[0088] about Figure 6 , describes ML model inference (e.g., ML model 520 based on ML model input, based on its trained or pre-trained output labels or output set) and use at a gNB or network node, provides some details of the main steps, and highlights the Figure 5 The difference in the operation of ML model inference at the UE is shown.

[0089] Step 1bis: Figure 5 Same as step 1 in .

[0090] Step 2bis: gNB 410 requests UE location (or UE positioning information) from UE or using a network-based positioning method. gNB 410 may receive positioning information of UE 512 from UE, from LMF, or using other techniques.

[0091] Step 3bis: The gNB 410 uses the set of N RSRP measurements from the SSB or wide beams determined and stored during step 1 and the UE positioning information obtained during step 2bis as inputs to the ML model 520. Thus, the gNB 410 performs ML model inference and saves the output or results of the ML model. Step 3bis may include ML model inference, and may include the UE obtaining or determining the ML model output (beam index and estimated or predicted AoA) for one or more gNB transmit narrow (e.g., CSI-RS) beams based on the SSB beam signal measurements and the UE positioning information. Thus, the UE 512 performs ML model inference and saves the results. The inputs to the ML model 520 may include: RSRP measurements of the N best SSB beams and UE positioning information; the output of the ML model 520 may include beam indices for the K (e.g., K=1, 2, or 5 (as in Table 2) or other number) best gNB narrow transmit beams, and an estimated (or ML model predicted) AoA for each of those beams. The K best beam indices are used by the gNB to select the gNB to transmit narrow beams during the next beam refinement phase.

[0092] The AoAs of the K best beams are sent to the UE side. The AoAs obtained from the ML model output may be expressed in a global coordinate system, and the UE may require a transformation matrix, which may be signaled by the gNB 410 to the UE 512 to transform the AoAs from the global coordinate system to the UE local coordinate system. Alternatively, the gNB 512 may convert the AoAs to a local coordinate system at the gNB side, using the device orientation communicated by the UE, to calculate the transformation matrix.

[0093] The UE 512 may then select the beam in the codebook that most closely matches the estimated or predicted AoA indicated by the gNB 410 for the UE to receive a narrow (e.g., CSI-RS) beam (e.g., based on the minimum angular separation between the predicted AoA in the local coordinate system and the angular direction of each receive (Rx) beam within the UE codebook). The UE may select the beam that the UE receives a narrow (e.g., CSI-RS) beam, for example, based on one of the two approaches described above, either directly using the beam with the estimated AoA and calculating the antenna weights, or by using the codebook (or set of discrete beams) beams (and associated antenna weights) with the AoA that most closely matches the estimated or predicted AoA received from the gNB 410. A fixed or known formula or lookup table may be used to calculate the beam weights (e.g., the amplitude and phase of each beam weight to be applied to the antenna system) based on the AoA, for example, such as where the UE is able to provide a large number of beams or beam directions. If the UE can use only a limited number or less of discrete beam directions (and sets of beam weights), then a codebook may be advantageously used, for example.

[0094] Step 4bis: gNB 410 sends an ML model feedback message to UE 512, which indicates the AoAs of the K best beams obtained from the ML model output. The message may include a list of K AoAs, i.e. To be used by UE 512 to receive and measure RSRP of received signals during the refinement phase of 5bis.

[0095] Step 5bis: During the second beam scanning phase, the gNB 410 scans through a set of gNB transmit narrow beams corresponding to the predicted K best beams. The UE adapts the UE receive narrow beam direction according to the AoA value in the sequence of K AoA values ​​contained in the ML model feedback message sent in step 4bis (or according to the codebook beam that most closely matches the AoA value sent to the UE). At the end of this phase, the UE 512 identifies the best UE receive narrow beam from the RSRP measurement and reports the best beam index to the gNB 410, and identifies the best gNB narrow beam from the RSRP measurement. The gNB 410 then receives a measurement report from the UE including one or more indices identifying the (multiple) best gNB transmit narrow beams.

[0096] Step 6bis: gNB 410 initiates data transmission in DL (or UL) using the best gNB transmit narrow beam index reported in step 5bis, and UE 512 uses the best receive narrow beam found in step 5bis. gNB 410 initiates data transmission in DL (downlink) (sends data to UE 512) using the best gNB transmit narrow beam reported in step 5bis (or may receive data from the UE in UL (uplink)). UE 512 may receive the data from gNB 410 using the UE receive narrow beam associated with the best gNB transmit narrow beam. The two beams form a beam pair that may be used for data transmission between gNB 410 and UE 512 (for uplink and / or downlink transmission).

[0097] Wireless network

[0098] An illustrative wireless network may include a sectorized deployment of gNBs (or network nodes) to cover outdoor areas, and which may include buildings. The gNB may be equipped with an A H ×A V A rectangular array of antennas, each spaced half a wavelength apart in both the horizontal and vertical dimensions and each having two polarizations. A set of UEs may be equipped with one or more B H ×B V A rectangular array of antennas, each with a half-wavelength spacing in both the horizontal and vertical dimensions and each with two polarizations. The UE can be located outdoors or inside a building on different floors. The UE position can be expressed in 3D coordinates as (x ( ,y ( ,z ( ), where index i identifies the UE.

[0099] DL (downlink) transmission (from gNB 410 to UE 512) can be based on beamforming. At gNB 410, the beam (gNB transmit beam for DL ​​transmission) can be selected from a predefined codebook of size W, expressed as V TX = {v " ,v $ ,…,v +}, which covers both the azimuth and elevation directions. TX The subset of includes N SSB (or N wide) beams, forming the set of wide beams used in P1. For the beam refinement process P2, a set of narrow (e.g., CSI-RS) beams may be used, e.g., with size M TX A set of narrow beams of , which can be within the angular space of the SSB beam adopted or selected in process P1. At the UE side, a fixed size codebook can be used and can be expressed as Contains a dimension M ,X A set of UE beams d covers both azimuth and elevation directions.

[0100] Training phase

[0101] Figure 7 is a diagram illustrating the training of a machine learning (ML) model. The ML (machine learning) model 520 may be trained based on a data set collected from s UEs located in different locations in a gNB sector and reporting each of the multiple measurements in time. The ML model may be implemented in different forms such as DNN (deep neural network), FNN (feedforward neural network) or CNN (convolutional neural network). In addition, the ML model architecture may be formed by a single layer or multiple layers. The ML model input data may be as follows. The input data (input to the ML model 520) may be considered as a vector It includes:

[0102] 1) UE positioning information, e.g., estimated 2D coordinates of the UE location Including the vertical position of the UE, that is as well as,

[0103] 2) A set of (e.g., SSB or wide) beams of size N {v " ,v $ ,…,v .}Measured and reported RSRP (reference signal received power) signal measurement. After performing the random access procedure, the UE can measure the RSRP value of the gNB for the best N SSB beams received from the gNB, where such RSRP value can be used as input to the ML model 520.

[0104] Using a combination of RSRP measurements from P1 and UE positioning information as input to the ML model 520 can increase or improve the robustness of beam and AoA prediction in case the UE positioning information becomes unreliable. Additionally, N can depend on the size of the wide beam codebook at the gNB and modify the ML model input data. The ML model 520 can be specific to the gNB GoB configuration.

[0105] Figure 8 is a diagram showing gNB beams and UE beams. Figure 8 As shown, a fixed receive beam can be used or assumed during P2 to identify the best gNB transmit narrow beam during the training phase. Then, during P3, the gNB can transmit on the fixed transmit (Tx) narrow beam while the UE scans the receive beams, measuring the best performance of all receive beams (M ,X ) and estimate the AoA for each receive beam. Figure 8 As shown, on the left is the gNB codebook with N=3 SSB beams and 6 CSI-RS for each SSB beam. On the right, the UE codebook and AoA estimation are shown, where the fixed beam points in the direction of Rx beam 7.

[0106] Therefore, the UE may record or store a report containing M-1 rows as described in Table 1, where M is determined by the number of transmit (Tx) narrow beams in P2 (M TX ) plus the number of receive (Rx) beams in P3 (M ,X ). Table 1 includes a table of CSI-RS measurements performed by the UE at P2 and P3 devices during ML model training. The UE measures the RSRP and estimated AoA of the Tx beam. The Rx (receive) beam can be excluded from the report because the ML model does not require this information. In Table 1 the best transmit / receive (Tx / Rx) beam is identified based on the maximum RSRP. Therefore, during the training phase (for pre-training the ML model), the UE can send back (transmit) the best beam index and the associated AoA to the gNB. On the gNB side, the best beam index is mapped to the Tx narrow beam, while the AoA is mapped to a discrete value The discrete value The size W is defined based on the number of independent Rx beams. ,X A predefined grid is selected. The ML model can use the estimated AoA, which makes the ML model applicable to different beam grid configurations. Therefore, the receive beam index of the configuration of the beam grid specific to the user equipment can be excluded from the report. During ML model training, the index of the best Tx beam and the associated AoA are used as labels (outputs to be trained) and can be expressed as: Table 1

[0107] about Figure 8 , on the left is the gNB codebook with N=3 SSB beams (wide beam) and 6 CSI-RS beams for each SSB beam. On the right, the UE codebook and AoA estimation are shown, where the fixed beam points in the direction of Rx beam 7.

[0108] Fig. 9 is a diagram illustrating the ML model inference phase at a UE or user equipment. ML model training may use several data samples with input data. The size of a given ML output layer is expressed as l=M TX ×W ,X , the classification cross entropy loss function used can be expressed as:

[0109] Among them, the lth ML model output is the same as the 2-element vector Correspondingly, where l tx is the beam index and is the estimated or predicted angle of arrival.

[0110] Inference phase:

[0111] Fig.10 2 is a diagram showing representation of the gNB and UE codebooks for the second beam scanning phase. Fig.10 An example of a gNB and UE codebook with a list of transmit (Tx) CSI-RS beams and AoA IDs (identifiers) is shown, which can be used in the second beam scanning phase. The trained ML model can be deployed online to predict the best beam pair to be used by the gNB / UE for data transmission. Fig. 9 As shown, the input of the ML model is the data collected for a single UE, and the ML model output can be expressed as:

[0112] Among them, P l =P(y=l|X) represents the probability of selecting the lth ML model output value given the ML model input vector X.

[0113] From the ML model output, the UE or gNB can use the predictions of the 2nd, 3rd and Kth beam and AoA pairs with the probability expressed as

[0114] Therefore, the ML model output may include a list of 2-element vectors:

[0115] Corresponding to the best K gNB narrow Tx (transmit) beam indices and the associated estimated (or predicted) AoA. Next, as Fig.10 As shown, during the second beam scanning phase, the ML model output is used as follows: 1) The gNB configures the UE to measure (e.g., measure the RSRP of the signals received via these beams) a set of narrow Tx beams specified in a list of 2-element vectors. 2) The UE calculates or determines a list of predicted Rx beams from the list of predicted AoAs and uses the sequence of Rx (receive) beams (e.g., the UE receives narrow beams) for measuring CSI-RS.

[0116] After the second beam scanning phase, the UE may compile or generate a report as shown in Table 2, where each CSI-RS may be measured taking into account the Tx (transmit) beam and Rx (receive) beam selected based on the predicted AoA at the UE side. Table 2 shows a list of CSI-RS measurements suggested by the ML model output. For each CSI-RS, the prediction from the ML model may specify the Tx (transmit) CSI-RS beam and the AoA Grid Id (indicating the AoA of the Tx beam), which may be used at the UE side to establish or determine the associated Rx (receive) beam. The list of measured RSRPs may then be used by the UE to find the best Rx beam and send an indication of the best gNB narrow transmit beam to the gNB, e.g., indicating the CSI-RS identifier associated with the best transmit narrow beam. Table 2

[0117] Embodiments may include numerous technical advantages, including one or more of the following:

[0118] Reduction of beam alignment procedures, signaling overhead and / or measurements. Performance gains can be achieved as the time required to establish the best gNB / UE beam can be significantly reduced compared to the standard 3GPP beam alignment procedures P2 and P3, since the proposed method narrows the number of CSI-RS measurements during the beam refinement phase. Furthermore, since fewer CSI-RS reports are requested by the gNB, the associated signaling overhead is significantly reduced.

[0119] The flexibility of embodiments with different UE device models is exploited since the same ML model can be applied to different configurations of UE beam grids and use the AoA information to select the UE device Rx beam.

[0120] Improved ML inference process on the UE side, as the associated embodiments do not require the UE positioning information to be fed back to the gNB, but consider downloading the ML model to the UE to make predictions locally on the UE side. Thus, sharing the UE location information with the gNB is avoided.

[0121] Significant improvement in ML model prediction accuracy on the UE side, as the combination of UE positioning information and RSRP measurements from SSB beams are used as input data increases the robustness and accuracy of the ML model and speeds up the ML model training phase.

[0122] As described above, ML models can be used. Typically, one or more nodes within a wireless network (e.g., BS, gNB, eNB, RAN node, user node, UE, user equipment, relay node or other node) can use or adopt ML models, for example, such as neural network models (e.g., which can be referred to as neural networks, artificial intelligence (AI) neural networks, AI neural network models, AI models, machine learning models or algorithms or other terms) to perform or assist in performing one or more functions. Other types of models may also be used. A neural network or ML model may be or may include a computational model used in machine learning composed of nodes organized in layers. Nodes are also referred to as artificial neurons, or simply neurons, and perform functions on the provided inputs to produce some output values. A neural network or ML model requires a training cycle to learn parameters (i.e., weights) for mapping inputs to desired outputs. Mapping occurs via a function. Therefore, weights are weights for the mapping function of a neural network. Each neural network model or ML model can be trained for a specific task.

[0123] In order to provide an output given an input, a neural network model or ML model should be trained, which may involve learning appropriate values ​​for a large number of parameters (e.g., weights) for a mapping function. Parameters are also often referred to as weights because they are used to weight terms in a mapping function. The training may be an iterative process, with the values ​​of the weights being adjusted over many (e.g., thousands) of training rounds until the optimal or most accurate values ​​(or weights) are reached. In the context of a neural network (neural network model) or ML model, the parameters may be initialized, typically with random values, and a training optimizer iteratively updates the parameters (weights) of the neural network to minimize the error in the mapping function. In other words, during each round or step of iterative training, the network updates the values ​​of the parameters so that the values ​​of the parameters eventually converge to the optimal values.

[0124] As an example, a neural network model or ML model can be trained in a supervised or unsupervised manner. In supervised learning, training examples are provided to a neural network model or other machine learning algorithm. The training examples include inputs and desired or previously observed outputs. Training examples are also referred to as labeled data because the inputs are labeled with the desired or observed outputs. In the case of a neural network, the network learns the values ​​of the weights used in the mapping function that most often result in the desired output given the training input. In unsupervised training, the neural network model learns to identify structures or patterns in the input provided. In other words, the model identifies implicit relationships in the data. Unsupervised learning is used in many machine learning problems and typically requires a large set of unlabeled data.

[0125] According to example embodiments, the learning or training of a neural network model or ML model may be classified into (or may include) two broad categories (supervised and unsupervised), depending on whether there is a learning "signal" or "feedback" available for the model. Thus, for example, within the field of machine learning, there may be two main types of learning or training of a model: supervised and unsupervised. The main difference between the two types is that supervised learning is performed using known or prior knowledge of what the output values ​​for certain data samples should be. Thus, the goal of supervised learning may be to learn a function that, given a data sample and a desired output, best approximates the relationship between observable inputs and outputs in the data. On the other hand, unsupervised learning does not have labeled outputs, so its goal is to infer the natural structure that exists within a set of data points.

[0126] Supervised learning: The computer is presented with example inputs and their desired outputs, and the goal can be to learn general rules that map inputs to outputs. Supervised learning can be performed, for example, in the context of classification, where a computer or learning algorithm attempts to map inputs to output labels or regression, where a computer or algorithm can map (multiple) inputs to (multiple) continuous outputs. Common algorithms in supervised learning can include, for example, logistic regression, naive Bayes, support vector machines, artificial neural networks, and random forests. In both regression and classification, the goal can include finding specific relationships or structures in the input data that allow us to efficiently produce correct output data. As a special case, the input signal may be only partially available, or limited to special feedback: semi-supervised learning: The computer is only given an incomplete training signal: a training set with some (often many) missing target outputs. Active learning: The computer can only obtain training labels for a limited set of instances (based on a budget), and must also optimize its selection of objects to obtain labels. When used interactively, these can be presented to the user for labeling. Reinforcement learning: Training data (in the form of rewards and penalties) is given only as feedback to the program's actions in a dynamic environment (e.g., using live data).

[0127] Unsupervised learning: No labels are given to the learning algorithm, leaving it to find structure in its input on its own. Some example tasks within unsupervised learning can include clustering, representation learning, and density estimation. In these cases, the computer or learning algorithm attempts to learn the inherent structure of the data without the use of explicitly provided labels. Some common algorithms include k-means clustering, principal component analysis, and autoencoders. Since no labels are provided, there may be no concrete way to compare model performance in most unsupervised learning methods.

[0128] Some additional examples will be provided.

[0129] Example 1. A method (e.g., Figure 2 ) includes: by a user device (for example, Figure 5 UE 512) determines ( Figure 2 210; for Figure 5 Phase P1) is for slave network nodes (e.g. Figure 5 signal measurements (e.g., reference signal received power (RSRP) measurements) for signals (e.g., for reference signals, such as synchronization signal block (SSB) reference signals or channel state information reference signal (CSI-RS) signals) received by a gNB 410 of the present invention for a plurality of first beams (e.g., for a plurality of wide beams or SSB beams); determining, by a user equipment (e.g., UE 512) based on the signal measurements for a set of one or more first beams and positioning information for the user equipment, (e.g., Figure 2 220) transmits a second beam to one or more network nodes (e.g., transmits a narrow beam to one or more network nodes or a gNB or transmits a CSI-RS beam to the gNB) and a beam index and an estimated angle of arrival (AoA); the user equipment (UE 512) transmits a second beam to one or more network nodes in the network node to determine (e.g., Figure 2 230) the associated user equipment receives a second beam (for example, the associated UE receives a narrow beam or the associated UE receives a CSI-RS beam), wherein the second beam received by the associated user equipment is at least one of the following: a beam from a codebook having an angle of arrival that most closely matches an estimated angle of arrival (AoA) for transmitting the second beam by the network node, or a beam having a beam weight (for example, which can be applied to the amplitude and / or phase of the antenna system) calculated based on the estimated angle of arrival for transmitting the second beam by the network node (for example, based on a lookup table or a formula); and the user equipment (for example, UE 512) controls the transmission of ( Figure 2 240) transmits a beam index of a second beam to one or more network nodes.

[0130] For example, a user equipment (eg, UE 512) determines (eg, 220, Figure 2 ) transmitting a beam index and an estimated angle of arrival (AoA) of a second beam for one or more network nodes may include step 3A( Figure 5 , including ML model inference (for Figure 5 The ML model 520) includes a UE 512 obtaining or determining an ML model output (beam index and estimated or predicted AoA) based on SSB beam signal measurements for one or more gNB transmitting narrow beams and UE positioning information.

[0131] Example 2. The method of Example 1, wherein determining the beam index and the estimated angle of arrival comprises: providing, by the user equipment, a signal measurement for a set of one or more first beams (e.g., an SSB beam) and positioning information for the user equipment (for UE 512) as a pre-trained machine learning model enabled (e.g., running or executing) on ​​the user equipment (on UE 512) (e.g., RSRP measurement and UE positioning information may be provided as input to a Figure 5 ML model 520); controlled by user equipment (UE 512) to receive as input from a pre-trained machine learning model ( Figure 5 The output of the ML model 520) is a beam index and an estimated angle of arrival (the beam index and the estimated AoA are) for transmitting a second beam to one or more network nodes. Figure 5 output of the ML model 520).

[0132] Example 3. The method of any one of Examples 1-2 further includes: Figure 5 The UE 512 of the present invention controls receiving a pre-trained machine learning model ( Figure 5 Step 2, ML model download); by the user equipment based on the user equipment for a plurality of first beams (for example, Figure 5 The method further comprises: selecting a set of one or more first beams by performing signal measurement in step 1); and determining positioning information for the user equipment by the user equipment.

[0133] Example 4. The method according to any one of Examples 1-2 further includes: Figure 5 UE 512) determines that the best or selected network node transmits the second beam based on the signal measurement of one or more network nodes transmitting the second beam received from the network node; and controls the user equipment to send the beam index of the best or selected network node transmitting the second beam to the network node (for example, see Figure 5 Step 5 of ). UE 512 measures (or performs other signal measurements) the RSRP of each transmit / receive beam pair (where, for the K best gNB transmit narrow beams output by ML model 520 and sent by gNB 410, the beam pair includes a gNB narrow transmit beam and an associated UE receive narrow beam). At the end of this stage, the UE identifies the maximum or highest RSRP value from the list of K measurements to select the best gNB transmit narrow beam among the K gNB transmit narrow beams. Therefore, the UE can retrieve and store the AoA of the best gNB transmit narrow beam (and the associated UE receive narrow beam). Also at step 5, as Figure 5As shown, UE 512 sends a beam measurement report to gNB 410 that identifies the best (having the highest RSRP as measured by the UE) gNB transmitting a beam (e.g., indicating a beam index) among the narrow (e.g., CSI-RS) beams.

[0134] Example 5. A method according to Example 4, wherein determining by a user equipment (e.g., UE 512) that the best or selected network node transmits the second beam based on a signal measurement of one or more network nodes transmitting the second beam received from the network node comprises: performing the following items for each of the one or more network nodes transmitting the second beam: receiving a signal from the network node via a beam pair controlled by the user equipment, the beam pair comprising the network node transmitting the second beam and the associated user equipment receiving the second beam; measuring by the user equipment a signal parameter of the received signal; and selecting by the user equipment, based on the measurement, the best or selected network node transmitting the second beam from among a plurality of network nodes transmitting the second beam. For example, see Figure 5 Step 5. For each beam in the sequence of K best gNB transmit narrow beams sent by the gNB, the UE adapts (changes or adjusts) the UE receive narrow beam direction according to the AoA of the selected or associated UE receive narrow beam (e.g., of the selected codebook beam). UE 512 measures (or performs other signal measurements) the RSRP of each transmit / receive beam pair (wherein, for the K best gNB transmit narrow beams output by the ML model 520 and sent by the gNB 410, the beam pair includes a gNB narrow transmit beam and an associated UE receive narrow beam). At the end of this stage, the UE identifies the maximum or highest RSRP value from the list of K measurements to select the best gNB transmit narrow beam among the K gNB transmit narrow beams. Therefore, the UE can retrieve and store the AoA of the best gNB transmit narrow beam (of the associated UE receive narrow beam). Also at step 5, as Figure 5 As shown, UE 512 sends a beam measurement report to gNB 410, which identifies the beam (e.g., indicating a beam index) among the best (having the highest RSRP as measured by the UE) gNB transmitting narrow beams (e.g., CSI-RS).

[0135] Example 6. The method according to any one of Examples 1-5 further includes: controlling, by the user equipment, to receive data from the network node via the beam pair (for example, see Figure 5 In step 6), the beam pair includes a best or selected network node transmitting a second beam and an associated user equipment receiving the second beam.

[0136] Example 7. A method according to any of Examples 2-6, wherein the pre-trained machine learning model (e.g., ML model 520) is not specific to a user device, but is common to several user devices and is trained based on information provided by multiple user devices.

[0137] Example 8. A method according to any one of Examples 1-7, wherein the angle of arrival (AoA) includes both an azimuth angle of arrival and an elevation angle of arrival, and the method further comprises: converting, by the user equipment, the angle of arrival from the first coordinate system to the second coordinate system based on the transformation matrix. For example, by UE 512( Figure 5 ) The AoA obtained from the ML model 520 output may be expressed in the global coordinate system, and the UE may require a transformation matrix signaled by the gNB 512 to convert the AoA from the global coordinate system to the local coordinate system. Alternatively, the gNB 512 may train the ML model to predict the AoA in the UE local coordinate system, and in this case the AoA may be used directly.

[0138] Example 9. The method according to any one of Examples 2-8 further includes: controlling, by a user equipment (UE 512), to send a signal to a network node ( Figure 5 The gNB 410 sends information to enable the network node to perform pre-training of the machine learning model (pre-training, such as Figure 4 520 in FIG. 5 ).

[0139] Embodiment 10. A method according to any one of Examples 1-9, wherein: the first beam (e.g., an SSB beam) includes a wide beam; and the second beam (e.g., an SSB beam or a CSI-RS beam or a narrow beam) includes a narrow beam narrower than the wide beam.

[0140] Example 11. A method according to any one of Examples 1-10, wherein: the first beam includes a wide beam; the network node transmits the second beam including a narrow beam; and the user device receives the second beam including a narrow beam; wherein the wide beam has a beam width wider than the narrow beam.

[0141] Example 12. A method according to any one of Examples 1-11, wherein: the first beam includes a synchronization signal block (SSB) beam or a channel state information reference signal (CSI-RS) beam; the network node transmits the second beam including a channel state information reference signal (CSI-RS) beam; and the user equipment receives the second beam including a channel state information reference signal (CSI-RS) beam.

[0142] Example 13. A method according to any of Examples 1-12, wherein the signal measurement includes a reference signal received power (RSRP) measurement of a reference signal.

[0143] Example 14. A device (eg, device 1200) comprising means (eg, processor 1204, memory 1206, and / or wireless transceiver 1202A) for performing the method of any of Examples 1-13.

[0144] Example 15. A non-transitory computer-readable storage medium (e.g., 1206) comprising instructions stored thereon, which when executed by at least one processor (e.g., processor 1204) are configured to cause a computing system to perform the method of any of Examples 1-13.

[0145] Example 16. A device comprising: at least one processor (e.g., processor 1204); and at least one memory comprising computer program code; the at least one memory and the computer program code are configured to, together with the at least one processor, cause the device to at least perform the method of any one of Examples 1-13.

[0146] Example 17. A device comprising: at least one processor; and at least one memory (e.g., 1206) comprising computer program code; the at least one memory and the computer program code are configured to, together with the at least one processor, enable the device to at least: determine, by a user device, a signal measurement for a signal for a plurality of first beams received from a network node; determine, by the user device, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurement for a set of one or more first beams and positioning information for the user device; determine, by the user device, for one or more of the network nodes transmitting the second beam, that an associated user device receives the second beam, wherein the second beam received by the associated user device is at least one of the following: a beam from a codebook having an angle of arrival that most closely matches the estimated angle of arrival for transmitting the second beam for the network node, or a beam having a beam weight calculated based on the estimated angle of arrival for transmitting the second beam for the network node; and control, by the user device, sending a beam index for transmitting the second beam for one or more network nodes to the network node.

[0147] Example 18. A method (e.g., see Figure 3 and Figure 6 ), including: controlling, by a network node (e.g., LMF or gNB 410), a plurality of first beams (e.g., Figure 6 Step 1bis, including gNB sending an SSB signal to UE) sends (for example, Figure 3 310 of them; Figure 6 Phase P1) signal; controlled by the network node (LMF or gNB 410) to receive from the user equipment (for example, Figure 3320) for a signal measurement of a set of one or more first beams (e.g., for an SSB beam or a wide beam) (e.g., an RSRP signal measurement sent via step 1bis after initial access); receiving is controlled by the network node (e.g., Figure 3 330) for positioning information of a user equipment (UE 512) (e.g., in Figure 6 The method further comprises: determining, by the network node (LMF or gNB 410), a beam index and an estimated angle of arrival (AoA) for transmitting a second beam for one or more network nodes based on a signal measurement for a set of one or more first beams and positioning information for the user equipment; and controlling, by the network node, sending ( Figure 3 350) indicates information indicating an estimated angle of arrival of a second beam transmitted by one or more network nodes.

[0148] Example 19. A method according to Example 18, wherein determining a beam index and an estimated angle of arrival includes: providing, by a network node, signal measurements for a set of one or more first beams and positioning information for a user device as input to a trained machine learning model enabled on the network node; and controlling, by the network node, receiving a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes as output from the trained machine learning model.

[0149] Example 20. The method according to any one of Examples 18-19 further includes: controlling, by the network node, one or more network nodes indicated by a pretrained machine learning model to transmit a second beam to send a signal; and controlling, by the network node, receiving, from a user device, a beam index of the best or selected network node transmitting the second beam among the one or more network nodes transmitting the second beam.

[0150] Example 21. The method according to any one of Examples 18-20 further includes: controlling, by a network node, to send data to a user equipment via a beam pair, wherein the beam pair includes an optimal or selected network node transmitting a second beam and an associated user equipment receiving the second beam.

[0151] Example 22. A method according to any one of Examples 18-21, wherein: the first beam comprises a wide beam; and the second beam comprises a narrow beam narrower than the wide beam.

[0152] Example 23. The method of any of Examples 18-22: the first beam comprises a wide beam; the network node transmits the second beam comprises a narrow beam; and the user equipment receives the second beam comprises a narrow beam; wherein the wide beam has a beam width wider than the narrow beam.

[0153] Example 24. A method according to any one of Examples 18-23, wherein: the first beam includes a synchronization signal block (SSB) beam or a channel state information reference signal (CSI-RS) beam; the network node transmits the second beam including a channel state information reference signal (CSI-RS) beam; and the user equipment receives the second beam including a channel state information reference signal (CSI-RS) beam.

[0154] Example 25. A method according to any of Examples 18-24, wherein the pre-trained machine learning model is not specific to the user device and is trained based on information received from multiple user devices.

[0155] Example 26. The method according to any one of Examples 18-25 also includes: training the machine learning model based on information received from multiple user devices to obtain a trained machine learning model.

[0156] Example 27. The method according to any one of Examples 18-26 further includes: receiving information by a network node from multiple user devices, the information including signal measurements for a set of one or more first beams and positioning information for the user devices; and training a machine learning model based on the information received from the multiple user devices to obtain a pre-trained machine learning model.

[0157] Example 28. A method according to any of Examples 18-27, wherein the arrival angle includes both an arrival azimuth angle and an arrival elevation angle, and the method further includes: converting the arrival angle from a first coordinate system to a second coordinate system based on a transformation matrix by the network node.

[0158] Example 29. A device (e.g., Fig.11 1200), comprising components (e.g., processor 1204, memory 1206, and / or transceiver 1202A) for performing a method as described in any of Examples 17-27.

[0159] Example 30. A non-transitory computer-readable storage medium (e.g., Fig.11 1206), including instructions stored thereon, which when executed by at least one processor are configured to cause the computing system to perform a method according to any one of Examples 18-28.

[0160] Example 31. A device (e.g., Fig.11 1200), including: at least one processor ( Fig.11 1204); and at least one memory (1206) including computer program code; the at least one memory and the computer program code are configured to, together with at least one processor, cause the apparatus to at least perform a method according to any one of Examples 18-28.

[0161] Example 32. An apparatus (e.g., 1200) comprising: at least one processor (1206); and at least one memory including computer program code; the at least one memory and the computer program code are configured to communicate with the at least one processor ( Fig.11 1204), so that the device at least: controls, by a network node, to send a signal to a user equipment via multiple first beams; controls, by a network node, to receive signal measurements for a set of one or more first beams from the user equipment; controls, by a network node, to receive positioning information for the user equipment; determines, by the network node, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurements for the set of one or more first beams and the positioning information for the user equipment; and controls, by the network node, to send information to the user equipment indicating the beam index and the estimated angle of arrival for transmitting a second beam for one or more network nodes.

[0162] Fig.11 1 is a block diagram of a wireless station or node 1200 (e.g., a network node, a user node or UE, a relay node, or other node) according to an example embodiment. The wireless station 1200 may include, for example, one or more (e.g., Fig.11 The wireless station also includes a processor or control unit / entity (controller) 1204 for executing instructions or software and controlling the transmission and reception of signals, and a memory 1206 for storing data and / or instructions.

[0163] The processor 1204 may also make decisions or determinations, generate frames, data packets or messages for transmission, decode received frames or messages for further processing, and other tasks or functions described herein. The processor 1204 may be a baseband processor, for example, a message, data packet, frame or other signal may be generated for transmission via the wireless transceiver 1202 (1202A or 1202B). The processor 1204 may control the transmission of signals or messages through a wireless network, and may control the reception of signals or messages, etc., via a wireless network (e.g., after down-conversion by the wireless transceiver 1202). The processor 1204 may be programmable and capable of executing software or other instructions stored in a memory or other computer medium to perform the various tasks and functions described above, such as one or more of the tasks or methods described above. The processor 1204 may be (or may include) for example, hardware, programmable logic, a programmable processor that executes software or firmware, and / or any combination of these. Using other terms, for example, the processor 1204 and the transceiver 1202 together may be considered a wireless transmitter / receiver system.

[0164] In addition, refer to Fig.11 , controller (or processor) 1208 can execute software and instructions and can provide overall control for station 1200 and can provide Fig.11 Other systems shown in the present disclosure provide controls, such as controlling input / output devices (e.g., display, keyboard), and / or can execute software for one or more applications provided on the wireless station 1200, such as an email program, audio / video applications, a word processor, an IP voice application, or other applications or software.

[0165] Additionally, storage media including stored instructions are provided that, when executed by a controller or processor, may cause the processor 1204 or other controller or processor to perform one or more of the functions or tasks described above.

[0166] According to another example embodiment, the RF or (multiple) wireless transceiver 1202A / 1202B can receive signals or data and / or transmit or send signals or data. The processor 1204 (and possibly the transceiver 1202A / 1202B) can control the RF or wireless transceiver 1202A or wireless transceiver 1202B to receive, send, broadcast or transmit signals or data.

[0167] Embodiments of the different technologies described herein may be implemented in a digital electronic circuit system, or in computer hardware, firmware, software, or a combination thereof. Embodiments may be implemented as computer program products, i.e., computer programs tangibly embodied in information carriers (e.g., in machine-readable storage devices or in propagated signals) for execution or control of the operation of data processing devices by data processing devices (e.g., programmable processors, computers, or multiple computers). Embodiments may also be provided on a computer-readable medium or a computer-readable storage medium, which may be a non-transient medium. Embodiments of various technologies may also include embodiments provided via transient signals or media, and / or programs and / or software embodiments downloadable via the Internet or other networks (wired networks and / or wireless networks). In addition, embodiments may be provided via machine type communications (MTC) and the Internet of Things (IOT).

[0168] A computer program may be in source code form, object code form, or some intermediate form, and may be stored in some carrier, distribution medium, or computer-readable medium, which may be any entity or device capable of carrying the program. Such carriers include, for example, recording media, computer memories, read-only memories, optical and / or electrical carrier signals, telecommunication signals, and software distribution packages. Depending on the processing power required, a computer program may be executed in a single electronic digital computer, or it may be distributed among multiple computers.

[0169] In addition, embodiments of the various techniques described herein may use cyber-physical systems (CPS) (systems of collaborative computing elements that control physical entities). CPS enables the implementation and development of a large number of interconnected ICT devices (sensors, actuators, processors, microcontrollers, ...) embedded in physical objects at different locations. Mobile cyber-physical systems are a subcategory of cyber-physical systems in which the physical systems involved have inherent mobility. Examples of mobile physical systems include mobile robots and electronic devices transported by humans or animals. The rise in the popularity of smartphones has increased interest in the field of mobile cyber-physical systems. Therefore, various embodiments of the techniques described herein may be provided by one or more of these techniques.

[0170] Computer programs (such as the computer program(s) described above) can be written in any form of programming language including compiled or interpreted languages, and can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or unit or part suitable for use in a computing environment. The computer program may be deployed to be executed on a computer, or executed on multiple computers at one site, or distributed across multiple sites and interconnected by a communication network.

[0171] The method steps may be performed by one or more programmable processors executing a computer program or computer program portion, which performs functions by operating on input data and generating output. The method steps may also be performed by dedicated logic circuits, and the apparatus may be implemented as, for example, a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC).

[0172] Processors suitable for executing a computer program include, for example, general and special purpose microprocessors, and any one or more processors of any kind of digital computer, chip or chipset. Typically, the processor will receive instructions and data from a read-only memory or a random access memory, or both. Elements of a computer may include at least one processor for executing instructions and one or more memory devices for storing instructions and data. Typically, a computer may also include or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices (e.g., magnetic, magneto-optical, or optical disks) for storing data. Information carriers suitable for embodying computer program instructions and data include all forms of non-volatile memory, including, for example, semiconductor memory devices, such as EPROM, EEPROM, and flash memory devices; magnetic disks, such as internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory may be supplemented or incorporated by special purpose logic circuitry.

[0173] To provide interaction with a user, embodiments may be implemented on a computer having a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information to the user and a user interface through which the user can provide input to the computer (such as a keyboard and a pointing device, such as a mouse or trackball). Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback; and the input from the user may be received in any form, including sound, voice, or tactile input.

[0174] Embodiments may be implemented in a computing system that includes a back-end component (e.g., as a data server), or includes a middleware component (e.g., an application server), or includes a front-end component (e.g., a client computer with a graphical user interface or a web browser through which a user can interact with the embodiments), or any combination of such back-end, middleware, or front-end components. The components may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks (LANs) and wide area networks (WANs), such as the Internet.

[0175] While certain features of the described embodiments have been shown as described herein, many modifications, substitutions, changes, and equivalents will now occur to those skilled in the art. It should therefore be understood that the appended claims are intended to cover all such modifications and changes that fall within the true spirit of the various embodiments.

Claims

1. A method comprising: determining, by the user equipment, a signal measurement for a signal for a plurality of first beams received from the network node; Determining, by the user equipment, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurement for the set of one or more first beams and positioning information for the user equipment; Determining, by the user equipment, that an associated user equipment receives a second beam for one or more of the network nodes transmitting the second beam, wherein the second beam received by the associated user equipment is at least one of the following: a beam from a codebook having an angle of arrival that most closely matches the estimated angle of arrival for transmitting the second beam by the network node, or a beam having a beam weight calculated based on the estimated angle of arrival for transmitting the second beam by the network node; as well as The user equipment controls to send the beam index for transmitting the second beam to the one or more network nodes to the network node.

2. The method of claim 1 , wherein determining a beam index and an estimated angle of arrival comprises: providing, by the user equipment, the signal measurements for the set of one or more first beams and the positioning information for the user equipment as input to a pre-trained machine learning model enabled on the user equipment; The user equipment controls receiving a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes as an output from the pre-trained machine learning model.

3. The method according to any one of claims 1 to 2, further comprising: Controlled by the user equipment to receive the pre-trained machine learning model from the network node; selecting, by the user equipment, the set of one or more first beams based on signal measurements performed by the user equipment for a plurality of first beams; The user equipment determines positioning information for the user equipment.

4. The method according to any one of claims 1 to 2, further comprising: determining, by the user equipment, a best or selected network node to transmit the second beam based on signal measurements received from the network node of the one or more network nodes transmitting the second beam; as well as The user equipment controls to send, to the network node, a beam index of the best or selected network node for transmitting a second beam.

5. The method of claim 4, wherein the determining, by the user equipment, that the best or selected network node transmits the second beam based on the signal measurement received from the network node of the one or more network nodes transmitting the second beam comprises: For each of the one or more network nodes transmitting the second beam, performing the following: receiving, by the user equipment, a signal from the network node via a beam pair, the beam pair comprising the network node transmitting a second beam and the associated user equipment receiving the second beam; Measuring, by the user equipment, a signal parameter of the received signal; as well as The user equipment selects, based on the measurement, a best or selected network node among the multiple network nodes transmitting the second beam to transmit the second beam.

6. The method according to any one of claims 1 to 5, further comprising: The user equipment controls receiving data from the network node via a beam pair, the beam pair comprising a best or selected network node transmitting a second beam and the associated user equipment receiving the second beam.

7. A method according to any one of claims 2 to 6, wherein the pre-trained machine learning model is not specific to the user device, but is common to several user devices and is trained based on information provided by multiple user devices.

8. The method according to any one of claims 1 to 7, wherein the angle of arrival comprises both an azimuth angle of arrival and an elevation angle of arrival, the method further comprising: The angle of arrival is converted from a first coordinate system to a second coordinate system by the user equipment based on a transformation matrix.

9. The method according to any one of claims 2 to 8, further comprising: The user equipment controls the sending of information to the network node so that the network node can perform pre-training of the machine learning model.

10. The method according to any one of claims 1 to 9, wherein: The first beam comprises a wide beam; The second beam comprises a narrow beam narrower than the wide beam.

11. The method according to any one of claims 1 to 10, wherein: The first beam comprises a wide beam; The network node transmits a second beam including a narrow beam; and The second beam received by the user equipment includes a narrow beam; The wide beam has a wider beam width than the narrow beam.

12. The method according to any one of claims 1 to 11, wherein: The first beam comprises a synchronization signal block (SSB) beam or a channel state information reference signal (CSI-RS) beam; The network node transmits a second beam comprising a channel state information reference signal (CSI-RS) beam; and The user equipment receives a second beam including a channel state information reference signal (CSI-RS) beam.

13. The method of any one of claims 1 to 12, wherein the signal measurement comprises a Reference Signal Received Power (RSRP) measurement of a reference signal.

14. An apparatus comprising means for performing the method according to any one of claims 1 to 13.

15. A non-transitory computer-readable storage medium comprising instructions stored thereon, which when executed by at least one processor are configured to cause a computing system to perform the method according to any one of claims 1 to 13.

16. An apparatus comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, together with the at least one processor, cause the apparatus to at least perform the method according to any one of claims 1 to 13.

17. An apparatus comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to at least: determining, by the user equipment, a signal measurement for a signal for a plurality of first beams received from the network node; Determining, by the user equipment, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurement for the set of one or more first beams and positioning information for the user equipment; Determining, by the user equipment, that an associated user equipment receives a second beam for one or more of the network nodes transmitting the second beam, wherein the second beam received by the associated user equipment is at least one of the following: a beam from a codebook having an angle of arrival that most closely matches the estimated angle of arrival for transmitting the second beam by the network node, or a beam having a beam weight calculated based on the estimated angle of arrival for transmitting the second beam by the network node; as well as The user equipment controls to send the beam index for transmitting the second beam to the one or more network nodes to the network node.

18. A method comprising: Controlled by a network node, to send a signal to a user equipment via a plurality of first beams; Controlling, by the network node, receiving, from the user equipment, signal measurements for a set of one or more of the first beams; Controlling, by the network node, receiving positioning information for the user equipment; determining, by the network node, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurement for the set of one or more first beams and positioning information for the user equipment; as well as The network node controls sending information indicating the estimated arrival angle of the second beam transmitted by the one or more network nodes to the user equipment.

19. The method of claim 18, wherein determining a beam index and an estimated angle of arrival comprises: providing, by the network node, the signal measurements for the set of one or more first beams and the positioning information for the user equipment as input to a trained machine learning model enabled on the network node; The network node controls receiving, as output from the trained machine learning model, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes.

20. The method according to any one of claims 18 to 19, further comprising: Controlling, by the network node, the one or more network nodes indicated by the pre-trained machine learning model to transmit a second beam transmission signal; as well as The network node controls receiving, from the user equipment, a beam index of a best or selected network node transmitting the second beam among the one or more network nodes transmitting the second beam.

21. The method according to any one of claims 18 to 20, further comprising: The data is sent to the user equipment via a beam pair controlled by the network node, the beam pair comprising the best or selected network node transmitting a second beam and an associated user equipment receiving the second beam.

22. A method according to any one of claims 18 to 21, wherein: The first beam comprises a wide beam; The second beam comprises a narrow beam narrower than the wide beam.

23. The method according to any one of claims 18 to 22: The first beam comprises a wide beam; The network node transmits a second beam including a narrow beam; and The second beam received by the user equipment includes a narrow beam; The wide beam has a wider beam width than the narrow beam.

24. A method according to any one of claims 18 to 23, wherein: The first beam comprises a synchronization signal block (SSB) beam or a channel state information reference signal (CSI-RS) beam; The network node transmits a second beam comprising a channel state information reference signal (CSI-RS) beam; and The user equipment receives a second beam including a channel state information reference signal (CSI-RS) beam.

25. A method according to any one of claims 18 to 24, wherein the pre-trained machine learning model is not specific to the user device and is trained based on information received from multiple user devices.

26. The method according to any one of claims 18 to 25, further comprising: The machine learning model is trained based on information received from multiple user devices to obtain the trained machine learning model.

27. The method according to any one of claims 18 to 26, further comprising: receiving, by the network node, information from a plurality of user equipments, the information comprising signal measurements for a set of one or more first beams and positioning information for the user equipments; as well as The machine learning model is trained based on information received from the multiple user devices to obtain the pre-trained machine learning model.

28. The method according to any one of claims 18 to 27, wherein the angle of arrival comprises both an azimuth angle of arrival and an elevation angle of arrival, the method further comprising: The angle of arrival is converted from a first coordinate system to a second coordinate system by the network node based on a transformation matrix.

29. An apparatus comprising means for performing the method according to any one of claims 18 to 28.

30. A non-transitory computer-readable storage medium comprising instructions stored thereon, which when executed by at least one processor are configured to cause a computing system to perform the method of any one of claims 18 to 28.

31. An apparatus comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, together with the at least one processor, cause the apparatus to at least perform a method according to any one of claims 18 to 28.

32. An apparatus comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to at least: Controlled by a network node, to send a signal to a user equipment via a plurality of first beams; Controlling, by the network node, receiving, from the user equipment, signal measurements for a set of one or more of the first beams; Controlling, by the network node, receiving positioning information for the user equipment; determining, by the network node, a beam index and an estimated angle of arrival for transmitting a second beam for one or more network nodes based on the signal measurement for the set of one or more first beams and positioning information for the user equipment; as well as The network node controls to send information indicating that the one or more network nodes transmit the second beam and the estimated arrival angle to the user equipment.

Images